Goddard v. Farmers Insurance

*235 GILLETTE, J. This case involves a constitutional challenge to a punitive damages award that a jury imposed on a liability insurance company, based on the company’s bad faith failure to settle a third-party’s claim against one of its insureds within policy limits. The underlying case is a wrongful death proceeding. The insured, John Munson, caused an automobile accident in 1987 while driving under the influence of alcohol. The accident killed Marc Goddard. Goddard’s mother (plaintiff) brought a wrongful death action against Munson. Munson’s insurer, Farmers Insurance Company of Oregon (defendant), undertook Munson’s defense, but failed to settle plaintiffs wrongful death action within policy limits, after which a jury returned a verdict that resulted in a judgment against Munson for $863,274. Munson, who asserted that defendant’s failure to settle was an act of bad faith, assigned his bad faith claim against defendant to plaintiff, who prosecuted the action. A jury awarded plaintiff $863,274 in compensatory damages and $20,718,576 in punitive damages. Defendant moved to reduce both awards. The trial court reduced the compensatory damages award in certain respects, but left the punitive damages award in place. Both sides appealed. The Court of Appeals reinstated two amounts that the trial court had deducted from the compensatory damages award, but also significantly reduced the punitive damages award. Goddard v. Farmers Ins. Co., 202 Or App 79, 120 P3d 1260 (2005), modified and adh’d to as modified on recons, 203 Or App 744, 126 P3d 682 (2006). Plaintiff sought review, which we allowed. For the reasons that follow, we affirm the decision of the Court of Appeals. I. FACTS Plaintiffs “failure to settle” action against defendant is based on defendant’s conduct in the wrongful death action against Munson, so we begin there. For the most part, the parties do not contest the facts as set out in the Court of Appeals opinion, 1 from which we quote liberally. *236 A. Wrongful Death Action The accident at issue was a collision between a truck (which Munson was driving) and a car driven by the deceased. The accident potentially was covered by two separate insurance policies issued by defendant. First, Munson himself had a $100,000 automobile insurance policy with defendant. Second, the truck that Munson was driving was owned by Helen Foley, who also had a $100,000 automobile insurance policy with defendant. Separate litigation would eventually determine that the accident was covered only by Foley’s policy. “The fatal accident occurred on October 29, 1987. Munson, who had consumed a large quantity of alcohol, made a left turn into a tavern parking lot and collided with Marc Goddard’s oncoming vehicle, killing Goddard. Shortly after the collision, Foley, who was to meet Munson at the tavern, arrived on the scene. Both Munson and Foley told a police officer that Munson had been driving Foley’s truck with Foley’s permission. Although Munson believed that he was not impaired by alcohol at the time of the collision, his blood alcohol level taken at the hospital was more than twice the legal limit for operating a motor vehicle.” Goddard, 202 Or App at 86. On November 2, 1987, Foley and Munson both told the defendant’s claim representative, Sellers, that Munson had been driving the truck with Foley’s permission. Id. Munson also admitted to Sellers that he had had “eight to ten beers” before the accident. Id. Despite strong evidence that Munson was at fault in the collision, defendant’s employees resisted plaintiffs attempts to resolve liability: “On November 5, 1987, [attorney] Carl Amala wrote to Sellers, stating that he had been retained by Marc Goddard’s estate and seeking to discuss liability, property *237 damage, and medical and burial expenses. Sellers spoke with Amala by telephone, indicating that Sellers did not believe Munson was at fault in the collision. By early December, Sellers had verified through the police that Goddard had his headlights on at the time of the collision. “On December 10, 1987, Amala again wrote to Sellers, stating that the physical evidence demonstrated that the collision occurred in Goddard’s lane of travel, that Goddard’s headlights were on, and that criminal charges against Munson were contemplated. Amala asked for any information that called into question Munson’s liability. He also asked for information about what Farmers policies were involved and what their limits were. By December 10, 1987, Sellers had concluded that Munson’s liability for the collision was clear. However, neither Sellers nor any other employee of defendant responded to Amala’s letter. “On December 16, 1987, Sellers wrote an investigation report for his supervisor, Doug Heatherington, indicating that both Foley’s and Munson’s policies were potentially involved, that Munson admitted to having consumed nine beers prior to the accident, and that the case “has policy limits potential.’ The following day, Sellers wrote a claim status report, stating that the Goddard estate ‘will want policy limits’ and that ‘[w]e have two policies involved with $200,000 exposure. I feel we should let things “ripen” a bit before making any settlement talk.’ “On January 5, 1988, Heatherington wrote to Don McClure, the regional claims manager in Portland, noting that both Foley’s and Munson’s policies were potentially involved but also noting that Munson’s policy might not apply if Munson used Foley’s vehicle regularly. That letter further stated that Munson admitted having eight to ten beers before the collision; that Marc Goddard had been 19 years old and was single with no children; and that Goddard was believed to have been recently released from jail. The letter concluded that, although ordinarily liability would be clear for making a left turn in front of an oncoming vehicle, Munson had suggested that Goddard might not have had his headlights on. Heatherington requested that a reserve, or settlement value, of $30,000 be established for the case and suggested that police reports, blood alcohol tests, and information on Goddard’s headlights be obtained. *238 “On January 12, 1988, Amala again wrote to Sellers, indicating that he had not received a response to his earlier correspondence and seeking a response. Again, defendant provided no response.” Id. at 87-88. Meanwhile, internal memoranda show that defendant’s employees had little doubt about Munson’s liability for Goddard’s death. “Sellers provided a status report to [Randy] Voth[, who had replaced Heatherington in the Salem office,] on January 22, 1988, indicating that Goddard’s vehicle had its lights on and that Munson had simply made a left turn in front of Goddard’s vehicle after consuming nine beers. Sellers stated, ‘I think we should get some offer out on this if possible soon, or perhaps we should answer a lawsuit and find out facts on [Marc Goddard] through discovery process.’ “Munson was charged with second-degree manslaughter. On March 3,1988, Amala wrote again to Sellers, noting that he had received no response to his three earlier letters, that the police reports indicated that Munson was at fault in the collision, and that Munson had been charged with manslaughter. He again sought information about the policies involved and asked that defendant respond in writing concerning the settlement of the property damage portion of the case. In late March, Amala spoke on the phone with Sellers, and Sellers confirmed that Foley and Munson each had $100,000 of coverage. Sellers also represented to Munson’s [criminal] attorney, Feitelson, that both Foley’s and Munson’s policies would apply.” Id. at 88. At that point, the noncooperative nature of the relationship between plaintiff and defendant became more formal. “On March 28, 1988, Amala sent a demand letter to Sellers, indicating that he would recommend to his client that she accept $200,000 to settle all claims against Munson, giving a three-week response time. On March 31, 1988, Sellers wrote a claim status report noting the $200,000 demand and requesting authority to settle for up *239 to $50,000. The report stated that Sellers was awaiting an accident reconstruction report and that Munson’s [criminal] attorney, Feitelson, was trying to obtain information on Marc Goddard’s criminal history.” Id. at 88-89. In April 1988, Sellers received an accident reconstruction report and information about Goddard’s criminal history. Id. at 89. The accident reconstruction report confirmed that Munson had turned in front of Goddard and that Goddard had been traveling at approximately the speed limit, but the report could not specify whether Goddard had been using his headlights. Id. The criminal history investigation by Feitelson indicated that, at the time of the collision, Goddard had been on probation and that his probation was in jeopardy. Id. “On April 25, 1988, Voth, Sellers’s supervisor in the Salem office, wrote to McClure, in the Portland regional office, indicating that the police were unable to determine whether Goddard’s headlights had been on and further indicating that [the accident reconstruction firm] would attempt to determine whether the headlights had been on. On May 3, 1988, Sellers submitted a status report to Voth, explaining that a state police report concluded that Goddard’s headlights had been on and that the present demand was for $200,000. Sellers concluded, T think the case has a value of $50,000 — and feel we should make the offer now.’ On May 12, 1988, Sellers called Amala, the attorney for Goddard’s estate, indicating that he should have an answer within a few days on settlement but that he knew that defendant was ‘not going to offer policy limits.’ Amala responded that the Goddard estate intended to file suit shortly. “On May 26, 1988, Voth wrote to McClure, stating that information showed that Munson had not used a turn signal before turning left and that Goddard’s headlights were on. Voth further indicated that he believed that Goddard had recently been released from jail and that he had an alcohol problem. Voth stated, T feel it is time that we make an offer to plaintiffs attorney and asked for authority in the amount of $50,000. McClure responded to Voth that he ‘did not feel we have enough information to evaluate our *240 insured’s exposure’ and instructed Voth to seek out negative information about Goddard.” Id. Munson was convicted of criminally negligent homicide in early June 1988. Around the same time, plaintiff filed her wrongful death action against Munson. Plaintiffs attorneys notified defendant that plaintiffs offer to settle the case for $200,000 would continue in effect for 30 days. In a letter to Sellers, one of plaintiffs attorneys stressed that, “because Munson had been convicted of criminally negligent homicide, defendant was in the position of‘having to admit liability in a case that has aggravated liability justifying punitive damages.’ ” Id. at 90. In spite of that advice, however, defendant never responded to plaintiffs $200,000 settlement offer. Id. at 91. Defendant hired the law firm of Parks & Bauer to defend Munson in the wrongful death action. Id. at 90. “Voth’s letter retaining that firm described Munson’s consumption of alcohol and the fact that Goddard’s headlights had been on, but also stated that there was information that Goddard had a criminal record. Voth further noted that plaintiffs demand was for $200,000 under both policies. Voth then wrote to Munson[,] stating that defendant had retained Parks & Bauer ‘who will enter a defense and protect your interests’ in the wrongful death case. That letter acknowledged that ‘[i]t is possible that a judgment could be secured in excess of the limits of your insurance policy.’ ” Id. (second brackets in original). Munson’s attorneys in the criminal action were surprised by defendant’s apparent willingness to take the civil wrongful death action to trial. One of those attorneys, McGehee, wrote to defendant that, in light of Munson’s “manifest criminal liability’ and the strong potential for a verdict in excess of $200,000, he was reluctant to recommend that Munson go to trial on the wrongful death case. McGehee warned defendant that, if the jury returned a verdict in excess of $200,000, Munson would “be looking directly to you for payment of the overage on the basis of bad faith settlement efforts.” Id. *241 As noted, defendant already had recognized that Munson’s policy might not apply if Munson used Foley’s vehicle regularly, because the policy contained a “regular use” exception. Id. at 87. Defendant now began to work on that angle. “On July 25, 1988, Amala spoke with Keith Bauer, the insurance defense attorney whom defendant had retained to defend Munson in the wrongful death action. Bauer told Ainala that he believed that only one policy of $100,000 was available and that the criminal sentencing would reduce any punitive damages. Shortly thereafter, Bauer, who had not yet conferred with Munson, filed an answer in the wrongful death action denying that Munson had been under the influence of intoxicants and that he had caused the collision, but did not assert as a defense that Goddard was negligent in any way in causing the collision. “On July 27, 1988, [Munson’s criminal lawyer,] Feitelson, wrote to defendant, stating that neither he nor [Munson’s other criminal lawyer,] McGehee[,] had heard back from defendant but that he understood that defendant was taking the position that only one $100,000 policy was involved. Feitelson requested a copy of any coverage opinion to that effect. “On July 29, 1988, defendant retained attorney Lynn Ashcroft to provide a coverage opinion. The letter retaining Ashcroft stated that defendant suspected that Munson had had frequent use of Foley’s truck and asked whether, under those circumstances, defendant was exposed to the full $200,000 limits of both policies.” 2 Id. at 91. In the fall of 1988, a second coverage issue, pertaining to whether, at the time of the accident, Munson was a permissive user of Foley’s truck, presented itself: “In early August, Billy Sime, [one of the attorneys retained to defend Munson], attended Munson’s deposition in the wrongful death action. During the deposition, Munson stated for the first time that Foley had told him not *242 to drive her pickup when he had been drinking and that Foley had picked him up at taverns a number of times when he was too drunk to drive home. Sime then spoke privately with Foley, who told him that Munson ‘could not even have one drink of alcohol and drive her truck’ and that she had taken the truck away from Munson on previous occasions when he had been drinking. Sime subsequently wrote to defendant, stating that, although Munson’s and Foley’s statements were consistent in that they agreed that Munson was not supposed to drive the truck when he had been drinking, their interests were potentially adverse because Foley was a potential defendant in the wrongful death action on the theory that she had negligently entrusted the truck to Munson despite her knowledge of his drinking problem. “Shortly thereafter, defendant sent claims representative Mary Kay Hendrickson to interview Foley. Foley told Hendrickson that Munson had used her truck probably 100 times while working for her and that, on several occasions, he had also borrowed her truck when he was getting his car fixed. Foley also stated that she told everybody ‘no drinking when you’re driving my pickup.’ Foley told Hendrickson that she had never had any reason to believe that Munson had been driving the pickup and drinking. Foley acknowledged, however, that she had planned to meet Munson at the tavern on the evening of the collision and that she knew he would be driving her truck. “In September 1988, because of the potential conflict between Munson and Foley, defendant retained attorney Sandra Haynes to represent Foley in the wrongful death action. In the letter retaining Haynes, Voth noted that the Goddard estate was planning to add Foley as a defendant on the theory that she negligently entrusted her vehicle to Munson. Haynes spoke with Foley, who denied any knowledge of Munson’s drinking problem and any knowledge that he had driven her truck while intoxicated. Because Foley’s statements were at odds with Munson’s deposition testimony that Foley would pick him up at taverns after he had had too much to drink, Haynes believed that one of the two — probably Foley — was lying. Shortly thereafter, an amended complaint was filed in the wrongful death action, naming Foley as a defendant.” Id. at 91-92. *243 With Munson’s and Foley’s claim that Foley’s permission was conditional, defendant saw an opportunity to escape liability on both policies. “[0]n October 17, Voth sent Munson a reservation of rights letter, stating that defendant was investigating whether there was coverage under either policy. In particular, Voth indicated that there were two potential coverage disputes: (1) whether, under Foley’s policy, Munson was a covered ‘permissive user’ of Foley’s vehicle; and (2) whether Foley’s vehicle was covered under Munson’s policy because it was not a vehicle that was being temporarily used and, thus, fell within that policy’s ‘regular use’ exclusion. Shortly thereafter, defendant obtained a coverage opinion from attorney William Hallmark, who concluded that a driver who drove after consuming alcohol contrary to the owner’s express instructions was not a ‘permissive user’ of the vehicle as defined in the policy.” Id. at 92-93 (footnotes and citation omitted). Plaintiff, meanwhile, continued to seek a settlement: “In November 1988, Harris, one of the attorneys for the Goddard estate, told Foley’s attorney, Haynes, that the Goddard estate would settle with Foley for her policy limits of $100,000, and Haynes sought from defendant the authority to settle the case for that amount. Haynes believed that a $100,000 policy limits settlement was reasonable, and she did not want the case to go to trial because of her concerns about her client’s lack of candor. Haynes also recommended that defendant adjust the claims against Foley and Munson separately due to the conflicts of interest.” Id. at 93 (footnote omitted). Defendant nevertheless moved to either avoid coverage entirely, or to settle the case well under policy limits. “Heatherington [now in the Portland office] and McClure then sought authority from defendant’s home office either to file [a] declaratory judgment action [respecting liability] or to settle for $100,000. In December 1988, Voth received authorization from the home office to file the declaratory judgment action, but settlement authority was limited to $30,000. At about the same time, in a letter to Hallmark, defendant noted that Bauer had estimated the *244 value at ‘$50,000 to $75,000, with a possible range of $100,000.’ “On December 21, 1988, Ashcroft filed the declaratory judgment action on defendant’s behalf, seeking a declaration that there was no coverage under either policy. Ashcroft noted in a letter to defendant that the existence of the declaratory judgment action did not free defendant from its obligations to act reasonably to negotiate settlement within policy limits, stating that, if the claim could exceed policy limits, ‘even if coverage is in question, [it may be appropriate] to pay the policy limits into the court to be held pending the outcome of the declaratory relief action.’ ” Id. at 92-93. Ashcroft echoed Haynes’s earlier recommendation that the Foley and Munson claim files be adjusted separately. Id. at 94. While defendant did assign the Foley file to adjuster David Strand, Voth nevertheless continued to have supervisory responsibility over both the Foley and Munson files. “* * * Strand sought and received Voth’s approval before authorizing Haynes to make a settlement offer of $30,000 to settle all claims against both Munson and Foley. Although Haynes conveyed the offer, she believed that it was unacceptably low and suspected that defendant was not negotiating in good faith. Plaintiff declined that offer.” Id. (footnote omitted). Defendant never told Munson about that offer or its rejection. The parties did not make any further attempts at settlement until just before trial. Acting on Munson’s behalf, attorney Feitelson contacted attorney Bauer several times in early 1989, asking that defendant not expose Munson to liability exceeding policy limits. Feitelson never received a response. Just before the trial started in 1990, Feitelson again wrote to Bauer, suggesting that defendant was not attempting in good faith to settle the matter and warning again that Munson would bring a bad faith claim if the wrongful death judgment exceeded policy limits. Id. at 94-95. Trial was scheduled to start in January 1990. Id. at 94. With less than two weeks to go, defendant resumed its efforts to settle the case — but still for small amounts of money. Moreover, the evidence suggests that defendant *245 attempted to protect itself against any subsequent action for failure to settle by “padding the file” with lower estimates of potential liability. “On January 4,1990, Voth requested that Bauer submit an evaluation of the case so that Voth could request settlement authority * * *. “On January 11, 1990, Bauer’s firm wrote and had hand-delivered to defendant an evaluation that concluded that ‘the case has a reasonable jury value between $75,000 and $150,000 for non-economic damages plus punitive damages.’ Later that day, however, Bauer had his staff retrieve that letter from defendant’s office and replace it with a letter, also dated January 11, that deleted the amounts $75,000 and $150,000 and substituted a much lower range of exposure — $35,000 to $60,000 — for noneconomic damages. Because of Goddard’s criminal record, Bauer did not believe that a jury would award economic damages. “The wrongful death case was tried from January 16 through January 24, 1990. On the first day of trial, Bauer offered $30,000 to settle the case, and the offer was rejected. At the end of the first day of trial, Bauer increased the offer to $50,000, with the option to arbitrate any additional damages up to the policy limits established in the declaratory judgment action. [Plaintiff] countered that it would accept $100,000 with an additional $100,000 to be paid only if the declaratory judgment action established that both policies applied. Bauer responded that defendant would not agree to such a settlement. At that point, Voth was of the opinion that Munson’s exposure was between $60,000 and $150,000 ‘and maybe more.’ “On the third day of trial, Bauer offered $100,000, with the understanding that, if the declaratory judgment action determined that both policies applied, arbitration would need to occur to establish the amount between $100,000 and $200,000 that defendant would pay. That is, in contrast to the estate’s counteroffer under which the estate would automatically recover the additional $100,000 if the arbitrator determined that both policies applied, under defendant’s final offer any payment in excess of $100,000 would be further conditioned upon the arbitrator’s determining that the estate was, in fact, entitled to recover wrongful *246 death-related damages in that amount. [Plaintiff] rejected that final offer. “On January 23, 1990, the case was submitted to the jury. Bauer’s trial notes indicate that, on that day, the trial court’s judicial assistant * * * told Bauer that he should settle the case because the jury was ‘talking big dollars.’ Bauer relayed that information to Voth, who relayed it to Heatherington and McClure. McClure decided not to increase the settlement offer. * * * “On January 24, 1990, the jury returned a verdict against Munson for $188,274 in economic damages, $425,000 in noneconomic damages, and $250,000 in punitive damages — a total of $863,274.” Id. at 95-96 (footnote omitted). Defendant appealed the ensuing judgment, but the Court of Appeals affirmed. As mentioned earlier, defendant then brought a separate declaratory judgment action seeking declarations that the accident that killed Goddard was not covered under either Munson’s or Foley’s insurance policies. After a number of decisions, appeals, and remands in that action, the Court of Appeals determined that the accident was covered by Foley’s policy, but not Munson’s. Defendant chose not to appeal that decision and finally, in 1998, paid plaintiff $175,960, representing the $100,000 limit in Foley’s policy, plus interest. See id. at 84 (noting payment). B. Action Against Defendant for Failure to Settle within Policy Limits After the trial court entered judgment against him in the wrongful death action, Munson assigned to plaintiff all of his potential claims against defendant. Id. at 83. In May 1990, plaintiff brought the present action against defendant under the assignment, contending that defendant breached its duty to Munson to make reasonable efforts to settle the wrongfiil death action. Id. at 83, 85. Plaintiff sought, as compensatory damages, the $863,274 awarded in the wrongful death action, plus $450 million in punitive damages. Id. at 83. *247 In addition to the evidence summarized above, plaintiff presented expert testimony regarding defendant’s conduct in handling the wrongful death case: “James Nelson, an attorney with extensive experience in insurance cases, testified that insurance companies sometimes engage in the practice of ‘stonewalling,’ which involves refusing to negotiate settlement because claimants often have very limited resources and mounting medical and legal bills; ‘stonewalling’ puts psychological and financial pressure on such claimants to accept settlements that are less than reasonable. Nelson further testified that another negotiation strategy, ‘low-balling,’ involves making minimal settlement offers to signal that the insurance company is willing to fight the claim and to lower the claimant’s expectations. In Nelson’s opinion, defendant had used both strategies in handling the Goddard claim. According to Nelson, the $30,000 offer was a ‘low-ball’ offer because the Goddard claim should, reasonably, have been valued at over $200,000. “Plaintiff also presented expert testimony from John Partlow, who rendered an opinion as to whether the claim against Munson was handled properly. Partlow had worked as a claims auditor for a large insurance company. In that position, he examined claims files to determine if the claims had been handled properly, and he subsequently performed similar functions as a consultant. Partlow testified that coverage limits should be communicated promptly and that mistakes about coverage limits should be disclosed promptly. He further testified that defendant’s failure to respond to the time-limited settlement demand violated the company’s own policies and did not comport with industry standards. Partlow concluded that defendant had failed to make a timely and sufficient investigation of the collision and had failed to negotiate fairly throughout its handling of the Goddard claim — and, indeed, that, in his 15 years of evaluating claims, the handling of the Goddard claim was among the worst he had ever seen. “Plaintiff also presented expert testimony from Jeffrey J acobs, an attorney who had significant experience working as in-house counsel in Oregon for an insurance company. Jacobs testified that defendant’s conduct in failing to respond to communications from the attorneys for the Goddard estate for many months after the collision fell *248 below the industry standard of care. He further testified that a decision to let the claim ‘ripen’ rather than respond to an attorney under these circumstances very seriously jeopardizes the protection of the person who has paid the insurance premiums. In Jacobs’s view, liability and the value of the case as in excess of policy limits was clear by June 1988, when defendant failed to respond to the time-limited demand that was made by the Goddard estate, and failing to respond was not in keeping with industry standards. According to Jacobs, the case should have been valued in the range of $500,000 to $750,000. “Plaintiff also presented evidence of defendant’s use of ‘low-balling’ tactics in another case involving a collision where an injury resulted from the conduct of a drunk driver insured by defendant. In that case, in which Bauer also represented the insured, the injured party accepted an offer of less than policy limits due to financial pressures, even though his attorney believed the case to be worth more than policy limits. That attorney testified that Bauer had told him, ‘Voth doesn’t pay policy limits.’ ” Id. at 96-98 (footnote omitted). The jury returned a special verdict for the plaintiff. The jury concluded that defendant was 80 percent at fault for entry of the $863,274 wrongful death judgment against Munson. The jury also concluded that Munson’s actions after the accident made him 20 percent responsible for the wrongful death judgment. The jury determined that, because of its wrongdoing in failing to settle the wrongful death case, defendant should pay plaintiff $20,718,576 in punitive damages. In its judgment, the trial court reduced the compensatory damages award in several ways. Starting with the $863,274 awarded against Munson in the wrongful death judgment, the trial court subtracted 20 percent for Munson’s comparative fault, as determined by the jury. The court also subtracted defendant’s 1998 payment of $175,960, i.e., the $100,000 policy limit plus interest. In doing so, the court credited $100,000 of that payment against principal, rather than interest. Finally, the trial court subtracted $325,000, an amount that had been paid to plaintiff on an underinsured motorist policy that she held through another insurer. On the *249 other hand, and although defendant contended that it was unconstitutionally large, the trial court refused to reduce the punitive damages award. In sum, the trial court entered judgment against defendant for “$265,619.20 economic damages, plus prejudgment interest thereon in the amount of $343,573.62; and $20,718,576 in punitive damages plus interest thereon.” Id. at 83-84. Plaintiff and defendant both appealed to the Court of Appeals. Defendant contended that the trial court erred in admitting evidence of a “pooling agreement” between defendant and other related insurance companies. Defendant also argued that the punitive damages award was unconstitutionally large. Plaintiff contended that the trial court erred in calculating the amount of compensatory damages. Id. at 84. The Court of Appeals rejected defendant’s evidentiary argument. Id. at 100-03. Respecting the issue relating to the calculation of compensatory damages, the court agreed with plaintiff. The court concluded that defendant was not entitled to a $325,000 credit for the amounts paid by the other insurance company. Id. at 103-07. The court also held that defendant’s entire $175,960 payment to plaintiff should have been applied to interest. Id. at 107-10. While leaving the final calculation of interest to be made by the trial court on remand, the court estimated “total economic damages” at approximately $1,280,000, representing $690,619.20 in principal (80 percent of the $863,274 awarded in the wrongful death judgment), plus approximately $589,000 in prejudgment interest. Id. at 110. The Court of Appeals also agreed with defendant that the punitive damages award was grossly excessive and therefore unconstitutional under the Due Process Clause of the Fourteenth Amendment to the United States Constitution. Id. at 111-22. Finally, the court concluded that the highest permissible punitive damages award that the jury could have made was three times the compensatory damages award (as correctly calculated). Id. at 121-22. Defendant petitioned the Court of Appeals to reconsider, arguing (among other things) that the court had erred in the way in which it had dealt with the roughly $589,000 in prejudgment interest on the $690,619.20 of principal. *250 Goddard v. Farmers Ins. Co., 203 Or App 744, 746, 126 P3d 682 (2006). Defendant noted that the trial court’s judgment identified the prejudgment interest separately from plaintiffs “economic damages” and that the court had refused a proposed form of judgment that would have lumped prejudgment interest and principal together. Defendant contended that punitive damages had to be calculated using only the $690,619.20 in principal identified as “economic damages” in the judgment. Doing otherwise, defendant claimed, gave plaintiff relief on a matter that she did not assign as error on cross-appeal. The Court of Appeals agreed that it should not have referred to the entire $1,280,000 as “economic damages.” The court altered its earlier opinion to refer to that amount as “compensatory loss.” But the court otherwise rejected defendant’s argument, concluding that the prejudgment interest should be included in the amount used to evaluate the constitutionality of the punitive damages award. Id. at 747-49. Plaintiff sought review, challenging the Court of Appeals’ determination that the punitive damages award was unconstitutional, and we allowed review. II. DISCUSSION As a preliminary matter, we note that, although defendant has not cross-petitioned this court for review of the Court of Appeals decision, it has renewed, in its arguments to this court, certain of the claims of error that it raised on appeal (and that the Court of Appeals rejected). Although our rules permit us to consider claims of error raised in that manner, ORAP 9.20(2), we decline to do so in this case: We allowed plaintiffs petition for review to consider a particular set of issues pertaining to punitive damages, and we choose to focus only on those issues. We thus affirm the Court of Appeals decisions with respect to claims of error that are unrelated to the punitive damages award (specifically, defendant’s claims that the trial court erred in admitting evidence of a “pooling agreement” and in declining to reduce plaintiffs damages by the amount that she received from her own underinsured motorist insurance policy), without expressing any opinion concerning the Court of Appeals’ analysis of those claims. *251 A. Determining When a Punitive Damages Award Violates the Due Process Clause We turn, then, to the question that prompted our decision to allow review — whether the Court of Appeals erred in determining that the jury’s punitive damages award was grossly excessive and that the maximum constitutionally permissible punitive damages award was three times the amount that the jury awarded in compensatory damages. Plaintiff contends that the Court of Appeals did err and that, if all of the relevant circumstances are taken into account, it is clear that the jury’s award was within the permissible range. Defendant responds that the punitive damages award was excessive and that, if anything, the maximum constitutionally permissible punitive damages award is even smaller than the Court of Appeals determined. Punitive damages awards that are “grossly excessive” violate the Due Process Clause of the Fourteenth Amendment to the United States Constitution, because excessive punitive damages serve no legitimate purpose and constitute arbitrary deprivations of property. BMW of North America, Inc. v. Gore, 517 US 559, 568, 116 S Ct 1589, 134 L Ed 2d 809 (1996). Excessive punitive damages also implicate the requirement of fair notice that is enshrined in the Due Process Clause: Due process dictates that “a person receive fair notice not only of the conduct that will subject him to punishment, but also of the severity of the penalty that a state may impose.” Id. at 574. The United States Supreme Court has identified three “guideposts” that should be considered by appellate courts charged with determining whether a jury’s punitive damages award comports with the Due Process Clause. They are: “(1) the degree of reprehensibility of the defendant’s misconduct; (2) the disparity between the actual or potential harm suffered by the plaintiff and the punitive damages award; and (3) the difference between the punitive damages awarded by the jury and the civil penalties authorized or imposed in comparable cases.” State Farm Mut. Ins. Co. v. Campbell, 538 US 408, 418, 123 S Ct 1513, 155 L Ed 2d 585 (2003). See also Williams v. Philip *252 Morris Inc., 340 Or 35, 55, 127 P3d 1165 (2006), vac’d on other grounds, 549 US 346, 127 S Ct 1057, 166 L Ed 2d 940 (2007), on remand, 344 Or 45, 176 P3d 1255 (2008) (setting out those guideposts and using them to analyze Oregon punitive damages award for excessiveness). Two of the Court’s recent cases — Gore and Campbell — elaborate on those factors and illustrate how they are to be applied in determining whether, and to what extent, a jury’s punitive damages award violates a tortfeasor’s due process rights. In Gore, the Court considered an Alabama jury’s verdict awarding punitive damages in an amount that was 1,000 times the amount of the plaintiffs compensatory damages. The case arose out of a lawsuit by a car owner, who sued the manufacturer of his new BMW automobile because the manufacturer had failed to disclose to him that the vehicle had been repainted prior to sale (the original paint job had been marred en route to the United States). 517 US at 563. The jury awarded the car owner $4,000 in compensatory damages and $4 million in punitive damages, id. at 565, but the Alabama Supreme Court reduced the punitive damages award to $2 million, largely because it concluded that the jury had unlawfully based much of its award on BMW’s out-of-state conduct, when it was unclear whether that conduct was unlawful where it occurred. Id. at 567. On review, the United States Supreme Court agreed with the Alabama Supreme Court that the punitive damages award could not be used to punish BMW for, or to deter, conduct that was lawful in the state where it occurred. Id. at 571-74. But the Court also concluded that, even with the Alabama Supreme Court’s adjustment to the jury’s award, the punitive damages award was grossly excessive. Although the court warned that there was no “ ‘mathematical bright line between the constitutionally acceptable and the constitutionally unacceptable that would fit every case,’ ” id. at 583 (quoting Pacific Mut. Life Ins. Co. v. Haslip, 499 US 1, 18, 111 S Ct 1032, 113 L Ed 2d 1 (1991)), it nevertheless set out a framework for determining when punitive damages awards are grossly excessive — specifically, the three “guideposts” mentioned above. *253 As to the first guidepost, reprehensibility, the Court described a number of “aggravating factors associated with particularly reprehensible conduct.” Gore, 517 US at 576. Those aggravating factors (as the Court later summarized them in Campbell) include whether “the harm caused was physical as opposed to economic; the tortious conduct evinced an indifference to or a reckless disregard of the health or safety of others; the target of the conduct had financial vulnerability; the conduct involved repeated actions or was an isolated incident; and the harm was the result of intentional malice, trickery, or deceit, or mere accident. The existence of any one of these factors weighing in favor of a plaintiff may not be sufficient to sustain a punitive damages award; and the absence of all of them renders any award suspect.” Campbell, 538 US at 419 (citation omitted). Applying those aggravating factors to the facts, the court concluded that BMW’s conduct had not been particularly reprehensible: No physical harm or threat to safety or health had been involved, the plaintiff was not financially vulnerable, and there was no clear evidence of intentional recidivism or a malicious motive. Gore, 517 US at 575-80. With respect to the second guidepost, the Court noted that even the reduced $2 million punitive damages award was 500 times the amount of actual harm, id. at 582— a ratio that the Court described as “breathtaking,” and which “must surely raise a suspicious judicial eyebrow.” Id. at 583 (internal quotation marks and citation omitted). Finally, with respect to the third guidepost, the Court observed that there was an enormous difference between the $2 million punitive damages award and the civil penalty that could be imposed in the only comparable class of cases — a $2,000 fine. Id. at 584. Based on its analysis, the Court concluded that the jury’s $4 million punitive damages award, and even the Alabama Supreme Court’s $2 million remitted award, were unconstitutionally large. Id. at 585-86. However, the court did not determine what size award would be constitutional. Rather, it left that question to the state court to decide in the first instance. Id. at 586. *254 Several years later, in Campbell, the Court again applied the three guideposts to decide whether a punitive damages award comported with due process. The basic fact pattern in Campbell, a Utah case, was similar to the present case: After an insurer failed to settle a liability lawsuit against its insured within policy limits, the case went to trial and the jury returned a verdict that was significantly in excess of the policy limits. 538 US at 412-13. The insured sued the insurer for bad faith, fraud, and intentional infliction of emotional distress. Id. at 414. A jury awarded the insured $2.6 million in compensatory damages and $145 million in punitive damages. The trial court reduced those awards to, respectively, $1 million and $25 million. The Utah Supreme Court reinstated the $145 million punitive damages award. Id. at 415. The United States Supreme Court’s analysis in Campbell essentially tracked its reasoning in Gore. The Court began with the first Gore guidepost, reprehensibility. In examining that factor, the court excluded from consideration much of the conduct that the state court had relied on in its analysis — specifically, the defendant’s out-of-state conduct and conduct that was not similar to the underlying tort, which the plaintiff had argued was evidence of a nationwide policy of capping payouts on claims of all kinds. 538 US at 420-24. The Court noted, as it had in Gore, that a state may not legitimately “punish a defendant for unlawful acts committed outside of the State’s jurisdiction.” Id. at 421. Furthermore, the Court concluded, the punitive damages award should have been based only on similar conduct: “A defendant should be punished for conduct that harmed the plaintiff, not for being an unsavory individual or business.” Id. at 423. Given those parameters, the Court concluded, “the conduct that harmed [plaintiffs was] the only conduct relevant to the reprehensibility analysis.” Id. at 424. With regard to that conduct, the Court suggested that a more “modest” punitive damages award “could have satisfied the state’s legitimate objectives, and the [state] courts should have gone no further.” Id. at 419-20. Regarding the second guidepost, the ratio between punitive damages and actual or potential harm, the Court *255 declined to “impose a bright-line ratio which a punitive damages award cannot exceed.” Id. at 425. Still, it observed that, “in practice, few awards exceeding a single-digit ratio between punitive and compensatory damages, to a significant degree, will satisfy due process. In [Pacific Mutual Life v.]Haslip, [499 US 1, 111 S Ct 1032, 113 L Ed 2d 1 (1991)] in upholding a punitive damages award, we concluded that an award of more than four times the amount of compensatory damages might be close to the line of constitutional impropriety. We cited that 4-to-l ratio again in Gore. The Court further referenced a long legislative history, dating back over 700 years and going forward to today, providing for sanctions of double, treble, or quadruple damages to deter and punish. While these ratios are not binding, they are instructive.” Id. (citations omitted). Having said that, the Court concluded that there was a strong presumption against a punitive damages award involving a 145-to-l ratio, like the one before it. Id. at 426. With respect to the third guidepost — the relationship between the punitive damages award and civil or criminal penalties for comparable conduct — the Court began by discounting the value of criminal penalties to the analysis: “Great care must be taken to avoid use of the civil process to assess criminal penalties that can be imposed only after the heightened protections of a criminal trial have been observed, including, of course, its higher standards of proof.” Id. at 428. In the case at issue, the Court concluded, the only relevant civil sanction was a $10,000 fine applicable to certain acts of fraud, which the Court characterized as “an amount dwarfed lay the $145 million punitive damages award.” Id. Taking all three of the guideposts into account, the Court concluded that the $145 million award was an irrational and arbitrary deprivation of property and, as such, a violation of due process. In addition, for the first time, the Court suggested what the constitutionally permissible maximum award might be. “An application of the Gore guideposts to the facts of this case * * * likely would justify a punitive damages award at or near the amount of compensatory damages.” Id. at 429. Still, the Court did not intend that estimate *256 to be binding. It held that “the proper calculation of punitive damages under the principles we have discussed should be resolved, in the first instance, by the Utah courts.” Id. B. Determining the Maximum Constitutionally Permissible Punitive Damages Award In Gore and Campbell, the Supreme Court has provided some guidance for determining when a jury’s punitive damages award is so grossly excessive as to violate the Due Process Clause. But, thus far, the Court has avoided the more difficult question that due process challenges often present, viz., how to determine the maximum constitutionally permissible punitive damages award that may be allowed in a particular set of circumstances. In Gore and Campbell — the only two cases in which the Court struck down punitive damages awards as grossly excessive — the Court left it to the state courts to determine on remand what amount of punitive damages would be constitutional. See Campbell, 538 US at 429 (“the proper calculation of punitive damages under the principles we have discussed should be resolved, in the first instance, by the Utah courts”); Gore, 517 US at 586 (“[w]hether the appropriate remedy requires a new trial or merely an independent determination by the Alabama Supreme Court of the award necessary to vindicate the economic interests of Alabama consumers is a matter that should be addressed by the state court in the first instance”). Thus, for purposes of civil actions arising in the courts of this state, it falls to this court to answer that question — that is, to refine the Court’s guideposts in a way that “yield[s] * * * real answers in * * * real cases.” Gore, 517 US at 606 (Scalia, J., dissenting). Of course, in doing so, we remain mindful of the fact that we are dealing with a federal challenge and must rely entirely on federal due process considerations. As we previously have observed, “[t]here is * * * no ‘state law excessiveness challenge’ under the Oregon Constitution.” Parrott v. Carr Chevrolet, Inc., 331 Or 537, 551, 17 P3d 473 (2001) (footnote, quotation marks, and citations omitted). It is clear that that analysis must begin, at least, with the aforementioned guideposts from Gore and Campbell. It is less clear how those guideposts can yield a *257 specific numerical figure that represents the maximum constitutionally permissible award. In that regard, we note that the first guidepost, reprehensibility, does not generate numerical answers at all, because the guidepost itself, and the “subfactors” that go into it, are all qualitative, not quantitative. Still, we may compare the level of reprehensibility exhibited in various cases, and that comparison may lead us to a conclusion that the constitutionally permissible limit in a particular case is “high” or “low,” relative to the limit in another case. The third guidepost, which examines comparable sanctions, also fails to provide a quantitative measuring stick. In Williams, this court discussed how Campbell had applied the third guidepost: “There, the most relevant civil sanction was $10,000; the Court found that civil sanction was ‘dwarfed’ by the $145 million punitive damage award. Yet the Court approved a punitive damage award ‘at or near the amount of compensatory damages,’ i.e., an award somewhere around $1 million. And a $1 million punitive damage award was still 100 times the comparable sanction.” 340 Or at 58 (citations omitted). Ultimately, we concluded that the Court must have intended that a pertinent civil sanction be used as a basis for comparing one punitive damages award to another, and not as a direct predictor of the constitutional limits of an individual punitive damages award: “The guidepost may militate against a significant punitive damage award if the state’s comparable sanctions are mild, trivial, or nonexistent. However, the guidepost will support a more significant punitive damage award when the state’s comparable sanctions are severe.” Id. at 58. We conclude that the second guidepost, i.e., the ratio between punitive damages and actual or potential harm to the plaintiff, comes closest to providing numerical limits. Although the Court has professed its reluctance to “identify concrete constitutional limits on the ratio between harm, or potential harm, to the plaintiff and the punitive damages award,” Campbell, 538 US at 424, it has also, at various *258 times, alluded to specific numerical ratios. In Haslip, 499 US at 23-24, for example, the Court stated that a punitive damages award that is more than four times the compensatory damages award “may be close to the [constitutional] line.” In Gore, 517 US at 581, the court interpreted a prior decision, TXO Production Corp. v. Alliance Resources Corp., 509 US 443, 113 S Ct 2711, 125 L Ed 2d 366 (1993), as suggesting that the ratio could not exceed ten to one. In Campbell, 538 US at 425, the Court suggested that, although a higher ratio may be justified when compensatory damages are small, “[w]hen compensatory damages are substantial, * * * a * * * ratio [that is] perhaps only equal to compensatory damages * * * can reach the outermost limit of the due process guarantee.” And, in the same opinion, the Court indicated that due process may be violated by “exceeding a single-digit ratio between punitive and compensatory damages, to a significant degree.” Id. at 425. But the Court also has conceded that the single-digit ratio may be exceeded in certain circumstances. It has indicated, for example, that higher ratios may be appropriate if “a particularly egregious act has resulted in only a small amount of economic damages,” and in “cases in which the injury is hard to detect or the monetary value of noneconomic harm might have been difficult to determine.” Gore, 517 US at 582. This court has identified one other circumstance in which an award that exceeds a single-digit ratio between punitive and compensatory damages might satisfy due process. In Williams, we held that “extraordinarily reprehensible” conduct on the part of the defendant may provide a basis for overriding concerns that may arise from an award that exceeds a single-digit ratio. 3 340 Or at 63. *259 In sum, the second guidepost, although far from perfect, does provide at least a rough numerical baseline for calculating a maximum constitutionally permissible punitive damages award in a given case. Specifically, it suggests that due process normally will not permit a punitive damages award in excess of a single-digit ratio to the compensatory damages award. Once that rough numerical reference point is established, the other guideposts come into play. As we have suggested, the first (reprehensibility) guidepost primarily is a comparative tool: If the level of reprehensibility exhibited in a case is high relative to other cases, then so will be the constitutionally permissible punitive damages award. The third guidepost also is comparative: If the comparable civil sanctions for certain conduct is high relative to other civil sanctions, then that fact may justify an upward adjustment in the constitutionally permissible amount of punitive damages. Such comparisons can be made to fit, albeit only approximately, into the numerical baseline that the second guidepost has established (which is based on the ratio between compensatory and punitive damages). Thus, punitive damages at what normally is the highest constitutionally permissible level (i.e., a nine-to-one ratio between punitive and compensatory damages) are justified only when the conduct at issue is highly reprehensible (based on the subfactors that make up the reprehensibility analysis) and, if relevant, when that conduct subjects the actor to civil sanctions that are high relative to other civil sanctions that the state may impose. Of course, the foregoing framework is still amorphous. We offer one refinement in the hope of making it less so. 4 By long tradition, and reinforced by existing legal standards at all levels of government, conduct that causes or creates a risk of physical harm to persons is punished more *260 harshly than conduct that causes or potentially causes economic harm. The Supreme Court has recognized that distinction in two of the “subfactors” that contribute to the analysis of reprehensibility (the first of the three Gore guideposts). As we already have observed, reprehensibility depends, to a large degree, on whether “the harm caused was physical as opposed to economic” and on whether “the tortious conduct evidenced an indifference to or a reckless disregard of the health or safety of others.” Campbell, 538 US at 419. We would expect that distinction to be strongly reflected in the punitive damages awards that attach to the two kinds of harms/risks, as well as in assessments by appellate courts of the level at which such awards become excessive. In fact, we think that, if a nine-to-one ratio between punitive and compensatory damages is at the limit of what the Due Process Clause normally will allow a court to impose on any tortious conduct, then the limit must be significantly lower for conduct that causes or risks only economic injury. But what is that limit? With respect to cases (like this one) involving injury to economic interests, our best estimate points to the four-to-one ratio between punitive and compensatory damages that the Supreme Court sometimes has identified as bordering on excessiveness. See Haslip, 499 US at 22-23 (stating that four-to-one ratio between punitive and compensatory damages was close to, but did not cross, the line of constitutional impropriety). See also Gore, 517 US at 581 (referring to four-to-one ratio discussed in Haslip). As the Court has observed, that ratio is consistent with the double, triple, and quadruple damages that English and American statutes have imposed for various analogous civil violations. Gore, 517 US at 581. We conclude that, as a very general rule of thumb, the federal constitution prohibits any punitive damages award that significantly exceeds four times the amount of the injured party’s compensatory damages, as long as the injuries caused by the defendant were economic, not physical. Of course, as with the single-digit ratio discussed above, 344 Or at 258-59, a higher ratio might be appropriate *261 in unusual circumstances, such as when the economic damages are relatively small in relation to the theoretical injury, where the injury is difficult to detect, where the conduct causes noneconomic harm that is hard to value, or where the conduct is “extraordinarily” reprehensible. C. Standard of Post-Verdict Judicial Review Having established the foregoing general standards for assessing punitive damages awards and the maximum award that due process will allow in given circumstances, we turn to still another preliminary question: What is the proper standard for post-verdict judicial review of a punitive damages award? On balance, and based on our review of the cases, we conclude that the proper methodology for performing a review for gross excessiveness is the one used by the Court of Appeals in its opinion in this case. See Goddard, 202 Or App at 112 (describing that process). First, to determine whether there is a factual predicate for a punitive damages award, the court reviews the evidence in the record that is relevant to such an award in the light most favorable to the party who won the award. 5 Second, the court applies the constitutionally prescribed guideposts to those predicate facts to determine if, as a matter of law, the award is grossly excessive. If *262 the court determines that the award is grossly excessive, it then uses the same guideposts to determine the highest lawful amount of punitive damages that a rational juror could award, consistent with the Due Process Clause, i.e., it reviews for “gross excessiveness.” Of course, in applying that standard, the reviewing court must in some sense reexamine the evidence in the record — not to redecide the historical facts as decided by the jury, but to decide where, for purposes of the Gore guideposts, the conduct at issue falls on the scale of conduct that does or might warrant imposition of punitive damages. D. Was the Jury’s Punitive Damages Award Grossly Excessive in this Case? Having described the methodology and standard of review that courts are to apply in reviewing a punitive damages award for gross excessiveness, we turn to the task of applying them in the present case. We begin by describing the historical facts that a rational juror could find, 6 based on the evidence in the record. In so doing, we agree with and adopt the excellent summary in the opinion of the Court of Appeals: “Given the evidence, a rational juror could find the following facts: (1) Defendant, in calculated fashion, engaged in a protracted course of conduct — from its initial ‘stonewalling’ and ‘low-balling’ tactics through its fraudulent manipulation of the claims evaluation process and its refusal to settle even at trial — that exposed its insured to the virtual certainty of a devastating excess verdict in a ‘no defense’ case. (2) Defendant willfully engaged in such conduct, heedless of its insured’s interests and in cynical violation of its obligations to its insured, for its own selfish purposes of building and maintaining a reputation for ‘toughness’ in claims adjustment and settlement. (3) Defendant’s conduct towards Munson was intentional, deceitful, and malicious. (4) Defendant’s hardball tactics in this case were typical, not merely an isolated instance * * Goddard, 202 Or App at 112-13. *263 Next, we employ the applicable legal criteria (including the three Gore guideposts) to determine if, as a matter of law, the jury’s punitive damages award is grossly excessive. The parties have framed their arguments entirely in terms of the three Gore factors. We turn to those factors and the parties’ arguments respecting them. For its part, defendant begins its discussion with an argument that, defendant asserts, applies generally to the Gore factors: Defendant notes that one of the due process concerns that the Gore analysis is designed to address is the idea of “fair notice.” 7 Defendant suggests that due process requires that defendants have fair notice that the law considers their conduct to be tortious and, as such, subject to punitive damages, before such damages may be imposed. In the alternative, defendant suggests that lack of fair notice of the possibility of punitive damages should at least “mitigate” the reprehensibility of the conduct under consideration. Defendant then argues that, when the conduct at issue here occurred, it had no reason to know that such conduct might be a proper subject of punitive damages. Defendant contends that, at that time, an insurer’s bad faith failure to settle an insurance claim was considered to be simply a breach of contract and did not carry the kind of moral and legal stigma that is associated with torts and that justifies the practice of awarding punitive damages. Defendant’s argument rests heavily on this court’s opinion in Farris v. U.S. Fid. and Guar. Co., 284 Or 453, 587 P2d 1015 (1978) — a case that, according to defendant, was this court’s final word on punitive damages in breach of insurance contract cases until this court offered an entirely contrary view in Georgetown Realty v. The Home Ins. Co., 313 Or 97, 831 P2d 7 (1992) — two years after defendant failed in its duty to plaintiff in this case. 8 In Farris, the court considered whether an insured’s action against his insurer for damages resulting from the insurer’s denial of liability insurance *264 coverage sounded in tort, so that the insured could recover for emotional distress caused by that denial. The court concluded that such denials of coverage are a breach of contract only and support only normal contract damages. However, the Farris court distinguished between an insurer’s bad faith failure to defend and a bad faith failure to settle: “Assuming, but not deciding, that a cause of action for failure to settle within the policy limits is one in tort, it is our opinion that the rationale of such an action has no application to the present situation and that the present action is not one in tort. In an action for failure to settle within the policy limits, the insurance company is charged with acting in a fiduciary capacity as an attorney in fact representing the insured’s interest in litigation. The company’s interest comes into conflict with that of the insured’s while representing him; and, arguably, acting in its own interests to the detriment of the insured’s interest while acting in such a fiduciary capacity is a tort. In the present case defendant did not undertake in this fiduciary duty to represent the insured’s interest in the litigation — it refused it. It did not, in the course of representing plaintiffs, violate its fiduciary duty arising out of the sole control of the settlement. It never undertook any fiduciary duty by purporting to act in the interests of the insured.” Id. at 459-60 (footnote omitted). Thus, although Farris did hold that a liability insurer’s refusal to defend its insured at all was purely a breach of contract, it did not apply that same holding to insurers who, like defendant, failed to settle claims against their insureds within policy limits. In fact, it expressly left that question open, and even suggested a rationale for treating a failure to settle within policy limits as a tort. It follows that, contrary to defendant’s assertion, defendant was on notice, at the relevant time, that its refusal to settle plaintiff s action against Munson within policy limits may well not have been immune from punitive damages. *265 We do not rest our holding respecting fair notice solely on distinctions between Farris and Georgetown Realty, however. The legislature also has spoken in a relevant way on this issue. Specifically, long before defendant engaged in the conduct at issue, the legislature had enacted the Oregon Insurance Code, which includes a provision that forbids an insurer to use the kind of tactics that this defendant used. See ORS 746.230 (providing that no insurer “shall commit or perform” an “unfair claim settlement practice],” including “[n]ot attempting, in good faith, to promptly and equitably settle claims in which liability has become reasonably clear”). Defendant was (and is) an insurance company, permitted to do business as such in Oregon. Thus, it was subject at all relevant times to the Code, the provisions of which it is presumed to be aware, and it therefore cannot claim that it was unaware that its actions in handling Munson’s claim were illegal. Defendant’s suggestion that there is a fair notice issue at play here that should diminish or altogether eliminate punitive damages that otherwise might be permissible under the three Gore factors is not well taken. We turn, then, to the first Gore factor — reprehensibility. As we already have observed, “reprehensibility” under Gore is a function of at least five separate aggravating factors: whether the resulting harm was physical as opposed to economic; whether the conduct evinced indifference to or a reckless disregard of the health or safety of others; whether the target of the conduct was financially vulnerable; whether the conduct involved repeated actions; and whether the harm resulted from intentional malice, trickery, or deceit. The parties disagree, to one extent or another, about the applicability of each of those factors. At the outset, defendant contends that its conduct did not cause physical harm to either plaintiff or Munson and did not involve a reckless disregard for the health or safety of others. Plaintiff argues, to the contrary, that when an insurer refuses to cover an insured’s loss, the insured is denied the peace of mind that the insurer promised and almost certainly suffers emotional and even physical trauma as a result. We think, however, that it is a sufficient answer to plaintiff’s argument in this case to note that plaintiffs complaint did not seek, and the trial court’s instructions to the jury and the special verdict form submitted with those instructions did *266 not authorize, the jury to award damages for physical harm or emotional distress to Munson, plaintiff, or anyone else. The first two aggravating factors from Gore are not a part of this case. We turn to the remaining three. Plaintiff contends that Munson was financially vulnerable and that defendant’s conduct left him with a large excess liability verdict that would have been financially devastating if he had been required to pay it. Defendant acknowledges that Munson had no appreciable assets at the time of the accident, but it contends that the “financial vulnerability” factor is not implicated because there is no evidence that defendant targeted Munson because of the condition of his finances. We agree that there is no evidence that defendant specifically targeted Munson for that reason, but there is plenty of evidence to permit the inference that defendant knew that Munson was vulnerable, and engaged in its stonewalling tactics despite Munson’s vulnerability. In any event, we do not read the cases to require that the defendant solely or even specifically targeted the victim because of his or her financial vulnerability. It was enough, we think, that defendant knew of Munson’s vulnerability, and therefore fairly may be deemed to have been indifferent to the harm that its actions would cause him. Munson met this criterion. The parties also dispute whether defendant’s conduct involved “repeated actions” within the meaning of Gore. Plaintiff contends that that aggravating factor applies because defendant engaged in a “course” of conduct that was injurious to Munson over a number of years. Defendant contends that a course of conduct against a single individual does not qualify as “repeated actions.” It argues that the “repeated actions” factor is directed at the problem of recidivism and applies only if the actor has engaged in identical transgressions against different individuals in the past and has been punished for or otherwise notified of the wrongfulness of that conduct. The federal cases provide little support for the theoretical foundation of defendant’s argument. As the United States Supreme Court stated in Gore: “[E]vidence that a defendant has repeatedly engaged in prohibited conduct while knowing or suspecting that it was *267 unlawful would provide relevant support for an argument that strong medicine is required to cure the defendant’s disrespect for the law.” 517 US at 576-77 (emphasis added). Many circumstances may supply the necessary warning. For example, as we have explained, defendant had adequate notice that its misconduct was unlawful in this state under ORS 746.230(l)(f) before it engaged in the conduct at issue here. We agree with plaintiff that, even treating Munson’s case in isolation, a modest level of “repeated action” is present here. Munson was not stonewalled once; he was stonewalled many times, over a period of years. Such actions by defendant seem “repeated” to us. Moreover, Munson’s case was not an isolated incident. A rational jury could have inferred from the evidence that, as one lawyer had said, “Voth doesn’t pay policy limits” — i.e., that “low-balling” insureds’ claims was a common practice of defendant. That leaves us to consider the last of the five reprehensibility factors — whether the harm the defendant’s conduct caused resulted from “intentional malice, trickery, or deceit.” Defendant essentially concedes that, under the facts as the jury was entitled to find them, it acted with malice, i.e., intentionally, without just cause or excuse, in its dealings with Munson. Plaintiff argues that defendant’s conduct and state of mind was far worse than that. Plaintiff insists, in fact, that the Court of Appeals did not sufficiently capture defendant’s malevolence when it recognized that defendant had engaged in a “calculated course of conduct betraying its obligation to its insureds,” that it acted in “cynical violation of its obligations to its insured and for its own selfish purposes,” and that its conduct was “intentional, deceitful and malicious.” We believe, however, that the Court of Appeals’ assessment sufficiently conveys the deplorable nature of the thought processes of defendant’s minions and correctly places them on the scale of reprehensibility. In summary, we conclude that defendant’s actions were directed at a financially vulnerable victim, were not confined to this victim alone, and involved intentional malice and deceit. On the other hand, defendant’s actions caused economic harm only and did not evince a reckless disregard *268 for the health or safety of others, although defendant is entitled to little credit for either factor — cases of economic harm like the present one seldom provide malefactors with an opportunity to meet either of those criteria. Taken together, our analysis of the five reprehensibility factors set out in Gore, considered in light of the economic nature of defendant’s wrongdoing, leads us to conclude that defendant’s actions were very reprehensible. Turning to the second Gore guidepost, we consider the ratio between the punitive damages award and the “actual and potential harm suffered by the plaintiff.” Campbell, 538 US at 418. To do so, we first must identify the amount of actual or potential harm, because that figure is a basis for calculating the constitutionally permissible punitive damages award. The Court of Appeals concluded that the correct amount was $1,280,000 — the part of the total economic damages award that the jury attributed to defendant ($690,619.20) plus interest. Both parties argue that we should use a different figure. Plaintiff argues that we should use the full economic damage verdict ($863,274) without subtracting the 20 percent comparative fault that the jury attributed to Munson. The full amount of the wrongful death judgment, plaintiff contends, is the measure of Munson’s (and hence plaintiffs) potential harm. We disagree. The jury determined that Munson was 20 percent responsible for the large judgment in the wrongful death case. Defendant is not responsible for paying that part of the compensatory damages that resulted from Munson’s misconduct; neither should defendant be made to pay extra punitive damages in proportion to Munson’s own misconduct. Plaintiff argues in the alternative that the potential harm includes the amount of compensatory damages that plaintiff sought in the wrongful death case ($318,000 more than the $863,274 that the jury awarded), on the theory that the jury in the wrongful death action might have awarded that higher amount. Plaintiff misconceives what the Court meant when it wrote of “potential” harm: That concept has nothing to do with the amount that a jury could conceivably have awarded to plaintiff. Rather, the actual and potential *269 harm suffered by a plaintiff is a fact to be decided by the jury — in this case, 80 percent of $863,274. Plaintiff next argues that, for purposes of identifying the range of permissible ratios, defendant should not be credited with its 1998 payment (a total of $175,960.08) that purported to satisfy the $100,000 policy limits plus interest. Plaintiff contends that that payment did not reduce the potential harm that Munson suffered when the trial court entered judgment in the wrongful death action for $863,274. We disagree. Defendant’s responsibility to pay the $100,000 policy limits was contractual, and defendant eventually paid it. The Court of Appeals did not err in crediting defendant for the $100,000 payment. 9 Defendant renews the argument that it made on reconsideration in the Court of Appeals, i.e., that the prejudgment interest, which the Court of Appeals estimated at $589,000, may not be counted in establishing the punitive damages ratio. That argument turns entirely on the phrasing of the trial court’s judgment, which labeled prejudgment interest as separate from “economic damages.” Although defendant does not dispute that it must pay plaintiff that prejudgment interest, it argues that, “unless the jury itself awards prejudgment interest, it does not and cannot become ‘damages.’ ” We reject defendant’s argument. Under United States Supreme Court precedent, the punitive damages award should be proportional to the “actual or potential harm suffered by the plaintiff.” Campbell, 538 US at 418; see also BMW, 517 US at 575 (“harm or potential harm”). By permitting a punitive damages award to be a multiple of “potential harm,” the Court demonstrated that the punitive damages award is not limited to some multiple of the compensatory damages actually awarded by the trial court. And we have no problem concluding that the prejudgment interest here, however labeled by the trial court, is part of Munson’s “actual *270 harm.” The wrongful death judgment against Munson included accrued interest that Munson is legally obligated to pay. Defendant’s 80 percent liability for the wrongful death judgment necessarily makes it responsible for the interest that accrued on that portion of that judgment. We conclude, then, that the constitutionally permissible punitive damages award in this case will be some multiple of the correctly calculated actual harm to Munson— $690,619.20 in compensatory damages, plus interest (which the Court of Appeals estimated at $589,000). Comparing that figure (approximately $1.28 million) to the jury’s punitive damages award of $20,718,576, it immediately becomes clear that the ratio of punitive damages that the jury awarded to the “actual and potential harm suffered by the plaintiff’ is somewhere around sixteen-to-one. That ratio significantly exceeds the four-to-one ratio that the Court described, in Haslip, 499 US at 22-23, as being close to the constitutional line, at least when only economic injury is involved. Indeed, the ratio even exceeds the single-digit ratio that the Court mentioned in Campbell, 538 US at 425 (“few awards exceeding a single-digit ratio between punitive and compensatory damages, to a significant degree, will satisfy due process”), as the putative limit for punitive damages in any kind of case. It may be true that, as we already have observed, a punitive damages award that exceeds the single-digit ratio may be acceptable in a few narrow circumstances: (1) when a particularly egregious act causes only a small amount of economic injury; (2) when the injury is hard to detect; (3) when it is difficult to place a monetary value on noneconomic harms; and (4) when “extraordinarily reprehensible” conduct— roughly comparable to the defendant’s conduct in Williams— is involved. None of those circumstances is present here: The jury awarded plaintiff substantial compensatory damages; the injury that defendant caused was evident; no claim of noneconomic harm was involved; and defendant’s conduct is in no way comparable to, for example, Philip Morris’s 50-year campaign to delude a large part of the population of Oregon about the potentially devastating physical effects of smoking its products, which this court described in Williams. Also, *271 plaintiff has not pointed to any other factor that we believe takes the case out of the normal realm of punitive damages cases that involve economic injury alone. Although there is no question that this is a case in which punitive damages at some level properly may be awarded, the case is not extraordinary in any way that would justify a punitive damages award that is more than four times, much less more than nine times, the amount that the jury found was the “actual and potential harm suffered by the plaintiff.” We proceed to the third Gore factor — consideration of comparable criminal or civil sanctions. In Williams, this court concluded that analysis of the comparable sanctions guidepost proceeds in three steps: “First, courts must identify comparable civil or criminal sanctions. Second, courts must consider how serious the comparable sanctions are, relative to the universe of sanctions that the legislature authorizes to punish inappropriate conduct. Third, courts must then evaluate the punitive damage award in light of the relative severity of the comparable sanctions. The guidepost may militate against a significant punitive damage award if the state’s comparable sanctions are mild, trivial, or nonexistent. However, the guidepost will support a more significant punitive damage award when the state’s comparable sanctions are severe.” 340 Or at 58. In the context of the calculation method described here, severe comparable sanctions — if they exist— could justify a higher ratio between actual or potential harm and the amount of punitive damages. Plaintiff argues — and we have recognized — that defendant’s misconduct implicates civil sanctions. At the time of the accident in this case — and today — defendant’s misconduct violated several provisions of ORS 746.230. Each such violation was (and is) subject to a civil penalty of up to $10,000. ORS 731.988(1). 10 Furthermore, the violator may also be required to disgorge the profits that resulted from the *272 violation. See also ORS 731.988(2) (additional civil penalty “not to exceed the amount by which such person profited in any transaction which violates any such provision [of the Insurance Code]”). Those sanctions are equivalent to an amount that the legislature has authorized to be imposed only for felony crimes (all lesser offenses are subject to fines less than $10,000). See, e.g., ORS 161.635(l)(a) (1987) (maximum fine of $2,500 for Class A misdemeanor); ORS 161.655(l)(b) (1987) (maximum fine of $5,000 for certain Class A misdemeanors committed by a corporation). In other words, $10,000 is an amount that the legislature thought appropriate to levy only in the case of felonies. See ORS 161.625(1) (1987) (for all classes of felonies, fine not to exceed $100,000); ORS 161.655(l)(a) (1987) (for felony by corporation, fine not to exceed $50,000). Measured by those criteria, a relatively severe punitive damages award would not violate due process. Even so, we have little difficulty in concluding that the $20 million punitive damages award that the jury awarded in this case is “grossly excessive,” as that term is used by the Supreme Court. Although our analysis of the first and third Gore factors suggests that a relatively high punitive damages award would be appropriate, it is the second Gore factor — the ratio of the punitive damages award to the actual and potential harm to the plaintiff — that tells us, in more or less absolute numerical terms, how high the punitive damages award may be. Our consideration of that factor forecloses a conclusion that the jury’s $20 million dollar award is consistent with the federal Due Process Clause: The sixteen-to-one ratio between the punitive damages award and plaintiffs actual and potential damages far exceeds the nine-to-one ratio that (with certain specified exceptions) marks the limits of due process. We affirm the Court of Appeals’ holding to that effect. E. What is the Maximum Constitutionally Permissible Award in this Case? That leaves us to decide the remaining issue: What is the highest amount of punitive damages that may be awarded in this case, without violating defendant’s rights under the Due Process Clause? *273 The Court of Appeals, after following much of the same tortuous path that we have reviewed here, reached the following conclusions respecting the appropriate ratio between the actual or potential damages suffered by Munson and the punitive damages award: “Beyond [Campbell’s] admonition that a ratio exceeding single digits is constitutionally permissible only in extraordinary cases and Gore’s earlier caution that a ratio of 4:1 might approach constitutional limits, there is a certain circularity to [the guidance offered by the Supreme Court’s cases] * * *. Presumably, the constitutionality of any award of punitive damages within the presumptive single-digit ratio range will depend on the application of the first (reprehensibility) and third (comparable civil penalties) guideposts. Beyond that, the only ratio-specific variable appears to be [Campbell’s] instruction that, in some cases involving relatively small amounts of compensatory damages, proportionately greater awards of punitive damages may be appropriate to effectuate the general purposes of punitive damages, with the obverse being true in some cases involving relatively large awards of compensatory damages. [Campbell], 538 US at 426. “Ultimately, as a practical matter, the ‘ratio’ guidepost serves as a sort of benchmark. Although fact-matching can be a fool’s errand, cases involving roughly analogous circumstances, particularly with respect to the ‘reprehensibility’ variables, should yield roughly similar ratios of compensatory and punitive damage awards. In that regard, it is instructive, although by no means conclusive, that, in [Campbell], which also involved a bad faith refusal to settle a clear liability third-party auto death claim, the Supreme Court suggested that a 1:1 ratio on a $1,000,000 compensatory award would comport with due process. 538 US at 429. Still, we hasten to emphasize that defendant’s course of conduct in this case was considerably more protracted and, in our view, much more malicious than was the insurer’s conduct in [Campbell]. “We return to the application of the Gore/[Campbell] guideposts in this case. Any such application, albeit ‘as a matter of law,’ is necessarily imprecise — even, at least arguably, impressionistic. Nevertheless, given the guidance of [Campbell], the distinctions of this case from *274 [Campbell], and the analysis of our post-[Campbell] decisions, we conclude that the maximum constitutionally permissible award of punitive damages in this case is three times the compensatory damage award. “Our reasoning is at least implicit from our foregoing discussion: Although the injury here was purely economic, without any threat to public health or safety, defendant’s conduct was far more reprehensible — a calculated and repeated course of cynical and malicious betrayal of its insured’s trust — than the Court found to be the case in [Campbell]. Nor was this an ‘isolated incident.’ Although certain features of defendant’s conduct here may have been ‘isolated’ or unique, there can be no question, viewing the evidence most favorably to plaintiff, that the ‘stonewalling,’ the ‘low-balling,’ the pressured manipulation of claims evaluations and the like, are all typical of how defendant did business: Defendant’s interests came first. On balance, the egregiously unethical character of defendant’s conduct justifies a proportionately, greater award of punitive damages than the 1:1 ratio suggested by the Court in [Campbell]. Conversely, the lack of serious physical injury or disregard for the health and safety of the consuming public dictates a proportionally lower award of punitive damages than in Waddill [c. Anchor Hocking, Inc. 190 Or App 172, 78 P3d 570 (2003) (involving personal injuries resulting from sale of dangerous product)]. Accordingly, a 3:1 ratio of punitive damages to compensatory damages in this case comports with due process.” Goddard, 202 Or App at 119-22. As is apparent, the Court of Appeals reached its determination by employing the Gore guideposts. We already have performed much of that same analysis. What remains is to synthesize what seem to us to be the relevant conclusions. With respect to the reprehensibility guidepost, we, like the Court of Appeals, have concluded that, because defendant’s conduct only caused economic harm and did not reflect indifference to the physical health or safety of others, it would not justify an award at or approaching the Gore (nine-to-one) ratio limit that we have discussed. Still, we acknowledge that, under certain of the relevant aggravating *275 factors, defendant’s conduct was very reprehensible. Specifically, the jury could infer that the person injured by defendant’s conduct (Munson) was financially vulnerable and defendant knew it, that defendant’s tortious conduct was not an isolated incident but, instead, was a deliberate pattern and practice, and that defendant acted deceitfully and with malice. With respect to the third guidepost, which looks at civil sanctions for comparable conduct, we concluded that, in the universe of civil penalties, the $10,000 civil penalties that applied to conduct like that at issue here all were relatively severe. Moreover, defendant’s acts were repeated, and the civil penalties could have been imposed for each separate illegal act. Those facts suggest that due process would permit a relatively high punitive damages award. In that regard, too, our view agrees with that of the Court of Appeals. But, as we already have noted above, 344 Or at 258-59, it is only the second Gore guidepost that offers any hard numerical guidance. Campbell suggests that, except in extraordinary circumstances, a punitive damages award that is more than nine times the amount awarded in compensatory damages violates due process, no matter what the tort. 538 US at 425. Gore, in our view, added another line of demarcation — that, when the conduct has solely economic effects and does not cause or risk physical harm to another, the ratio between the punitive damages award and compensatory damages normally cannot exceed four to one. Because, in this case, there is no basis for concluding that defendant caused or created a risk of physical harm to Munson, that four-to-one ratio is the relevant constitutional limit on punitive damages awards. In the end, we are persuaded that those factors demonstrate that requiring a one-to-one ratio, like that commended in Campbell, would exceed our constitutional role. On the other hand, we believe that approving an award at the outside limits (four to one, under Gore) would be permissible. Several reasons influence our choice, but the primary one is the degree of defendant’s betrayal of its obligation to protect Munson. The drumbeat march of defendant’s disregard for Munson’s interests began with Munson’s initial claim and *276 persisted, despite the most obvious reasons that it should end, through the calamitous verdict against him. There may be possible courses of behavior in the area of economic wrong that would be even worse, but none leaps to mind. This case fully justifies the highest permissible award, viz., an award that is four times the amount of plaintiffs actual and potential harm (which, as we have discussed, is $690,619.20 plus prejudgment interest). 11 We therefore modify the Court of Appeals opinion to that effect. On remand, the trial court should finally calculate the amount of prejudgment interest — and, thus, precisely calculate the maximum permissible punitive damages award — and present that amount to plaintiff as the maximum punitive damages award that, on this record, the Due Process Clause will bear. Then, the court should grant defendant’s motion for a new trial, unless plaintiff agrees to entry of an amended judgment that reduces the punitive damages award to that amount. The decision of the Court of Appeals is affirmed as modified by this opinion. The judgment of the circuit court is vacated and remanded with instructions to grant defendant’s motion for new trial, limited to punitive damages, unless plaintiff agrees to remittitur of punitive damages to four times compensatory damages award. Footnotes [1] Defendant asks this court simply to adopt the Court of Appeals’ statement of the facts. Plaintiff agrees that the Court of Appeals opinion is accurate as far as it *236 goes, but she argues that the Court of Appeals does not go far enough in describing defendant’s reprehensible conduct: Much of her brief to this court is devoted to highlighting evidence that, according to plaintiff, "depict a [defendant] more treacherous than the malicious and deceitful [defendant] seen in the Court of Appeals opinion.” We do not feel compelled to include plaintiffs detailed additions in our explication of the facts. Her suggested additions confirm the malicious and deceitful nature of defendant’s conduct, but do not alter our overall assessment of the level of reprehensibility involved. [2] As noted earlier, a subsequent declaratory judgment action would vindicate defendant on this point, determining that the accident was covered under Foley’s policy but not Munson’s. Id. at 82 and n 3. [3] We concluded, in fact, that a $79.5 million punitive damages award — an award more than 29 times the jury’s compensatory damages award — did not violate due process when the defendant, Philip Morris Inc., had “engaged in a massive, continuous, near-half-century scheme to defraud the plaintiff and many others, even when Philip Morris always had reason to suspect — and for two or more decades absolutely knew — that the scheme was damaging the health of a very large group of Oregonians — the smoking public — and was killing a number of that group.” Williams, 340 Or at 63. In cases like Williams, where a party has engaged in a prolonged pattern of egregious and deceitful conduct that poses an extreme threat to the health and safety of a significant segment of the population of the state, the *259 state’s interest in punishing and/or deterring such conduct is heightened, and a punitive damages award that exceeds the single-digit ratio may be justified. [4] As previously noted, the Supreme Court in both Gore and Campbell declined to determine what level of punitive damages would be constitutionally permissible for the conduct at issue in those cases, and expressly left it to the states to make that determination “in the first instance.” We take that charge seriously. We cannot reasonably expect trial courts to review punitive damages awards for gross excessiveness without providing some concrete limits or, at least, general rules of *260 thumb, beyond the single-digit ratio that the Court mentions in passing. This is our first effort in that regard. *262 from the Restatement (Second) of Torts § 908 (1979) and from this court’s opinion in State ex rel Young v. Crookham, 290 Or 61, 72, 618 P2d 1268 (1980). [5] The trial court in this case instructed the jury that, to award punitive damages, it must find that defendant “was guilty of wanton misconduct that caused the judgment to be entered against * * * Munson.” The court also instructed the jury that, in fixing the amount of punitive damages, it could properly consider “(A) The likelihood that serious harm would arise from the defendant’s misconduct, “(B) The degree of the defendant’s awareness of that likelihood, “(C) The profitability or potential profitability of defendant’s misconduct, “(D) The defendant’s motive, “(E) The duration of the misconduct and any concealment of it, “(F) The attitude and character of the defendant’s conduct upon discovery of the misconduct, “(G) The number and position of employees involved in causing or covering up the misconduct, “(H) The sum of money that would be required to discourage the defendant and others similarly situated from engaging in such conduct in the future, “(I) The income, assets and financial condition of the defendant.” No one argues that those instructions are not a correct statement of the law relating to punitive damages in ordinary tort actions. They appear to be drawn [6] See note 5, describing the standard that the jury was instructed to apply in making its punitive damages decision. [7] When introducing the three guideposts in Gore, the Supreme Court began by noting that “[e]lementary notions of fairness enshrined in our constitutional jurisprudence dictate that a person receive fair notice not only of the conduct that will subject him to punishment, but also of the severity of the penalty that a state may impose.” 517 US at 574. [8] In Georgetown Realty, this court held that, when a liability insurer undertakes to defend its insured against a claim by a third party, it may be hable in tort *264 to the insured if, without justification, it fails to settle a claim against its insured within policy limits. The Georgetown Realty court did not announce that holding as a change in the common law. Instead, the court appeared to believe that the holding was consistent with a long line of cases that distinguish between (1) claims that are based solely on a breach of provision in a contract, which itself spells out the party’s obligations, and (2) claims that one party to a contract caused damage to the other by negligently or tortiously performing its obligations under the contract. 313 Or at 102-11. [9] In determining the amount of compensatory damages, the Court of Appeals concluded that all of the payment should be applied to interest, not to principal. Goddard, 202 Or App at 107-10. The parties do not challenge that ruling on review; the only issue here is whether that amount counts for purposes of calculating the constitutional limit on punitive damages. [10] ORS 731.988(1) provides, in part: “Any person who violates any provision of the Insurance Code * * * shall forfeit and pay to the General Fund of the State Treasury a civil penalty in an amount determined by the director of not more than $10,000 for each offense * * *. Each violation shall be deemed a separate offense.” [11] Some uncertainty exists about the precise amount of Munson’s actual harm in this case, because the prejudgment interest has not yet been specifically calculated. As noted, the Court of Appeals merely estimated prejudgment interest at $589,000, leaving it to the trial court on remand to precisely calculate the amount. Goddard, 202 Or App at 110. For purposes of our review, we also leave such calculations to the trial court on remand.