with whom CARPENETI, Justice, joins, dissenting.
I disagree with the opinion’s conclusions concerning the admissibility of evidence of Boyles’s insurance coverage and the propriety of denying FleegePs alternative motion to exclude for cause all prospective jurors who were State Farm policyholders. I would reverse on those points.
EVIDENCE OF BOYLES’S INSURANCE COVERAGE
The opinion concludes that the superior court correctly denied Fleegel’s motion to exclude evidence of Boyles’s insurance coverage, reasoning that Alaska Evidence Rule 411 did not preclude this evidence because evidence of a defendant’s wealth has relevance independent of fault in a punitive damages case: “The trial court focused on the relevance of insurance to Boyles’s financial condition, a purpose not excluded under Rule
*1279411. Therefore, the trial court did not err in failing to exclude evidence of insurance coverage.” 1 But the opinion’s analysis overlooks persuasive authority to the contrary and skips two important questions in the chain of admissibility: whether insurance coverage actually is relevant evidence of wealth and, if it is, whether its prejudicial effect outweighs its probative value.
While the opinion correctly observes that evidence of the defendant’s wealth is normally relevant in a punitive damages case,2 it neglects to ask if evidence of insurance coverage for punitive damages is relevant evidence of wealth under the specific facts of this case. Here, assuming that Boyles had shown that he was covered for punitive damages,3 evidence of his coverage would have been irrelevant to prove his wealth. For as the New Mexico Supreme Court recognized, “[Pjunitive damages liability coverage is not an asset which can be used to measure true punishment and ... therefore, it should not be considered by the jury in assessing a defendant’s financial standing.”4 Wisconsin has recently ruled the same way: “Although we note that when assessing punitive damages a jury is permitted to know evidence of the wrongdoer’s wealth, insurance coverage is not evidence of wealth.”5
And even if evidence of coverage were deemed relevant to the issue of wealth in a punitive damages case, Evidence Rule 403 would preclude its admission if the trial court found that the probative force of that evidence was outweighed by “the danger of unfair prejudice, confusion of the issues, or misleading the jury.”6 Here, the trial court never undertook the balancing process prescribed by Rule 403. But it seems to me that the potential prejudicial effect of the disputed evidence far outweighed whatever limited probative value it might have had. As evidenced by a juror’s all-too-predictable request to know whether “punitive damages [are] directfed] to Mr. Boyles or to the insurance company,” as well as by the jury’s anomalous punitive damages verdict (which found Boyles deserving of punishment but imposed no punitive damages), the trial court’s unfortunate decision to allow the jury to learn of Boyles’s insurance coverage opened the door to precisely the kind of danger that Rule 403 was designed to prevent: causing confusion and misleading the jury-
New Mexico and Wisconsin have both suggested that evidence of coverage would have to be excluded as more prejudicial than probative, even if it were relevant.7 Texas agrees.8 And most recently, Idaho has reached the same conclusion.9 Indeed — barring a narrow exception carved out for a situation not presented here — it appears that no other court addressing this issue has *1280found such evidence admissible.10 The narrow exception is a curative one: it allows plaintiffs to admit evidence of coverage to rebut a defense that a defendant is impoverished and therefore unable to pay punitive damages.11
Shane v. Rhines, which today’s opinion holds out as its guiding authority,12 is just such a case: it considered whether evidence of coverage was admissible to rebut the defendant’s affirmative claim of poverty.13 Neither Shane’s per curiam plurality nor either of its separate opinions purports to hold that evidence of coverage should be admitted except as actually needed to cure a misleading impression left by affirmative evidence of poverty. And notably, in Schaefer v. Ready, a more recent punitive damages case in which admission of evidence proving a defendant’s coverage was sought to cure affirmative evidence of poverty, the Idaho Supreme Court held that the better approach would be to exclude with one stroke both the affirmative defense and the consequent need for rebuttal.14
Shane and Schaefer, then, both serve to illustrate that necessity plays an indispensable role in Evidence Rule 403’s balancing process. Evidence that is arguably relevant but has an obvious potential to cause prejudice should be admitted under the rule only when a realistic, case-by-case assessment reveals an apparent need for its admission that predominates over its potential to confuse, mislead, or unfairly prejudice the jury.
In the present case, there was no conceivably legitimate need for evidence of Boyles’s insurance coverage. For Boyles’s counsel himself injected the fact of coverage; and he did so not to refute evidence that Boyles was unable to afford punitive damages. To the contrary, defense counsel relied on this information to forward a novel theory: that because Boyles — the only truly guilty party— could not pay punitive damages, his innocent insurer should not be targeted for punishment in his place.
Yet this theory of relevance — the only theory that the disputed evidence of coverage realistically tended to support — is disingenuous and fundamentally subversive to the public interest. Insurance is a regulated industry in Alaska, as it is almost everywhere else in the nation, because the state recognizes a strong public interest in fostering trust in insurers and in protecting consumers who choose to buy coverage.15 Once insurers contract to pay claims for punitive damages, then, the state has a strong public interest in ensuring that coverage is fairly provided and claims fairly paid. It would be inimical to this public interest if insurers could avoid legal responsibility for contractual obligation by the simple expedient of insisting that they are innocent parties and that their insured drivers should be punished only to the extent of their personal ability to pay.
Correspondingly, there is simply no factual basis for arguing that the insurer’s liability to pay punitive damages amounts to punish*1281ment for the insurer: to an insurer that accepts premiums to cover punitive damages, the obligation to pay an insured driver’s punishment is simply a business debt that the company has contracted to pay. In my view, then, there was no legally permissible reason to allow Boyles’s counsel to inform the jury that Boyles was insured.
Nor can the record sustain a finding of harmless error. Appellate courts have long recognized that evidence of insurance coverage can easily inflame juries; courts have likewise recognized that the risk of prejudice inherent in such evidence poses especially grave dangers in punitive damages cases,16 where the prejudice can cut either way, acting as “a two-edged sword depending on the wealth or poverty of the defendant.”17 And courts have further recognized that this sword of unfair prejudice is no less dangerous when wielded by an insured defendant than when in the hands of an injured plaintiff.18
Here, Boyles’s attorney artfully wielded the sword against Fleegel. After convincing the trial court at the outset of the proceedings to deny Fleegel’s motion to exclude evidence of coverage, Boyles’s counsel immediately told the jury during voir dire that he had been hired by Boyles’s insurer. Fleegel did everything she could to contain the inevitable damage flowing from this revelation: she downplayed the issue, allowed Boyles’s attorney no obvious opening to dwell on Boyles’s coverage during the trial, and stipulated to an instruction at the end of trial that mirrored the court’s pretrial ruling and told the jury no more than defense counsel took pains to tell it at the outset-that Boyles was insured and was represented by a lawyer hired by his insurer.
Despite Fleegel’s best efforts, defense counsel used Boyles’s coverage as the foundation for a successful empty-chair defense: a tactic that emphasized Boyles’s absence and poverty and that invited the jury to conclude that his insurer had been unjustly left holding the bag-abandoned by the only truly guilty party. Defense counsel’s first words in closing argument reminded the jury that Boyles’s insurer was the real party in interest: “It’s pretty silly to argue that an insurance company likes a hit-and-run driver or a drunk driver.” Defense counsel proceeded to emphasize that Boyles would not be deterred by an award of punitive damages, since “[h]e doesn’t have any assets.” And the point was cemented in defense counsel’s final words to the jury: “[I]t’s quite clear that they’re not really trying to get money back from Mr. Boyles. He doesn’t have any capacity to pay.... [Hje’s not the target here.”
Yet in the eyes of the law Boyles was Fleegel’s only target — whether Boyles had insurance or not. Evidence of his coverage — or possible coverage — was both immaterial as a matter of law and irrelevant as a matter of fact. But as illustrated by the jury’s question concerning punitive damages and its, ensuing verdict, Boyles’s strategy nonetheless had its intended effect.
Today’s opinion nonetheless declines to find error, professing an inability to locate any cases holding that “introduction of insurance evidence is reversible error because it is prejudicial to a plaintiff.”19 Yet this absence of precedent is beside the point.20 A novel tactic whose sole purpose is to prejudice an opponent is hardly proper merely because it has never before been condemned; and the *1282fact that the tactic prejudices a plaintiff rather than a defendant is wholly irrelevant.
More to the point is the opinion’s refusal to follow well-settled law that allows evidence of coverage to be admitted only when offered by a party seeking to refute an opponent’s affirmative claim of inability to pay punitive damages. And more telling still is the opinion’s failure to identify any legitimate purpose that evidence of coverage could conceivably serve when offered on behalf of a party like Boyles, who actively asserted that he was too poor to pay punitive damages.
On this record, then, it is simply unrealistic to find either an absence of error or harmless error.
DISQUALIFICATION OF STATE FARM POLICYHOLDERS
I also disagree with the opinion’s decision to affirm the trial court’s denial of Fleegel’s alternative motion to excuse State Farm policyholders from serving on her jury.
As a contingent remedy to be used in the event that the court denied her motion to preclude evidence of Boyles’s insurance coverage, Fleegel asked the superior court for voir dire examination to identify and excuse for cause all members of the jury panel who were insured by State Farm. The court denied the motion, instead precluding any mention of State Farm by name and barring any inquiry as to the identity of prospective jurors’ insurers. In upholding this ruling, today’s opinion posits that “[w]hether a juror is dismissed for cause is within the discretion of the trial court.”21 The opinion then concludes that the superior court did not abuse its discretion.22
But the opinion is mistaken in assuming that Fleegel’s challenge for cause raised a discretionary issue. Alaska Civil Rule 47(c)(10) and (12) categorically require a challenge for cause to be granted as to any prospective juror who is a client of a party or who has a financial interest in the outcome of the case.23 Our decisions applying these provisions explicitly recognize that they prohibit “certain relationships between jurors and parties” and that “[a] trial judge does not have discretion to deny a challenge for cause once that relationship has been established.” 24
In my view, when a court presiding over a punitive damages claim allows an insured defendant to tell the jury that the defendant’s insurance company is the real party in interest — the plaintiffs real “target” — this information inevitably creates a genuine risk that jurors who are policyholders of the same insurer, or who think that they might be, will perceive a financial interest in the outcome. The financial interest here is certainly no more attenuated than the one at issue in Reich v. Cominco Alaska, Inc., where we held that Rule 47(c)(12) required automatic disqualification of all prospective jurors who were shareholders in an Alaska Native corporation that was not a party to the litigation but had financial ties to the named defendant and stood to benefit indirectly from the outcome of the case.25 And this interest is at least as substantial as the one in Malvo v. J.C. Penney Co., where we held that Rule 47(c)(10) categorically required exclusion of all prospective jurors who had accounts with J.C. Penney and therefore technically qualified as the company’s “debtors.”26
*1283In reaching the opposite conclusion here, today’s opinion posits that “[t]he financial interest that State Farm insured jurors have in this case is speculative at best.”27 But the opinion’s analysis of the closeness of the relationship between policyholders and their insurers is misdirected, for Rule 47(c)(10) expressly preempts the issue. As already mentioned, Rule 47(c)(10) categorically excuses all prospective jurors who are “clients” of a party or an attorney.28 Since policyholders are undeniably clients of their insurers, Rule 47(c) unequivocally deems the financial relationship between insurance companies and their policyholders sufficiently close to require automatic disqualification in any action where a policyholder’s insurer is a party or an attorney.
Suppose for example that State Farm was actually named as a defendant in a lawsuit: there would be no doubt that its policyholders would have a sufficiently close financial interest to require their automatic disqualification as “clients” under Rule 47(c)(10). Because the financial relationship between State Farm and its policyholders is no different here than it would be if State Farm were a named party, it follows that the policyholder/insurer relationship itself is sufficiently close to warrant disqualification, provided that the other requirements of Rule 47 are met. The critical question, then, is whether the role State Farm played in this case justifies treating it as a party.
In the run-of-the-mill personal injury case against an insured party, of course, an insurer’s role might be insufficient to warrant treating it as a party under Rule 47. But when a policyholder acting through counsel retained by an insurer expressly reveals the existence of coverage and portrays the insurer as the plaintiffs true “target,” it seems to me that, in terms of triggering concerns for potential jury bias, the insurer’s role becomes virtually indistinguishable from the role it would play as an actual party. Hence, in this situation I would conclude that the insurer must be treated as a party under Rule 47(c)(10)’s provisions governing automatic disqualification of clients.
In the proceedings below, the superior court’s order restricting jury voir dire precluded Fleegel from establishing the very relationship that would have triggered Rule 47(c)’s mandatory right to a challenge for cause. True, the order also prevented the jury panel from learning the name of Boyles’s insurer. But this precaution hardly cured the potential for prejudice: by allowing Boyles to tell the jury that he was insured by an unnamed insurer, the trial court effectively left all insured jurors — not just those insured by State Farm — to speculate about what impact a punitive damages verdict might have on their own insurance premiums.
Thus, in my view, concealing State Farm’s identity was no more acceptable here than shielding J.C. Penney’s identity would have been in Malvo v. J.C. Penney Co. or shielding the Native corporation’s identity would have been in Reich v. Cominco Alaska, Inc. As shown by the jury’s question and its ensuing verdict, the superior court’s approach simply spread the danger of prejudice; it failed to protect Fleegel from potential jury bias as required under Rule 47(c).
Accordingly, given the superior court’s decision allowing defense counsel to tell the jury of Boyles’s insurance coverage, I would hold that Rule 47(c)(10) required the court to grant challenges for cause as to all prospective jurors who were State Farm policyholders.
CONCLUSION
I therefore dissent.
. Op. at 1272.
. See id. at 1271.
. Preliminarily, it is worth noting that evidence of coverage would be categorically irrelevant to the issue of a defendant's ability to pay punitive damages absent a specific showing that the defendant was actually covered for punitive damages. Cf. Owens-Coming Fiberglas Corp. v. Malone, 972 S.W.2d 35, 40-41 (Tex.1998) (distinguishing between evidence of actual and potential wealth in punitive damages case). Here, Boyles's counsel equivocated on this issue when asked whether Boyles was covered for punitive damages. Although hired by State Farm to defend the case, counsel insisted that he could not speak for the company and said only "[m]y position is that there is [coverage].”
. Baker v. Armstrong, 106 N.M. 395, 744 P.2d 170, 173 (1987).
. City of West Allis v. Wis. Elec. Power Co., 248 Wis.2d 10, 635 N.W.2d 873, 889 (2001) (citations omitted), review denied, 643 N.W.2d 93 (Wis. 2002).
. Alaska R. Evid. 403.
. See Baker, 744 P.2d at 173; City of West Allis, 635 N.W.2d at 888-89.
. See Owens-Coming Fiberglas Corp., 972 S.W.2d at 41.
. Schaefer v. Ready, 134 Idaho 378, 3 P.3d 56, 59 (Idaho App.2000):
ITJhe [trial] court ruled that evidence of the Readys’ insurance coverage for punitive damages awards was not admissible because "the probative value was substantially outweighed by its potential prejudicial impact.” Specifically, the court felt that the "mention of insurance invites higher awards than are justified by the facts.” This ruling is in accord with [Idaho Rule of Evidence] 41![.]
. The court's opinion tries to stand the point on its head by observing that “no court has held that introduction of insurance evidence is reversible error because it is prejudicial to a plaintiff[.]” Op. at 1272. Yet the absence of authority simply reflects the reality that few trial courts have ever thought to admit evidence of coverage for purposes comparable to those at issue here: to prove that a defendant deserving of punishment should not be punished because the defendant’s innocent insurer should not have to pay-even though the insurer presumably has contracted and been paid to do just that. The salient point, then, is that case law uniformly rejects the admission of such evidence; no authority supports its admission on the theory that prejudicing a plaintiff’s case is somehow more acceptable than prejudicing a defendant’s case.
. See, e.g., Wheeler v. Murphy, 192 W.Va. 325, 452 S.E.2d 416, 424 (1994).
. See Op. at 1271-1272.
. 672 P.2d 895, 899 (Alaska 1983).
. See 3 P.3d at 59. Cases in jurisdictions cited by Schaefer as applying the same rule indicate that those jurisdictions routinely exclude evidence of coverage in punitive damages cases. See id.
. See, e.g., AS 21.03.010 (requiring "[a]ll persons transacting a business of insurance in this state” to comply with the Alaska Insurance Code); AS 21.06.010 (authorizing appointment of statewide director of insurance); AS 21.06.020 (establishing division of insurance); and AS 21.06.080 (empowering director of insurance to enforce Alaska Insurance Code).
. See, e.g., City of West Allis v. Wis. Elec. Power Co., 248 Wis.2d 10, 635 N.W.2d 873, 889 (2001).
. S. Life & Health Ins. Co. v. Whitman, 358 So.2d 1025, 1027 (Ala. 1978) (Jones, J,, concurring).
. See Schaefer, 3 P.3d at 59 (quoting Kemezy v. Peters, 79 F.3d 33, 37 (7th Cir.1996)):
[I]t is bad enough that insurance or indemnification reduces the financial incentive to avoid wrongdoing.... It would be worse if the cost of insurance fell, reducing the financial disincentive to engage in wrongful behavior, because the insurance company knew that its insured could plead poverty to the jury. (Quotation marks omitted.)
. Op. at 1272.
. The observation is also incomplete in that it simply ignores the Idaho Supreme Court’s carefully reasoned opinion in Schaefer v. Ready, which would categorically exclude evidence of insurance coverage even if offered to refute a claim of inability to pay. See 3 P.3d at 59.
. Op. at 1272.
. See id. at 1272-1273.
. Alaska Civil Rule 47(c) states:
Challenges for Cause. After the examination of prospective jurors is completed and before any juror is sworn, the parties may challenge any juror for cause. A juror challenged for cause may be directed to answer every question pertinent to the inquiry. Every challenge for cause shall be determined by the court. The following are grounds for challenge for cause:
[[Image here]]
(10) That the person is the guardian, ward, landlord, tenant, employer, employee, partner,
client, principal, agent, debtor, creditor, or member of the family of a party or attorney. ...
[[Image here]]
(12) That the person has a financial interest, other than that of a taxpayer or a permanent fund dividend recipient in the outcome of the case.
. Reich v. Cominco Alaska, Inc., 56 P.3d 18, at 22 (Alaska 2002); see also Malvo v. J.C. Penney Co., 512 P.2d 575, 579 (Alaska 1973).
. 56 P.3d at 18-20, 27.
. 512 P.2d at 579.
. Op. at 1273-1274.
. See supra note 23 and accompanying text.