Yung v. Grant Thornton, LLP
Annotate this Case
Justia Opinion Summary
Want to stay in the know about new opinions from the Kentucky Supreme Court?
Sign up for free summaries delivered directly to your inbox. Learn More ›
You already receive new opinion summaries from Kentucky Supreme Court. Did you know we offer summary newsletters for even more practice areas and jurisdictions? Explore them here.
In this fraud and professional negligence case the Supreme Court reversed the decision of the Court of Appeals that a $80 million punitive damage award was unreasonable and reinstated the trial court’s award, otherwise affirming the appellate court’s judgment, holding that the $80 million award was not grossly excessive and was constitutionally acceptable.
Plaintiffs participated in a tax shelter marketed by Defendant, their accounting firm. After the IRS disallowed the tax shelter, Plaintiffs settled with the IRS, paying a total of $20 million for back taxes, interest and penalties and amounts paid to Defendant for fees. Plaintiffs then commenced this action to recoup the $20 million. The trial court found Defendant liable for fraud and gross professional negligence and awarded $20 million in compensatory damages and $80 million in punitive damages. The Court of Appeals affirmed the judgment on liability and compensatory damages but reduced the punitive damage award to equal the compensatory damage award. The Supreme Court (1) affirmed the Court of Appeals’ decision that Defendant was liable for fraudulent conduct and the compensatory damage award; but (2) reversed the remittitur and reinstated the trial court’s punitive damage award, holding that the facts supported an $80 million punitive damage assessment and that an award of that magnitude was constitutionally acceptable.
Download PDF
RENDERED: DECEMBER 13, 2018 TO BE PUBLISHED 2016- SC-000571-DG AND 2017- SC-000151-DG WILLIAM J. YUNG, MARTHA A. YUNG, AND THE 1994 WILLIAM J. YUNG FAMILY TRUST V. APPELLANTS / CROSS-APPELLEES ON REVIEW FROM COURT OF APPEALS CASE NO. 2014-CA-001957-MR KENTON CIRCUIT COURT NO. 07-CI-02647 GRANT THORNTON, LLP APPELLEE ! CROSS-APPELLANT OPINION OF THE COURT BY JUSTICE HUGHES AFFIRMING IN PART AND REVERSING IN PART William J. Yung, Martha A. Yung, and the 1994 William J. Yung Family Trust (collectively, the Yungs) participated in a tax shelter marketed by their accounting firm. Grant Thornton LLP (GT or the Firm). The shelter, Lev301, purportedly would allow funds held in the Yungs’ Cayman Island-based companies to be distributed to shareholders in the United States without federal tax liability. After the Internal Revenue Service (I.R.S.) disallowed the tax shelter, the Yungs settled with the I.R.S. in early 2007. Later that year, the Yungs commenced this action to recoup approximately $20 million, the combined total paid to the I.R.S. in back taxes, interest and penalties and paid to GT for fees. Following a bench trial, the trial court found GT liable for fraud and gross professional negligence in the marketing and sale of the tax shelter and awarded approximately $20 million1 in compensatory damages and $80 million in punitive damages. The Court of Appeals affirmed the circuit court’s judgment in favor of the Yungs on liability and compensatory damages. Partially reversing, that court reduced the punitive damage award to equal the compensatory damage award, having concluded that a punitive damage award in excess of the approximately $20 million compensatory damage award (a 1:1 ratio) was manifestly unreasonable and exceeded the amount justified to punish GT and to deter like behavior.2 The Yungs moved for discretionary review seeking to reinstate the $80 million punitive damage award and GT requested discretionary review regarding its liability and the compensatory and punitive damages. Having granted both motions, we affirm the Court of Appeals’ decision that GT is liable for its fraudulent conduct and approximately $20 million in compensatory damages. We reverse, however, the appellate court’s decision that the $80 1 The trial court awarded $4,682,786 in compensatory damages to William and Martha Yung, inclusive of GT’s $900,000 engagement fee, and $14,632,441 in compensatory damages to the 1994 Yung Family Trust, as well as prejudgment interest on the refunded $900,000 engagement fee from June 11, 2007 through the date of the judgment. This Opinion will refer to the award as approximately $20 million. 2 The Court of Appeals also remanded this case for entry of a corrected judgment setting the prejudgment interest rate at 8% per annum; the Yungs do not contest this correction. GT does not advance to this Court all the issues raised before the Court of Appeals, such as the trial court’s failure to enforce the engagement letter’s limitationof-liability clause; reduction of the compensatory damage award pursuant to Kentucky Rule of Civil Procedure (CR) 8; and modification of the post-judgment interest award. million punitive damage award is unreasonable and reinstate the trial court’s award. RELEVANT FACTUAL3 AND PROCEDURAL BACKGROUND 1. THE PARTIES AND LevSOl William J. Yung (Yung) is an experienced businessman who owns hotels and casinos in the Cayman Islands and in the United States, Columbia Sussex Corporation (CSC), owned by Yung and the 1994 William J. Yung Family Trust (Family Trust), is a privately held hospitality company headquartered in Crestview Hills, Kentucky. CSC is the primary organization for the Yungs’ hotel businesses, and at the time of trial owned approximately 40 hotels in the U.S. Yung also owns hotels and casinos through two Cayman Island holding corporations, Wytec, Ltd. (Wytec) and Casuarina Cayman Holdings, Ltd. (Casuarina). Casuarina is owned by William J. Yung and the Family Trust. Wytec is owned by Yung and two Grantor Retained Annuity Trusts (GRATs) created in 1997, one for the benefit of Yung and one for Martha A. Yung. The Cayman corporations are not obligated to make distributions to their shareholders, and consequently, profits accumulate in the Caymans without 3 In a bench trial, the trial judge weighs the evidence and makes findings of fact. Here, the trial court’s findings of fact following a 22-day trial comprise 168 pages of a 210-page order. We will not disturb the trial court’s factual findings on appeal unless they are clearly erroneous. Caudill v. Acton, 175 S.W.Sd 617 (Ky. 2004). Where quotation marks are employed in this Opinion, they refer to the trial court’s findings unless otherwise indicated. federal tax consequences. In 2000, the Yungs, in conjunction with these Cayman corporations, purchased the Grant Thornton Leveraged Distribution Product (LevSOl), which is the tax shelter at the center of this litigation. Grant Thornton LLP (GT) is a public accounting firm headquartered in Chicago, Illinois, with revenues in excess of $1 billion from 2000 through 2003. GT provided tax advice to the Yungs and their business organizations from the mid-1990s through some time in 2007, and the parties developed what the trial court found to be “a comfortable and trusting business relationship.” GT created the Lev301 as a strategy designed to allow corporations to make certain types of monetary distributions with a minimum of tax consequences to their shareholders. GT marketed the Lev301 to the Yungs and other clients beginning in 2000.4 As to the Yungs, the Lev301 involved moving money from the Cayman Islands into the U.S. by distributing the Cayman corporations’ profits to the shareholders as fully encumbered securities. First, the Cayman corporations bought $30 million in Treasury notes (T-notes) using borrowed money, with the T-notes serving as security for that debt. Next, the corporations transferred the T-notes to the shareholders in the U.S. Because they were 100% encumbered, the T-notes ostensibly had no taxable value, and accordingly, the shareholders would not report the distributions on their federal tax returns. The Cayman corporations would then pay off the debt six months to a year later, but the 4 GT used the Yungs’ experience to market the Lev301 to thirteen other customers, and sold a similar product to twenty-two other customers. loan repayment would also not result in reportable income to the shareholders because they were not co-obligors for the loan’s repayment. This tax shelter strategy, Lev301, theoretically allowed the shareholders to avoid tax consequences on $30 million in profits brought into the U.S. by means of the eventually unencumbered T-notes. Prior to the events at issue in this litigation, the Yungs brought income into the U.S. from the Cayman corporations when it could be done in “a tax efficient manner.” The Yungs looked for ways to accelerate this process, but, with a concern for the risks involved, vetted possible means closely; consequently, they did not engage in various tax strategies presented to them by other accounting firms. Because the Yungs were in the casino business, a highly-regulated industry, they were particularly sensitive to tax issues.5 The Yungs maintained a very conservative risk level as to income tax reporting. II. TAX SHELTER SCRUTINY A. The BOSS Notice * At the time GT began to develop Lev301, the U.S. Treasury Department (Treasury) was cracking down on products perceived as abusive tax shelters. In December 1999, the I.R.S. issued Notice 99-59 (BOSS Notice). 1999-52 I.R.B. 761. The BOSS (Bond and Option Sales Strategy) Notice described a tax product designed to create an artificial tax loss that was being sold at that time 5 The Cayman corporations owned casinos and Yung, through a separate corporation, owned casinos in Mississippi, Nevada and Louisiana. by several accounting firms.6 A BOSS transaction (described in the footnote) allegedly did not create taxable income under Internal Revenue Code (l.R.C.) § 301. Distribution of encumbered securities by a foreign corporation to a partnership where the securities were distributed “subject to” the bank debt, meant the value of the securities was reduced by the amount of the bank debt. With the bank debt equal in value to the securities, the value of the securities under l.R.C. § 301 was zero for tax purposes. The l.R.S. Notice warned that the BOSS transaction tax loss was not allowable for federal income tax purposes and that the l.R.S. could impose various penalties, including the accuracy-related penalty. B. New Tax Shelter Disclosures In early 2000, the Treasury issued additional regulations targeting the promotion of, and participation in, abusive and potentially abusive tax shelters. The Notice describes a typical BOSS arrangement: 6 Taxpayers, acting through a partnership, contribute cash to a foreign corporation (FC) set up solely to effectuate the transaction. In exchange for its cash contribution, the partnership receives the common stock of the FC. Another party contributes additional capital to the FC in exchange for the preferred stock of the FC. The FC borrows money from a bank and grants the bank a security interest in securities acquired by FC. The securities have a value equal to or greater than the amount of bank debt. FC then distributes the encumbered securities to the partnership, which has the effect of reducing the value of the FC’s common stock (held by the partnership) to zero or a minimal amount. The FC has sufficient other assets to pay off the debt with the other assets. The partnership is then treated as having disposed of the common stock, creating a tax loss equal to the “excess of the partnerships” original basis in the stock . . . over the fair market value of the common stock after the distribution of securities (zero). FC, typically in a later tax year, pays off the bank debt from its other assets, leaving the securities unencumbered in the hands of the taxpayers. T. D. 8875, 2000-11 I.R.B. 761; T.D. 8876, 2000-11 I.R.B. 753; T.D. 8877, 2000-11 I.R.B. 747. As a result, organizers and promoters of tax shelters were required to maintain a list of investors in potentially abusive tax shelters'7 and to make the list available for inspection upon the Treasury’s request. Organizers and promoters of corporate tax shelters were also required to register with the I.R.S. tax shelters which met the requirement of being “listed transactions,” a term the I.R.S. used to identify transactions that were the same or substantially similar to BOSS transactions. Corporate taxpayers filing U. S. income tax returns were also obligated to disclose participation in “reportable transactions.”8 In August 2000, the I.R.S. issued the Son of BOSS Notice (Notice 200044). 2000-36 I.R.B. 255. Addressing a BOSS derivative, the Notice declared that arrangements which purport to give taxpayers an artificially high basis in partnership interests and thereby give rise to deductible losses on disposition of those interests will not be recognized as bona fide losses reflecting actual economic consequences. Therefore, the losses are not allowable as deductions and may be disallowed under other I.R.C. provisions. The Notice also provided 7 “ A transaction for which a significant purpose of the structure of the transaction is avoidance or evasion of Federal income tax.” T.D. 8875, 2000-11 I.R.B. 761, 762; Treas. Reg. § 301.6111-2T (2000). 8 “Any transaction that is the same as or substantially similar to any of the specified types of tax avoidance transactions that the IRS has identified by published guidance as a listed transaction for purposes of § 6011 and that is expected to reduce the taxpayer’s Federal income tax liability by more than $1 million in any single taxable year or by a total of more than $2 million for any combination of taxable years.” T.D. 8877, 2000-11 I.R.B. 747, 748; Treas. Reg. § 1.6011-4T (2000). 7 that such arrangements are listed transactions and are subject to tax shelter registration and list maintenance requirements. During this timeframe, GT was aware of an April 2000 “Tax Notes” article entitled Corporate Tax Shelters: More Plain Brown Wrappers, written by noted tax law commentator Lee Sheppard. Sheppard described a so-called “Bossy” product being marketed by Arthur Andersen LLP and anticipated the l.R.S.’s disallowance of the product. The Bossy transaction involved the following steps: 1) A corporation borrows to buy a T-note that has a term of three to five years. 2) The corporation would then distribute the T-note, subject to bank debt, to its shareholders who would hold the note and collect the principal at maturity. 3) After the distribution, but before the note matures, the corporation would pay off the debt. Obviously, the Bossy product had the same essential characteristics as Lev301. C. The “More Likely Than Not” Tax Opinion Customarily tax shelter participants obtain a “more likely than not” tax opinion to satisfy the standard set forth in the Internal Revenue Code to avoid penalties for a substantial understatement of income tax. The U.S. Court of Federal Claims, in Alpha I, L.P. v. United States, 93 Fed. Cl. 280 (2010), effectively explains the role the tax opinion serves for tax shelter items. If an item is determined to be a tax shelter item, it may nevertheless be eligible for a reduction [in the amount subject to the understatement penalty] if there is substantial authority for the tax treatment of that item and the taxpayer reasonably believed at the time the return was filed that the tax treatment of the item was more likely than not the proper treatment. [Treas. Reg.] § 1.6662-4(g)(1). A taxpayer is considered reasonably to believe that the tax treatment of an item is more likely than not the proper tax 8 treatment if “[t]he taxpayer analyzes the pertinent facts . . . and in reliance upon that analysis, reasonably concludes in good faith that there is a greater than 50-percent likelihood that the tax treatment will be upheld if challenged by the Internal Revenue Service.” Jd. § 1.6662-4(g)(4)(i)(A). [Alternatively,] a taxpayer is considered reasonably to believe that the tax treatment of an item is more likely than not the proper tax treatment if . . . [t]he taxpayer reasonably relies in good faith on the opinion of a professional tax advisor, if the opinion is based on the tax advisor’s analysis of pertinent facts and authorities . . . and unambiguously states that the tax advisor concludes that there is a greater than 50-percent likelihood that the tax treatment of the item will be upheld if challenged by the Internal Revenue Service. Id. § 1.6662-4(g)(4)(i). 93 Fed. Cl. at 304. As the Alpha I, L.P., court noted [i]t is well established since [United States v. Boyle, 469 U.S. 241 (1985)] that reliance on the advice of a competent and independent professional adviser is a common means of demonstrating reasonable cause and good faith. Boyle, 469 U.S. at 251, 105 S.Ct. 687; see Treas. Reg. § 1.6664-4{b)(l); American Boat [Co., LLC v. United States, 583 F.3d 471,481 (7th Cir. 2009)]; Stable Creek [Investments, LLC v. United States, 82 Fed. Cl. 636, 717-18 (2008), affd, 608 F.3d 1366 (Fed. Cir. 2010)]. Id. at 315. III. LevSOl: GT’S LEVERAGED DISTRIBUTION PRODUCT In the spring of 2000, GT began developing Lev301, which it introduced to local offices of the firm in June 2000. Structured to avoid tax liability on shareholder distributions, the steps of the Lev301 were substantively identical to those of the “Bossy” product described in the Sheppard article. GT’s identified risks and exposures associated with the product included; 1) § 357(c) and legislative regulations thereunder; 2) I.R.S. Notice 99-59 (BOSS transaction); 3) corporate tax shelter regulations/I.R.S. Notice 2000-15; 4) the business purpose doctrine;9 5) the economic substance doctrine;106) the sham transaction doctrine; 11 and 7) the potential retroactivity of Congressional action. 12.13 About six months before GT introduced the Lev301, Sara Williams, previously employed by GT, returned to GT after serving as the Yungs’ tax director. During her tenure with the Yungs, Williams became aware of the Cayman corporations’ cash and the Yungs’ tax risk reasons for rejecting proposals (including a KPMG proposal) for transferring that cash to the U.S. Upon her return, she shared her knowledge with John Michel (J. Michel) of GT, a central figure in this litigation. This is a requirement that a transaction be entered into for a valid business 9 purpose to successfully utilize the strategy. See, e.g., Falconwood Corp. u. United States, 422 F.3d 1339, 1349-52 (Fed. Cir. 2005). 10 This doctrine dictates disregarding transactions that comply with the literal terms of the tax code but lack economic reality. Id. 11 This is a requirement that there be economic substance to the transactions for them to receive favorable tax treatment. Id. 12 The Sheppard “Tax Notes” article stated: [Arthur Andersen’s] BOSS variant should be further impetus to the government to exercise its new section 357 regulatory power. The government could retroactively import to section 301 the new definition of an assumed liability under the amended section 357 to section 301(b). Section 357 was amended ... to eliminate the “subject to” concept and to ask whether someone who assumed a liability was really on the hook for it. 13 The Son of Boss Notice was not issued until August 2000. 10 J. Michel,14a tax partner in GT’s Cincinnati office, began marketing Lev301 to his clients immediately upon its release. He communicated to others within GT that the product had a “short shelf life” and that all concerned had to react to opportunities quickly to ensure a successful sale.15 IV. GT MARKETS, SELLS, AND MAINTAINS LevSOl AS A VIABLE TAX SHELTER TO THE YUNGS AND OTHERS In July 2000, J. Michel and Dean Jorgensen16 met with Ted Mitchel (T. Mitchel)17 at the Fort Mitchell CSC office and introduced the Lev3Ol product. They did not disclose that Lev3Ol was substantially similar to the BOSS; that the February 2000 tax shelter regulations imposed disclosure and listing requirements on corporate participants in such transactions; that the Treasury would likely retroactively make Arthur Andersen’s equivalent “Bossy” product 14 J. Michel joined GT’s Cincinnati office in 1994 and became a full equity partner in 1998. Between 2001 and 2003, J. Michel was a member of the Firm’s Federal Tax Products Group. He was the primary relationship contact with Yung and Yung’s business personnel. He was the point person for the Lev3Ol sale to Yung and for the Yungs’ tax return preparation and review. J. Michel was involuntarily terminated by GT in 2009. 15 Approximately twenty GT non-ancillaiy personnel (tax partners, tax associates, lawyers) participated in meetings and other communications regarding the Lev301, but to avoid a cumbersome factual presentation, we identify only those necessary to provide context to GT’s misconduct. 16 Jorgenson became a full equity partner of the Firm in 1998. By 1999, he was GT’s National Tax Office’s sole sub-chapter C-corporation tax specialist. He was instrumental in the Lev3Ol’s development and was the head of the sales team which marketed, advised, and sold the Lev3Ol to the Yungs. He was the initial opinion writer for the Yungs’ opinions. 17 T. Mitchel joined the Yungs in 1989 and advanced to Chief Financial Officer and a vice-president of CSC. T. Mitchel worked previously at a public accounting firm providing audit related services but did not provide tax advice. He was the primary contact between the Yungs and GT. 11 unlawful; or that GT believed that there was a 90% chance that the l.R.S. would disallow the Lev301 tax benefits on audit. Jorgensen led the presentation of the Lev301 to William J. Yung himself nineteen days later. During the interim, Lev301 ’s so-called “Think Tank” members, inclusive of Jorgensen, met. They anticipated that in two weeks an internal tax opinion letter would be written and reviewed and, additionally, that an outside law firm’s review of the opinion would be obtained. At this point, J. Michel was sending emails to high level partners and others concerning Lev301’s “short shelf life” and the need for its fast delivery for the client. Despite GT’s internal deliberations and reservations, Lev301 was presented to the Yungs as a lawful tax strategy. While Jorgensen and J. Michel explained the need for a non-tax related “business purpose” to use Lev301, the Yungs were never told that the non-tax business purpose had to be the primary motivation. A. The “Worst Case Scenario” Representation Early on, Jorgensen and J. Michel represented to the Yungs and their advisors that with a Lev301 opinion letter from GT (which at this time had not been written), the “worst case scenario” was that if the l.R.S. audited them it could require the Cayman corporations’ shareholders to pay taxes and interest on the Lev301 distributions — but the l.R.S. would not assess penalties. As the trial court later found, Jorgensen and J. Michel understood this advice would be relied upon not only by the Yungs on behalf of the Cayman corporations but also by the shareholders of the companies. Jorgensen and J. 12 Michel further knew when they made the “worst case” representation that it was untrue, and that because of the BOSS Notice, there was a 90% likelihood that the I.R.S. would disallow the tax benefits and assess penalties regardless of whether the participant had an accounting firm’s opinion letter. They also knew that federal authorities were likely to change the law retroactively so as to foreclose GT’s interpretation of § 301’s “subject to” language, 18 which its opinion was relying on. B. The GE and P&G Representation After being informed that GT had yet to complete a Lev3Ol transaction, Yung told Jorgensen and J. Michel that he did not want to be its “guinea pig.” At some point afterward, without any factual basis, J. Michel told Joe Yung19 that a local jet-engine manufacturer (which Joe Yung understood to be General Electric (GE)) and a local consumer products manufacturer (which Joe Yung understood to be Proctor & Gamble (P&G)) had successfully used the strategy to transfer funds to the U.S. 18 I.R.C. § 301(b) provided that the defined distribution of property shall be reduced by liabilities by 1) “the amount of any liability of the corporation assumed by the shareholder in connection with the distribution,” and 2) “the amount of any liability to which the property received by the shareholder is subject immediately before, and immediately after, the distribution.” Under GT’s § 301 interpretation, there would not be a taxable distribution to the Yungs because the amount of the distribution would be fully reduced by the amount of corresponding debt, the amount of debt it was “subject to.” 19 William J. Yung’s son and advisor who served as CSC’s Vice President of Development. He was a proponent of the Lev301 product as he believed it would eliminate much of the international travel required by his position. 13 C. GT’s Internal Discussion and Halt of LevSOl Sales At the time a draft Engagement Letter with GT was being considered by the Yungs, the August 11, 2000 l.R.S. Son of Boss Notice and modifications to the February 28 regulations were issued. GT’s “Think Tank” members exchanged emails expressing concern. Jorgensen tried to distinguish LevSOl. He admitted Lev301 created an artificial high basis in the hands of the shareholders that did not reflect economic reality but relied on the application of an unambiguous statute with its loopholes and the doctrine of “judicial restraint” to support his thesis that LevSOl would work. Jorgensen’s draft opinion letter was sent to Richard Voll20 for review shortly thereafter. Within a matter of days, on August 21, 2000, the Wall Street Journal (WSJ) published an article about the BOSS transaction and Price Waterhouse Cooper’s decision to stop selling it. GT removed the Lev3O1 from its “Client Matrix,” which in effect stopped all sales of the Lev3OI.21 In light of the Son of Boss Notice and the WSJ article, Jorgensen indicated a decision needed to be made if the LevSO 1 “is a go or not” and suggested an approach for increasing fees for the Lev3Ol opinions, which would now need to be updated, and alerting the taxpayer to “assume ideas such as ours to be on the l.R.S. radar screen.” The record contains no 20 Voll, a lawyer by training, joined GT in 2000 to provide experience in the area of corporate tax shelters and opinion letter writing. He also became a partner. He was involuntarily terminated by the Firm in 2003. 21 GT stopped and restarted sales of the Lev3Ol several times during the period from August 2000 until November 2004 when sales were permanently halted. 14 evidence that such concerns were shared with the Yungs. Instead, when T. Mitchel of CSC, having read the WSJ article, contacted Jorgensen expressing concern about the legality of the Lev3Ol, Jorgensen conveyed there was no cause for concern. Jorgensen likewise represented that the Son of Boss Notice caused no concern. He did not inform T. Mitchel that GT suspended Lev3Ol sales in response to the WSJ article. D. The Final Engagement Letter In September 2000, after this conversation, J. Michel sent a revised version of the draft Engagement Letter between GT and the Yungs to the Yungs, having discussed with them their concerns about risk sharing in the event the final outcome was not successful and also CSC’s unwillingness to pay for something which they had not seen fully written up. Prior to sending the letter to the Yungs, Jorgensen emailed J. Michel that GT, because of the BOSS notices, could not back up its representation that its opinion letter would preclude the l.R.S. from assessing penalties. Jorgensen and J. Michel, however, knew that without this “no penalties” representation the Yungs would not go through with the LevSOl transaction. Jorgensen suggested alternatives to soften the removal of the guarantee language.22 22 The three suggested alternatives were: “Reliance by a taxpayer in good faith on our written tax opinion often avoids the successful imposition of penalties . . . .” “Reliance by a taxpayer in good faith on our written tax opinion is intended to preclude the imposition of penalties . . . .” “Our written tax opinion should preclude but without any guarantee the successful imposition of penalties . . . .” 15 The Final Engagement Letter language was altered; it did not put T. Mitchel of CSC on notice that GT would not limit the downside risk to “taxes and interest” as previously represented. The Letter included the following new statement: “Our written tax opinion should preclude the successful imposition of penalties by the U.S. Internal Revenue Service against the shareholders or Companies.” T. Mitchel understood the language to be a reaffirmation of the “worst case” representation and recommended the Yungs execute the Final Engagement Letter on September 15, 2000. GT’s engagement fee for the LevSOl was $900,000, with the bulk of the payment not due until after GT delivered its post-transaction opinion letters. The Final Engagement Letter also stated, “If, based on preliminary conclusions, the Firm cannot express an opinion on the federal income tax matters specifically identified in this engagement letter, the Firm reserves the right to withdraw from this engagement.” This Final Engagement Letter obligated GT to determine prior to the LevSO 1 distributions that it could issue an opinion in support of the transaction. GT was also obligated to provide the Yungs with its “preliminary conclusions” regarding the tax matters before advising the Yungs to proceed with the distributions. After the Son of Boss Notice and the changes to the regulations, GT concluded that individual investors in the LevSO 1 would have to be included on GT’s promoter list.23 Although J. Michel was advised to disclose the list 23 See fn. 7 and accompanying text. 16 maintenance requirement to the Yungs, J. Michel decided not to because he knew that disclosure would kill his sale. While other GT personnel insisted on written notice of the list maintenance requirement so no one played “professional roulette,” J. Michel argued that GT should make a business decision to not maintain a list so that they (as a Firm) would not need to disclose the requirement to potential Lev301 clients. The record contains no evidence that anyone at CSC had knowledge of, or had been informed of, a list maintenance requirement. The Final Engagement Letter did not include any disclosures regarding the risks stemming from the LevSO 1 ’s substantial similarity to BOSS or reflecting that the l.R.S. was likely to deem the LevSOl to be an abusive tax shelter. E. The “More Likely Than Not” Representation In September 2000, GT had outside legal counsel at the New York City office of Baker & McKenzie (B&M) review the first draft of the LevSO 1 opinion letter. Two B&M tax attorneys expressed serious concern about the LevSO 1 ’s ability to satisfy the “business purpose, economic substance and step transaction”24 judicial doctrines and neither was willing to opine that GT had reached the targeted “more likely than not” confidence level for surviving an l.R.S. challenge. GT ignored the legal advice and continued its Lev3Ol sales 24 The step transaction doctrine is the principle that tax liability should be determined by evaluating the transaction as a whole, disregarding non-substantive steps taken to achieve the final result. See, e.g., Falconwood Corp., 422 F.3d at 134951. 17 course. The Yungs were not informed that the Lev3Ol was reviewed by outside counsel or that it received adverse feedback. Although GT had a November 30 delivery date for the opinion letter, delivery did not occur at that point. Discussions were ongoing among GT personnel about how a “more likely than not” opinion could be reached. For example, although questions remained as to whether the lien on the securities was a liability from the point of view of the shareholders and whether it could be said that the parties viewed the step of encumbering the securities as anything other than a transitory step, Voll maintained that I.R.C. § 301’s unambiguous language was the strongest argument for the product. The trial court would later find that the discussions between Jorgensen and Voll, the primary opinion writers, reflected the use of different words to describe the borrowing instruments and, as a consequence, the nature and structure of the debt was not clarified. The requirement that the Lev301 must have a nonrecourse liability was never clearly defined or communicated within GT. The trial court found this failure to define “loan” versus “lien” versus “liability” led to “confusion in determining whether the distribution of the Treasury notes is ‘subject to’ the lien and/or ‘assumed by’ the shareholders.” The trial court found this was “a fatal confusion for the determination of the taxation of the distribution and, thus, the viability of the Lev301 product.” After receiving outside legal counsel’s opinion (the B&M review), Jorgensen communicated that “creative hats” should be put on to further CSC’s business purpose for the bank debt and more specifically the business 18 purpose for distributing encumbered property. The trial court found GT’s decision to get creative and strengthen the “business purpose” for CSC (which GT had acknowledged as a requirement at the July 5 meeting) reflected GT’s concern that they did not have sufficient authority “to author a ‘more likely than not’ opinion and that they would have to fabricate a CSC ‘business purpose’ to try to accomplish that goal.” In preparation for obtaining the bank loans for the Lev3Ol transaction, Jorgensen and J. Michel of GT and T. Mitchel of CSC met on October 3, 2000. “Business purpose” was discussed and T. Mitchel cited it as working capital for future actions, a purpose no more substantial than that discussed prior to B&M’s review. Although GT’s September communications confirmed that GT was aware that the Lev3Ol was a step transaction (like the BOSS transaction), GT did not discuss with T. Mitchel the multiple steps necessary to complete the transaction, the recourse/nonrecourse nature of the lien/loan/liability or the issues of list maintenance and the reporting requirement. The day after this meeting, Voll communicated within GT that he had yet to reach a “more likely than not” confidence level conclusion. Voll indicated that B&M counsel suggested using a lower standard and Voll thought that threshold could be reached using at least the statutory construction argument, the argument which Voll used to describe the “more likely than not” opinion as hovering around 50%. When others expressed surprise that the statutory construction argument did not provide a 60% or 70% probability of surviving an I.R.S. challenge and would prevent a court from moving through to attack 19 the “business purpose” argument, Voll advanced another approach (60-70% probability combined with a 30-40% confidence interval) which would establish a percentage of over 50.1% to reach the level of “more likely than not.” In November, Voll and others were still researching the statutory premise to achieve a “more likely than not” confidence level. As compared to the draft initially provided and reviewed by B&M in September, Voll sent a significantly revised opinion to Jorgensen on December 12. The morning before the December 29, 2000, Lev301 bank loan closing, GT promised to send the Yungs that afternoon its “model opinion” which would serve as GT’s “preliminary conclusions” pursuant to the parties’ Final Engagement Letter. On December 28 at 8:56 p.m., J. Michel was still communicating about an attachment which was the draft of the December 28, 2000 opinion letter; the letter contained “Background” and “Proposed Transaction” sections, but not an “Opinion” section. The “Opinion” section was not contained in prior drafts because of concerns about disseminating it electronically. The December 28 letter, or “short form opinion,” was not delivered to the Yungs until the following morning. J. Michel signed the letter on behalf of GT. The December 28 letter represented that GT intended to issue its written tax opinion by January 15, 2001, and that the opinion would be in substantially the same form as the model opinion. GT represented in the “Opinion” section of the letter that the Firm was of the opinion that it was “more likely than not” that its conclusions would be upheld in litigation with 20 the I.R.S. The letter did not contain any disclosures regarding, and the Yungs were not told before the loan closing about, the list maintenance requirement or the risks stemming from BOSS-like transactions. After the “more likely than not” statement, GT listed these seven opinions: 1) The Company will recognize taxable income as a result of the leveraged distribution only to the extent that the fair market value of the assets distributed exceeds the Company’s tax basis in such assets. For this purpose, the fair market value of the distributed assets is deemed to be at least equal to the amount of liabilities to which the distributed assets are subject. 2) The Company will reduce its earnings and profits by the excess, if any, of the fair market value of the assets distributed over the amounts of the liabilities to which the distributed assets are subject (“an excess distribution”). 3) A shareholder will recognize taxable income only to the extent of an excess distribution. Furthermore, a shareholder will reduce his or her tax basis of the stock held in the Company as a result of the leveraged distribution only to the extent of an excess distribution. 4) A shareholder will have a tax basis in the assets distributed equal to the fair market value of such assets at the date of the distribution. Such tax basis will not be reduced by any liabilities to which the distributed assets are subject. 5) A shareholder will not be in constructive receipt of a distribution, e.g., a dividend, upon any later payment by the Company of the liabilities to which the distributed assets are subject. 6) Judicial doctrines will not override opinions expressed on the aforementioned issues. 7) A shareholder or the Company will not be subject to any tax penalties in relying in good faith upon the opinions expressed on the aforementioned issues. (Emphasis supplied). On December 29, 2000, the Yungs carried out the first two steps of the Lev301 strategy in reliance on GT’s representations regarding the Lev301. 21 They executed a $30 million loan from Firstar Bank and purchased $30 million in T-notes with the loan money. On the same day, the Cayman corporations held board meetings to declare a dividend of the encumbered securities, and the T-Notes were transferred from the Cayman corporations’ bank accounts to the custodial accounts for the Yungs. (Later the same day, another Lev301 client also made a distribution to its shareholders using the Lev301 strategy pursuant to J. Michel’s advice.) At the time GT made its December 28 representations, Voll had not yet concluded that a “more likely than not” confidence level was possible regarding favorable l.R.S. treatment of the transaction. At the same time the Lev301 distributions were being made, the Yungs were in the process of acquiring Lodgian, Inc., a publicly traded hospitality company. As the trial court later found, that acquisition was not the “business purpose” underlying the Lev301, and the Yungs would not otherwise have transferred the Cayman corporations’ cash at that time. F. The Sheppard Predictions Become Reality Five days after the Cayman corporations’ Lev301 distributions to their shareholders, the U.S. Treasury Department issued temporary and proposed regulations. The effect was to invalidate GT’s argument that a shareholder who received a T-Note distribution “subject to” a liability that was recourse to the distributing corporation could reduce the value of the T-Notes by the amount of the liability. GT internal discussions lacked consensus, but several GT partners expressed the view that the regulations killed the Lev301. Jorgensen 22 circulated a “301 regulation” memo; among other concerns, he stated that under the new regulations, the shareholders would probably be treated as receiving a “constructive dividend” when the corporation paid on the loan and consequently, this outcome would destroy the Lev301. J. Michel received this memo but continued to promote the Lev301 product. A few days later, GT again ended the sale of the Lev301 until further notice. G. The “It’s All Good” Representation Days before GT’s final opinion was due to the Yungs, and even though GT had not reached these conclusions, J. Michel nevertheless told the Yungs that GT was of the opinion that the January 4, 2001 regulations did not adversely affect the Cayman corporations’ Lev301 distributions because the distributions were finalized prior to the effective date of the regulations. J. Michel also represented that GT believed that “such Regs may more favorably impact the favorable tax status for the 2000 year transaction.” J. Michel did not disclose to the Yungs that GT was no longer selling the Lev301 in response to the January 4 regulations. J. Michel later admitted that he made the statement that the January 4 regulations did not impact the Lev301 transaction to “buy time with the client.” Voll concluded on or about January 23, 2001, that the January 4 regulations were retroactive and applicable, i.e., the Lev301 was substantially similar to the BOSS. GT did not inform the Yungs or their advisors about this determination. Specifically, GT did not advise the Yungs to unwind the transaction at this time, even though that was a viable course of action. Three 23 days after Voll’s conclusion that the regulations were retroactive, the Yungs, however, emailed J. Michel asking him whether the Cayman corporations should pay off the bank loans since no opinion had been delivered yet and whether GT would pay the interest “if the deal never finalize[d] as planned.” Voll then shared a draft of the opinion letter with J. Michel. Voll informed J. Michel not to share the draft opinion as he was still researching for ways to strengthen the Yungs’ “business purpose.” J. Michel pressed Voll not to over-emphasize “business purpose” in discussions with the Yungs as he was unsure if they would sign the representation they had already been given. The draft Opinion Letter did not have the December 28 letter’s seventh tax opinion (quoted and emphasized above) regarding penalties, and it also did not disclaim the prior representation regarding penalties. On February 6, 2001, to prevent the Yungs from paying off the loan and terminating the Lev3Ol engagement, J. Michel sent the Yungs an incomplete draft of the Wytec Opinion Letter. This was the first full-length opinion provided to the Yungs. J. Michel explained missing verbiage would be incorporated related to the new regulations which would only serve to strengthen the opinion. T. Mitchel did not review the Wytec opinion to identify risks related to the transaction because he assumed that any material risks would have been disclosed to them by GT, their accounting firm and tax expert, before the Yungs authorized the Lev3Ol distribution. On February 19, GT’s Jorgensen warned in internal emails that for completed Lev3O1 transactions GT should consider a “rescind and restore” 24 strategy and that prospects should be told of the material and significant risk that the regulations would result in a constructive dividend to shareholders. He stated that in the Yungs’ case, accelerating the bank debt retirement would assure completion of the Lev301 strategy before the issuance of the constructive dividend regulation, which would not have a retroactive effect. Voll responded that the clients were already told about potential changes to the law and regulations, so “That covers us. . . . We advise the client on regulatory activity. Beyond that - what we say about what could happen is conjecture.” GT never told the Yungs that they might have to accelerate payment on the bank debt to save the transaction. On February 20, Voll agreed with Jorgensen to leave the Lev3Ol product out there. But while Jorgensen’s position was to further inform clients of litigation risks, Voll’s position was that in the unlikely event the l.R.S. caught the transaction, the GT opinion should stop a penalty, which is “what the client paid for.” Emails among various GT personnel acknowledged that the viability of the LevSO 1 had not yet been determined because Jorgensen and Voll had not yet agreed on the constructive dividend issue and, further, that undoing a LevSO 1 transaction might not be simple. In mid-February, J. Michel informed others at GT that the Yungs’ first payment to GT was soon due. He stated the remainder was due at the June bank debt retirement but it could be due sooner than planned in light of the threatened issuance of regulations. 25 After having advised T. Mitchel in February how to prepare the tax return regarding the LevSOl, in March, J. Michel monitored the Yungs’ tax returns to assure they did not disclose income due to the Lev3Ol transaction. Discussions continued about the business purpose representations that the Yungs were to make for the Lev3Ol product. In April 2001, GT decided to start selling the Lev3Ol product again. Internally, GT’s Voll expressed concern about strength of a “business purpose” in sales to assure the integrity of a “more likely than not” opinion and referenced the Yungs as having a weak business purpose which GT overcame by using other devices (meaning it was after December 31, 2000 and GT’s further inquiry that other shareholder investments were used to develop the Yungs’ business purpose). In early July, Voll informed J. Michel he was still working on the Yungs’ opinions and still concerned about “business purpose.” While these internal discussions continued, no one at GT ever expressed their concern about “business purpose” to the Yungs. More than five months after the Cayman corporations’ transaction closed, internal GT discussions were ongoing about the list maintenance requirement. Nevertheless, GT did not inform the Yungs of the possibility/probability of that issue even though GT notified its field agents of the requirement to keep an internal list. Selling the LevSO 1 product again, GT continued to omit crucial details about the product’s risks and the weaknesses of GT’s legal arguments. J. Michel had sold three Lev3Ol engagements to his clients by this time; he continued to push back on Voll’s “business purpose” concerns and stressed 26 pushing “business purpose” to an extreme would result in making the product non-saleable. GT changed the product in the Spring of 2001 in response to adverse feedback from a prospective client who sent the LevSO 1 proposal to outside legal counsel for review. At that juncture, GT required that the loan between the bank and the distributing company in LevSO 1 transactions be nonrecourse to avoid application of the January 4 regulations, and required a representation to that effect from the client. Both Voll and J. Michel had reviewed the Yungs’ loan documents and knew that GT was repudiating the version of the Lev3Ol it had advised the Yungs to execute in December 2000. On May 21, 2001 and June 19, 2001, the Yungs signed a “Representation Letter for the Opinion Letter of Grant Thornton LLP” as to Wytec and Casuarina, respectively. Although later required for LevSO 1 product purchasers, the representations did not include a representation that the note/loan/mortgage was “nonrecourse” or the term “bona fide business purpose.” On July 11, 2001, besides the “business purpose,” GT was also working on the “economic substance” portion of the Yungs’ opinion and internal discussions continued about “constructive dividend” and the intricacies of state law defining the nature of the loan transaction and its relationship to the shareholders. On July 13, 2001, the Family Trust filed its 2000 U.S. Income Tax Return which GT reviewed. On August 8 and 13, 2001, GT delivered the Final Wytec and Casuarina Opinion Letters. Unlike the Draft Wytec Opinion Letter, the August Opinion Letters contain several attachments, including the bank loan documents. 27 Those documents confirmed the recourse nature of the loans. The six opinions contained in the August Opinion Letters were substantively identical to those in the 2001 Draft Wytec Opinion Letter. T. Mitchel relied on GT as the tax expert and did not review the “Analysis” section or the two appendices of either opinion. To account for the January 4 regulations, Voll characterized the loan obligation as “nonrecourse” in both Opinion Letters, although that was a determination of Ohio law which GT as a public accounting firm was not qualified to make. GT’s opinions were also premised on the existence of a motivating non tax corporate business purpose, although the Yungs had made no representation of such. Jorgensen, J. Michel and Voll were all aware that the Yungs’ primary motivation for making the Lev3Ol distributions was the avoidance of U.S. federal income tax on the transfer of the Cayman corporations’ cash. Although GT could not have reasonably believed this statement, GT’s August Opinion Letters were based on the conclusion that the LevSOl would survive application of the step transaction judicial doctrine. On August 21, 2001, a WSJ article was published which again called BOSS strategies into question. GT stopped LevSOl sales again pending a full consideration of the article. On September 26, 2001, the Yungs paid off the loan which was the basis of the Lev3Ol transaction. On October 7, 2001, the Yungs’ l.R.S. form 1040 was filed after review by GT. GT also prepared the Yungs’ 2001 federal income tax returns in September 2002. The Yungs relied on the August Opinion 28 Letters when deciding to not report the $30 million distribution on the 2000 tax return and to not report the repayment of the $30 million on the 2001 federal income tax return. H. The l.R.S. Exam of GT and the Yungs’ Audit In early 2002 the l.R.S. initiated an exam of GT. The exam expressly targeted GT’s promotion of potentially abusive tax shelters. GT partners were aware that the exam substantially increased the risk that the l.R.S. would obtain the names of its Lev301 clients and increased the clients’ audit risk. Nevertheless, GT did not inform the Yungs of the exam until it became public in September 2003. In June 2002, additional tax shelter regulations were issued, and GT again took the Lev301 off the Client Matrix in July. Once again, the Yungs were not informed. Despite internal GT discussion of the regulations’ application to the Cayman corporation transactions, no one told the Yungs of these concerns. In September 2002, GT again resumed sales of the Lev301. The product was not permanently removed from the Client Matrix until late November 2004. At the same time, GT stopped issuing opinions to clients for transactions that occurred prior to the removal. As before, the Yungs were not informed in 2004 that GT had stopped selling or defending the Lev301 transaction. The l.R.S. exam of GT led to GT identifying to the l.R.S. the Cayman corporations and the Yungs as participants in the Lev301. In Spring 2004, the l.R.S. audited the Yungs concerning the Lev301, and in Spring 2005, the l.R.S. 29 assessed back taxes and penalties. GT continued to represent the Yungs. In February 2007, the Yungs and the I.R.S. reached a settlement. I. The Yungs’ Lawsuit Against GT Shortly afterward, in August 2007, William J. Yung and Martha A. Yung25 filed their complaint against GT in Kenton Circuit Court requesting a declaration that their September 2000 engagement contract did not limit recovery of damages from GT to the engagement fee or bar reimbursement of reasonable legal fees. Mr. and Mrs. Yung also alleged fraud, negligence and breach of contract. In the June 2008 amended complaint, Mr. and Mrs. Yung requested punitive or exemplary damages as appropriate. The bench trial in this action lasted 22 days and included over 40 witnesses and more than 600 documents. The trial court concluded the Final Engagement Letter did not protect GT from negligence, nor did it limit the Yungs’ monetaiy damages, and that GT committed fraud and gross negligence in its scheme to sell and maintain the Lev3Ol product to the Yungs. The trial court awarded the Yungs the $900,000 engagement fee paid to GT with prejudgment interest; approximately $19 million in taxes, interest, and penalties ($3,782,786 to William and Martha Yung and $14,632,441 to the Family Trust); and $80 million in punitive damages ($55,000,000 to William and Martha Yung and $25,000,000 to the Family Trust). Following the trial 25 The 1994 William J. Yung Family Trust was named as an additional party. As the trial court noted, although the 1994 William J. Yung Family Trust is designated a defendant in the complaint, its interests are more aligned with those of William J. Yung and Martha A. Yung. 30 court’s denial of GT’s CR 52.02 motion for additional findings of fact and CR 59.05 motion to alter, amend, or vacate the judgment,26 GT appealed the trial court’s judgment. As noted, the Court of Appeals in a 2-1 decision reduced the punitive damage award to $20 million, but affirmed the trial court on the other issues raised by GT. On discretionary review, the Yungs assert that the $80 million punitive damage award does not violate due process. GT asserts that the Yungs cannot show that they justifiably relied on GT’s advice; that the compensatory damage award is an improper windfall; and that even if GT is liable, the punitive damage award is improper. For the reasons stated below, we affirm the Court of Appeals’ decision except for the reduction of the punitive damage award. Additional facts are presented below as necessary. ANALYSIS 1. LIABILITY AND COMPENSATORY DAMAGES A. Liability for Fraud The Yungs alleged that GT engaged in fraud by misrepresentation and by omission.27 The trial court concluded GT had committed both, and the Court 26 The trial court modified its award of prejudgment interest. 27 To succeed on a fraud by omission claim, the proof must show: “(1) the defendant had a duty to disclose the material fact at issue; (2) the defendant failed to disclose the fact; (3) the defendant’s failure to disclose the material fact induced the plaintiff to act; and (4) the plaintiff suffered actual damages as a consequence.” Giddings & Lewis, Inc. u. Indus. Risk Insurers, 348 S.W.3d 729, 747 (Ky. 2011) (citing Rivermont Inn, Inc. v. Bass Hotels & Resorts, Inc., 113 S.W.3d 636, 641 (Ky. App. 2003)). Fraud by omission is grounded in a duty to disclose. Id. The trial court concluded that the professional accountant-client relationship between GT and the Yungs was a relationship requiring confidentiality and full disclosure under Kentucky Revised Statute (KRS) 325.440, 26 U.S.C. § 7225, and 31 of Appeals upheld that conclusion.28 GT argues before this Court that the circuit court’s judgment should be vacated because the Yungs cannot, as a matter of law, establish justifiable reliance on its misrepresentations.29 Fraud by misrepresentation is established by clear and convincing evidence: (1) that the declarant made a material representation to the plaintiff, (2) that this representation was false, (3) that the declarant knew the representation was false or made it recklessly, (4) that the declarant induced the plaintiff to act upon the misrepresentation, (5) that the plaintiff [reasonably or justifiably] relied upon the misrepresentation, and (6) that the misrepresentation caused injury to the plaintiff. Treasury Department Circular No. 230. The trial court found GT failed to disclose to the Yungs 1) the list maintenance requirement, 2) Baker 86 McKenzie’s unfavorable opinion of the Lev301, 3) GT’s removal of the Lev301 from its product offerings in response to the August 2000 and January 2001 regulations, and 4) the differing views as to Lev301’s substantial similarity to BOSS. Finally, the trial court found that these omissions induced the Yungs to enter into and complete the Lev301 strategy and caused them to incur associated federal income tax, interest and penalty damages. Although the fraud by omission claim is not at issue before this Court, the reprehensibility of the associated misconduct is nevertheless part of the totality of the circumstances considered under the punitive damage analysis. 28 Before the Court of Appeals, GT argued that the trial court’s fraud by omission determination was error because the trial court improperly concluded that GT was the Yungs’ fiduciary and GT had a duty to immediately disclose concerns raised in the course of internal deliberations. GT does not maintain this fraud by omission argument before this Court, although it continues to assert the trial court erred in concluding that GT was the Yungs’ fiduciary. In fact, the trial court only once referred to the relationship as “fiduciary” otherwise treating it as “a trusting” professional relationship. At this juncture, GT argues that the erroneous “fiduciary” finding cannot be considered when evaluating the reprehensibility of its actions. 29 Although GT makes a broad statement that all claims should be dismissed because the Yungs cannot show justifiable reliance, as it did in the Court of Appeals, GT does not directly challenge the trial court’s findings and conclusions of law that it was grossly negligent. 32 Flegles, Inc. v. TruServ Corp., 289 S.W.3d 544, 549 (Ky. 2009) (citing United Parcel Service Co. v. Rickert, 996 S.W.2d 464 (Ky. 1999)); McHargue v. Fayette Coal & Feed Co., 283 S.W.2d 170 (Ky. 1955); Restatement (Second) of Torts § 537 (1977)). Because the misrepresentation must relate to a past or present material fact, a “mere statement of opinion or prediction may not be the basis of an action.” Flegles, 289 S.W.3d. at 549 (quoting McHargue, 283 S.W.2d at 172). However, a fraud claim may be based on a “deceptive” opinion or prediction where “the opinion either incorporates falsified past or present facts or is so contrary to the true current state of affairs that the purported prediction is an obvious sham.” Id. at 549. But, if the recipient of a fraudulent misrepresentation has the opportunity to verify a representation through ordinary vigilance or inquiry and does not do so, the false representation, even when made with the intention to deceive, has no legal effect on the rights of contracting parties. Id. (citing Ky. Elec. Dev. Co.’s Receiver v. Head, 68 S.W.2d 1, 3 (Ky. 1934)). Finally, the recipient of a fraudulent material misrepresentation is justified in relying upon a representation when the opinion giver: (a) purports to have special knowledge of the matter that the recipient does not have, or (b) stands in a fiduciary or other similar relation of trust and confidence to the recipient, or (c) has successfully endeavored to secure the confidence of the recipient, or (d) has some other special reason to expect that the recipient will rely on his opinion. 33 Id. at 551 (quoting § 542 of Restatement (Second) of Torts (1977-2008)). The trial court found that J. Michel, Jorgensen and Voll made numerous material misrepresentations to the Yungs regarding the nature and risks of the Lev301 product from July 2000 until the termination of their relationship and that the Yungs reasonably relied on the misrepresentations. The trial court identified four specific misrepresentations: 1) the “worst case scenario” representation,30 2) the GE and P&G representation,31 3) the “more likely than not” confidence level representation,32 and 4) the “it’s all good” representation of January 2001.33 GT insists that a constellation of facts precluded the Yungs from justifiably relying on any alleged misrepresentations about the tax and penalty risks of the Lev301 strategy, i.e., the Yungs knew that dividends were ordinarily taxable; undertook the strategy for the specific purpose of avoiding taxes; knew GT’s tax opinion would provide an interpretation of the Lev301 as only slightly 30 The trial court found reasonable reliance for the “worst case scenario” representation because the Yungs’ relationship with GT was one of established trust and GT consistently represented that an IRS audit in the “worst case” scenario would result in the Wytec and Casuarina shareholders paying the taxes on the distribution with interest but not penalties. 31 The trial court found the GE and P&G representation was reasonably relied upon because Joe Yung, one of the Yungs’ closest advisors and investment advisor for the Trust, was encouraged to use the Lev301 strategy by GT, a trusted tax advisor, alluding to other clients entitled to confidentiality (seemingly GE and P&G), as using the product. Joe Yung could not verily these representations. 32 This misrepresentation, a key issue in this case, is discussed infra. 33 As to the January 2001 “it’s all good” representation, the Yungs reasonably relied upon it when questioning GT regarding the effect of new regulations on the Lev301 strategy. In addition to being told, “it’s all good,” they were told it “may even be better for the strategy.” Also, at this point, GT failed to advise the Yungs of the possibility, and perhaps necessity, of unwinding the Lev301 transaction to avoid adverse tax effects on the distribution. 34 more than 50% likely to be upheld in court; and obtained the opinion to allow them to argue that they should not be assessed penalties for failing to report the dividend if the l.R.S. disallowed the strategy. GT takes particular aim at the trial court’s conclusion that the Yungs reasonably relied on the “more likely than not” opinion for the Lev301. Before turning to GT’s “matter of law” argument regarding justifiable reliance, we note that although this Court has never expressly addressed the issue, whether reliance is justified (or as sometimes stated, reasonable) is a question of fact in all but the rarest of instances. See, e.g., IAS Service Group, L.L.C. V. Jim Buckley & Associates, Inc., 900 F.3d 640, 649 {5th Cir. 2018) (the issue of justifiable reliance is generally a question of fact); PharMerica Chicago, Inc. V. Meisels, 772 F. Supp. 2d 938, 957-961 (N.D. Ill. 2011) (justifiable reliance is a question of fact but can be determined as a matter of law “when no trier of fact could find that it was reasonable to rely on the alleged statements or when only one conclusion can be drawn”); McColgan v. Mutual of Omaha Insurance Co., 4 F. Supp. 3d 1228, 1230-34 (E.D. Cal. 2014) justifiable reliance is a fact question, “except in the rare case where the undisputed facts leave no room for a reasonable difference of opinion”). In Hanson v. American National Bank & Trust Co., 865 S.W.2d 302, 307 (Ky. 1993), this Court recognized that “[i]t was for the jury to decide if it was reasonable for Hanson to rely upon such inducements” to enter a contract and whether he “in fact did rely.” If the jury answered affirmatively as to both questions, “a tort action for fraud is the appropriate remedy.” Id. In this bench trial, the court considered 35 the evidence and made factual findings of justifiable reliance. As discussed below, we find substantial evidence supporting those findings and no grounds for deeming them clearly erroneous, but we nonetheless address GT’s claim that the reliance question should be disposed of as a matter of law. 1. Justifiable Reliance: GT Tax Opinion The trial court described numerous events related to the “more likely than not” opinion misrepresentations, including the following: On December 28, 2000, GT had its final pre-transaction opportunity to correct its prior errors regarding “listing,” “recourse,” and its inability to issue a “more likely than not” standard opinion. As no financial transactions had occurred, GT could have notified the Yungs at this point that GT did not have a “more likely than not” opinion and the Yungs should not go through with the transaction until that occurred or should not go through with the transaction at all. GT misrepresented in writing on December 28, 2000, that it was “of the opinion” that it was “more likely than not” that a court of law would uphold the non taxability of the Lev301 transaction. GT’s confidence level in its opinion was extraordinarily material to the Yungs’ decision to authorize the LevSO 1 distributions and to the decision to not report the income on the Yungs’ 2000 and 2001 federal income tax returns. In the Februaiy 2001 draft Wytec Opinion Letter, GT had not yet determined what the impact of the January 4, 2001 regulations would be on its opinion. At this point GT was no longer selling the Lev301 to clients. GT had not yet released and did not believe that they had a “more likely than not” opinion on the Lev301. GT sent the incomplete draft letter to prevent the Yungs “from terminating the LevSOl engagement.” In the August 2001 Final Opinion Letters, GT represented that it held a “more likely than not” confidence level for the Wytec and Casuarina Lev301 distributions. The opinions were based on the premise (unjustified) that the loan documents for the Lev3Ol transaction were nonrecourse. Furthermore, J. Michel and Voll knew that the Yungs did not satisfy the “business purpose” doctrine as the transaction was motivated by a tax purpose. S6 The trial court found that the Yungs were entitled to rely, and did reasonably rely, upon trusted tax advisor GT’s representations that the Lev3O1 product was guaranteed at a “more likely than not” standard and that no taxable income would result from the Lev3Ol transaction. In the trial court’s view, the Yungs reasonably relied upon the representations considering a) the December 28, 2000 “short form opinion” letter issued by GT before the first step of the LevSO 1 strategy was executed stated “The Firm is of the opinion that under the current U.S. federal income tax law it is more likely than not (i.e., there is a greater than 50% likelihood) that, that if challenged by [the I.R.S.] the following conclusions will be upheld . . . .” Those conclusions included no income tax liability and no exposure to tax penalties. The letter stated: “You may rely on this representation to complete the transaction as proposed and discussed most recently on 12/28/2000;” b) the August 2001 Wytec and Casuarina “Opinion Letters” stated that the LevSO 1 strategy is guaranteed at a standard of “more likely than not.” Although never brought to the Yungs’ attention, the substance of the letters, as well as the approach to “business purpose,” had changed significantly since the July 2000 meetings and the December 28, 2000 Opinion Letter; and c) because GT and the Yungs never shared the same definitions of “recourse” and “nonrecourse” throughout the process, Voll’s use of “nonrecourse” in the final August 2001 Opinion Letters did not put the Yungs clearly on notice of the requirement that the loans had to be nonrecourse. The Yungs were not aware “that they were required to have GT’s professional advice analyzed” and, in fact, GT “thwarted and discouraged” the Yungs from getting a second opinion because it feared another accounting firm might steal the product. Furthermore, the Yungs reasonably relied upon GT’s advice when it filed tax returns in 2000 and 2001 that did not report the $30,000,000 distribution. GT insists that as a matter of law, the Yungs were presented with information in the tax opinion itself sufficient to cause them to doubt the propriety of the tax shelter for tax avoidance purposes but they nevertheless 37 willfully blinded themselves to that information. Having reviewed GT’s tax opinion — which GT held out as satisfying the Internal Revenue Code’s criteria regarding an opinion on which a taxpayer may reasonably rely — we cannot conclude that it served as a red flag and negated the Yungs’ reliance upon the opinion. GT’s argument is also premised on evidence that the Yungs, sophisticated business owners, knew that the “tax shelter strategy,” engaged in for the sole purpose of avoiding taxes on dividends, was risky — if audited, there was nearly a 50% chance of the strategy being disallowed — and that their purpose in buying the “more likely than not” tax opinion was, if audited, to avoid penalties for failing to report the income. Consequently, GT argues that, as a matter of law, having been on notice of the risk of taxes and penalties, the Yungs could not justifiably rely on claimed misrepresentations or omissions that allegedly understated the risk that the l.R.S. would impose taxes and penalties.34 In short, GT maintains that its “more likely than not” opinion reflects a weak probability that the LevSOl would pass l.R.S. muster. The Yungs respond that the phrase “tax shelter” was never used by GT when the Firm approached them about the LevSOl (a point the Yungs stress), but instead LevSO 1 was presented as a transaction that would allow them to 34 This argument suggests that successful business people cannot be found to justifiably rely on their own tax advisors if any element of risk is involved in a transaction. Obviously, successful business people do have a right to rely on professional advice that they believe represents their advisors’ best professional judgment. 38 bring the Cayman corporations’ profits into the U.S. without income tax consequences. In the Yungs’ view, [t]his case is not about a “more likely than not” tax opinion that turned out to be wrong through an error in professional judgment. This case is about GT’s deliberate fraudulent misrepresentations and omissions concerning the legitimacy of its opinion. The ‘more likely than not’ nature of GT’s opinion did not alert the Yungs, and would not alert any client, to the risk that they were being lied to and deceived. GT sidesteps this response. The Firm acknowledges that justifiable reliance is often a fact question but asserts this is a case in which, under Kentucky law, undisputed red flags about the risks of the transaction preclude justifiable reliance. None of the Kentucky cases GT cites is remotely similar to this case. 35 GT relies more particularly on three tax fraud cases from other jurisdictions, Shalam u. KPMG LLP, 89 A.D.3d 155 (N.Y. App. Div. 2011), Salt Aire Trading LLC v. Enterprise Bank and Trust Corp., 2013 WL 775747 (N.Y. Super. Ct. Feb. 25, 2013), and DDRA Capital, Inc. v KPMC, LLP, No. l:04-cv00158 (D.V.I. June 6, 2014) (citing Shalam and Salt Aire),36 to argue that the 35 Examples, cited by GT and all distinguishable, include: Keith v. Robinson, 2006 WL 3524056 (Ky. App. Dec. 8, 2006) (no justifiable reliance on oral assurance contrary to written documents); SMA Portfolio Owner, LLC u. Corporex Realty & Inv., LLC, 112 F. Supp. 3d 555 (E.D. Ky. 2015), affd, 661 F. App’x 305 (6th Cir. 2016) (same); and Davis v. Siemens Med. Solutions USA, Inc., 2007 WL 710133 (W.D. Ky. Mar. 6, 2007), affd, 279 F. App’x 378 (6th Cir. 2008) (same); Repub. Bank & Tr. Co. v. Bear, Steams & Co., 707 F. Supp. 2d 702 (W.D. Ky. 2010), affd, 683 F.3d 239 (6th Cir. 2012) (lack of justifiable reliance due to written disclosures that informed plaintiff of the risk associated with securities being purchased). 36 The Third Circuit issued its non-precedential decision, DDRA Capital, Inc. v. KPMG, LLP, 710 F. App’x 522 (3d Cir. Sept. 6, 2017), during the briefing for this Court. As discussed below, the Third Circuit affirmed the district court’s grant of summary judgment on DDRA’s fraud and negligent misrepresentation claims, but on different 39 taxpayer who contracts to receive a “more likely than not” tax opinion for tax shelter strategies for the purpose of penalty protection is necessarily on notice of the risks of I.R.S. scrutiny and potential tax and penalty liability and, therefore, cannot have justifiably relied on the opinion. We find these tax shelter cases distinguishable. In each case, multiple factors placed the plaintiff on notice that he could not reasonably rely on claimed assurances either that the tax shelter itself was legal or that it was more likely than not to be allowed by the I.R.S. In each case, the court granted summary judgment to the defendant. In Shalam, the taxpayer faced tax liability on approximately $50 million in capital gains resulting from the public offering of securities for a company he founded. 89 A.D.Sd at 156. To offset the gains, he was solicited by an accounting firm for participation in a tax shelter, which the I.R.S. subsequently disallowed. The New York Appellate Division concluded that the plaintiffs own statements, admissions, and possessed documents “conclusively establishefd] that he knew or should have known that he was participating in a scheme of doubtful legality.” Id. at 157-58. The plaintiff 1) understood that the tax shelter was not an investment but “an artificial transaction to create paper grounds. The circuit court concluded that the crucial red flag was that DDRA did not believe they or their entities would suffer actual losses under the proposed transaction but still planned to claim those artificial losses as deductions. Intricate tax law knowledge was not needed to see this red flag. Consequently, GT’s reasons for relying on DDRA Capital changed from it being a case which exemplifies that a “more likely than not” tax opinion puts the taxpayer on notice of the risks of I.R.S. scrutiny to a case which exemplifies that just one “crucial red flag” was sufficient to preclude fraud. We note, as described below, that red flag was not the “more likely than not” opinion, the discussion of which was relegated to a footnote. 40 losses against which he could offset his capital gains so as to avoid paying taxes on those gains;” 2) signed loan documents as part of the tax avoidance scheme that reflected purported loans that were not actually made; and 3) understood that the provided “more likely than not” official opinion letter was not a true assessment of the tax shelter’s legality, but rather an effort to protect participants from I.R.S. penalties. Id. GT emphasizes the Shalam court’s observation that a “more likely than not” tax opinion puts “an ordinary person on notice that the odds in favor of legality could be as slim as 51% to 49%,” and these odds indicate “substantial risk that his tax avoidance strategy will not pass muster with the I.R.S.” Id. at 158. Although the court did so comment, it further concluded that the “[p]laintiff was presented with information sufficient to cause him to doubt the propriety of the [tax shelter] scheme for tax avoidance purposes, and willfully blinded himself to that information by failing to ask questions, pay attention to details, or read the documents he signed. Thus, he [could not] demonstrate reasonable reliance.” Id. at 159. In Salt Aire, the plaintiff, Kelly, sought to reduce the tax consequences on approximately $36 million in income. 2013 WL 775747, at *1. The Salt Aire court listed multiple reasons that Kelly, a successful corporate executive with a legal education, certainly “knew or should have known that he was participating in a scheme of doubtful legality.” Id. at *6 (citing Shalam, 89 A.D.3d at 157). Kelly 1) knew that the purpose of the tax shelter was to create artificial losses to offset his income, not to profit from foreign currency trading; 41 2) created and controlled a number of entities to implement the necessary transactions for the tax shelter and was an active participant in formulating the terms of the shelter; and 3) knew there was a risk that federal and state taxing authorities would disallow the tax shelter losses on his tax returns but obtained a “more likely than not” opinion letter to try to avoid penalties. Id. at *6-7. The Salt Aire court held that the plaintiffs (Kelly and his wife) could not invoke the opinion letter as proof of their reasonable reliance for the same reasons the Shalam plaintiff could not. That court concluded; “All that matters is that plaintiffs had sufficient information and knowledge to cause them to doubt the propriety of the [tax shelter] scheme. Plaintiffs harbored those doubts from the inception of their involvement with defendants’ tax shelter.” Id. at *6. “[Pjlaintiffs’ understanding of the shelter was sufficient in all material respects, and no information that was allegedly concealed from them could have blinded them to its obvious and palpable risks.” Id. at *8. In DDRA Capital, the taxpayers collectively earned income of over $84 million and sought advice from different advisors to minimize their tax burden before deciding upon a tax shelter suggested by KPMG. The district court identified five danger signals which should have caused the plaintiffs to doubt the propriety of the tax shelter: 1) the “[pjlaintiffs’ decision to execute the [tax shelter] was primarily tax-driven, not profit-driven;”37 2) the “[pjlaintiffs 37 This danger signal is based upon the foundational principle of tax law that deductions are allowed from tax income only for “losses incurred in any transaction entered into for profit.” I.R.C. § 165(c)(2) (2010). Losses from sham transactions, a 42 executed paperwork containing false representations that they knew or should have known was being processed to give [the tax shelter] an appearance of investment purpose;” 3) “Defendant advised Plaintiffs to report their taxes in a way that would avoid I.R.S. detection of the [tax shelter] transaction;” 4) “the [tax opinion letters (not from KPMG but rather a law firm)] contained a ‘more likely than not’ opinion;” and 5) “[the tax shelter] produced results that Plaintiffs should have realized were too good to be true.” No. l:04-cv-00158 at 37-49. Given the Third Circuit’s September 2017 decision, DDRA Capital now stands apart from Shalam and Salt Aire-, the Third Circuit identified a single “crucial red flag” which kept the plaintiffs from justifiably relying on the defendant’s misrepresentations. 710 F. App’x at 526. The crucial red flag was “the plaintiffs did not believe they or their entities would suffer actual losses but still planned to claim those losses as deductions.”38 As for the other danger signals, the circuit court stated that although not sufficient in themselves, they “should have contributed to the suspicion the knowledge of loss generation should already have triggered.” Id. at 528. Discussing the suspicions and accompanying investigations which should have been transaction which has no purpose or economic effect other than the creation of tax deductions, are not allowable under I.R.C. § 165(c)(2). DDRA Capital, Inc. v KPMG, LLP, No. l;04-cv-00158 at 37 (D.V.I.) (citing Bail Bonds by Marvin Nelson, Inc. v. C.I.R., 820 F.2d 1543, 1548 (9th Cir. 1987)). 38 DDRA Capital, 710 F. App’x at 526 (“We think no reasonable juror could conclude that plaintiffs’ knowledge of a fictional loss generated without economic reality would not have been a red flag . . . .” Id. at 527.). 43 generated, the court referenced the tax opinion (provided by a law firm KPMG had engaged) in a footnote: Given that the very nature of the transaction—loss generation— was a red flag, and that it would seem to be the I.R.S.’s views that would matter, plaintiffs should have viewed comments like these [-“as part of the paperwork being in order, you should have an opinion from a recognized law firm saying that whatever you’re doing is correct and, you know, legal and good;” the transaction “was fully within the law, but there was no guarantee the I.R.S. would say that it was fully within the law or whatever”-] with increased suspicion. Apparently, they did not. Instead, Baldwin thought that the tax opinion was “a magic shield” that would prevent him from paying any penalties or interest in the event the I.R.S. disallowed the transaction. Scott did not recall ‘ask[ing him]self why isn’t KPMG giving [him] a tax opinion if this transaction is so good.’ DDRA Capital, 710 F. App’x at 529 n.6 (citations to record omitted). Our Court of Appeals differentiated these cases from the case at bar, noting Shalam, Salt Aire, and DDRA Capital, No. 1:04-0158 (D.V.I.) (as noted, the Third Circuit decision had not issued at that time), are cases in which the taxpayer admitted to possessing information which would conclusively establish that he knew or should have known that the tax shelter scheme was of doubtful legality. The Court of Appeals concluded that in the absence of a taxpayer’s admission to actual or constructive knowledge of information showing that the tax opinion letter was fraudulent, the question of justifiable reliance is generally an issue of fact. GT counters that these tax fraud cases do not require the plaintiff taxpayer to make such an admission; instead, as Shalam instructed, the question is whether the plaintiff was presented with information sufficient to cause him to doubt the propriety of the tax shelter for tax avoidance purposes, but he willfully blinded himself to that information. 44 GT emphasizes the Shalam plaintiffs “admission” that he undertook the transaction solely for tax avoidance, based on a “more likely than not” opinion that he obtained for the purpose of defending against tax penalties. A fair reading of the New York and Virgin Island decisions does not support GT’s premise that the “more likely than not” tax opinion alone serves as a red flag negating a plaintiffs justifiable reliance on statements that a tax shelter strategy is proper. Although the Shalam court commented on the high risk that an ordinary person would understand to be associated with a “more likely than not” tax opinion, the court found that Shalam was presented with “information sufficient,” including an “artificial” loan transaction, to cause him to question whether the tax shelter was a proper tax avoidance scheme. 89 A.D.Sd at 158. The Shalam court noted the tax opinion’s first flaw which impacted the taxpayer’s justifiable reliance was the taxpayer knew it was not a true assessment of the tax shelter’s legality, but rather an opinion created to provide cover if the l.R.S. challenged the shelter. The actual analysis of the probabilities of the shelter succeeding in an l.R.S. proceeding was secondary. In Salt Aire, the tax opinion was likewise not singularly identified by the court but was discussed as part of the totality of “information sufficient” to have caused the plaintiff to question whether the tax shelter was a proper tax avoidance scheme. Further, in DDRA Capital, the Third Circuit considered the “more likely than not” tax opinion (from a law firm retained by KPMG) as a danger signal which was not sufficient by itself to negate justifiable reliance, 45 but which should have contributed to the taxpayer’s suspicion of the tax shelter’s legitimacy. While these fraud cases cited by GT suggest that a taxpayer’s reliance on a tax professional’s “more likely than not” tax opinion may play a role in establishing a lack of justifiable reliance, that determination is ultimately based upon the totality of the circumstances. As the New York courts stated, the issue is whether there was “sufficient information” to raise doubt about the tax avoidance scheme. Salt Aire, 2013 WL 775747, at *6. Needless to say, this entails a factual finding. These foreign cases also differ from the present case because the plaintiffs knew the transactions underlying the opinion letters were a sham. In Shalam and Salt Aire, the taxpayers understood that both the tax shelter scheme and the tax opinion itself were a sham. In DDRA Capital, although it does not appear the law firm’s tax opinion was a sham, the taxpayer was put on notice by the tax shelter strategy alone that it would not survive l.R.S. scrutiny. The tax opinion only served as one of several facts which cumulatively increased suspicion about the legality of the scheme. Here, the Yungs definitely were not part of the scheme to create an abusive tax shelter; they were approached by GT with the seemingly fullydeveloped Lev301. Also, they were never put on notice that the Lev301was not a proper tax avoidance strategy and they did not know the tax opinion was a sham. They reasonably believed that the shelter’s promoted tax treatment was more likely than not the proper tax treatment. The trial court specifically found that the Yungs were led to believe and did believe that GT’s opinions 46 were honest expressions of the Firm’s professional conclusions. Even when the Yungs later questioned the viability of the strategy because of information learned on their own, GT pressed forward knowing the “more likely than not” opinion did not exist. We cannot conclude that the trial court erred in its factual finding that the Yungs reasonably relied upon the “more likely than not” tax opinion since they believed it was being prepared in accordance with the expected standards of experienced tax practitioners. The trial court’s factual finding that the Yungs were not presented with sufficient information to cause them to question whether the tax shelter was a proper tax avoidance scheme is fully supported by the evidence. Unlike GT’s cited cases, here the trial court in its findings of fact enumerated multiple misrepresentations by GT that kept the Yungs from doubting the tax shelter’s legality and that induced the Yungs to complete the LevSOl transactions. While reliance on an opinion or prediction ordinarily is unreasonable as a matter of law, an opinion or prediction may be actionable if it falsely implies that the speaker has facts that form the basis of it or knows of no facts to the contrary. Restatement (Second) of Torts § 539 (1977); see Flegles, 289 S.W.Sd at 549. Liability then attaches to that false implication, not to the opinion or prediction in itself. Professionals who hold themselves out as having expertise in rendering opinions or predictions may be held liable for misrepresenting the true state of affairs when communicating such opinions or predictions, apart from any facts they might state or imply. See Flegles, 289 S.W.Sd at 549-51. 47 As the Yungs maintain, the issue here is not that GT’s “more likely than not” tax opinion turned out to be wrong through an error in professional judgment, but rather GT’s deliberate fraudulent misrepresentations and omissions concerning the legitimacy of its opinion, as reflected in the plethora of evidence that showed the Firm did not believe its own advice. The trial court properly found GT made multiple misrepresentations upon which the Yungs relied in both completing the LevSO 1 transaction and then continuing on the course charted by GT. We find no basis for overturning those fully-supported factual findings as a matter of law. As to the other facts which GT identifies to show that the Yungs could not have relied upon its advice — they knew the dividends were ordinarily taxable; they undertook the strategy for the specific purpose of avoiding taxes; and they obtained the opinion to allow them to argue against imposition of penalties for failing to report the dividend if the l.R.S. disallowed the strategy — we fail to see how this knowledge and conduct would put them on notice that the LevSOl was an improper tax avoidance scheme. Tax shelter participants in general expect the tax shelter promoter to be offering a product that satisfies the Internal Revenue Code and the l.R.S. that the product is not an economic sham. Under these circumstances, the Yungs are not to be faulted for attempting to take advantage of lawful methods to reduce their tax burden.39 39 Over and over again courts have said that there is nothing sinister in so arranging one’s affairs as to keep taxes as low as possible. Everybody does so, rich or poor; and all do right, for nobody owes any public duty to pay more than the law demands: taxes are enforced exactions, not 48 They participated in Lev301 as a purportedly legal means of bringing their Cayman corporations’ cash into the U.S. tax free and justifiably relied on their trusted tax professionals’ opinion, not realizing that the opinion was a facade to induce the Yungs to buy GT’s LevSO 1 product. In any event, all of the facts identified by GT were before the trial court and, as repeatedly noted, the eventual findings of fact emanating from the bench trial are supported by substantial evidence and are not clearly erroneous. 2. Justifiable Reliance: Advice from “Outside” Law Firm Additionally, GT claims the Yungs may not have reasonably relied upon GT’s tax opinion for another reason, i.e., advice the Yungs received from their own law firm. Due to what it deems an erroneous discovery ruling, GT insists it was unable to pursue whether the Yungs received independent professional tax advice which informed them that the LevSO 1 was not a strategy which supported a “more likely than not” opinion. Citing this erroneous denial of relevant discovery, GT argues the judgment must be vacated, and this case remanded. We disagree. The record reflects that the Yungs consulted an outside tax firm, Katz, Teller, Brant & Hild, regarding GT’s February 2001 draft long-form Opinion Letter. The Katz Teller review of the Opinion Letter was provided to the Yungs voluntary contributions. To demand more in the name of morals is mere cant. Comm’r v. Newman, 159 F.2d 848, 850-51 (2d Cir. 1947) (Learned Hand, J., dissenting), cert, denied, 331 U.S. 859 (1947); quoted with approval, U.S. v. Thompson./Center Arms Co., 504 U.S. 505, 511 n.4 (1992). 49 after the Yungs had already engaged in the Lev3Ol transaction. Katz Teller advised the Yungs that due to GT’s reputation and the Yungs’ long-standing relationship with GT, it was reasonable to rely on advice given to them by GT. Later, the Yungs informed the l.R.S. of the Katz Teller advice as support for its “reasonable cause” defense to penalties when the l.R.S. was investigating the Yungs’ Lev3Ol transaction. During discovery, the two Katz Teller attorneys who advised the Yungs were deposed. Although together they provided over 400 pages of deposition testimony, they refused to answer certain questions that went beyond the information the Yungs had provided the l.R.S., asserting the attorney-client privilege.40 GT moved to compel the attorneys to answer all questions relating to the subject matter of the advice given by the Katz Teller law firm to the Yungs regarding the GT opinion. The trial court denied that motion and, at trial, GT did not call the Katz Teller witnesses or introduce their depositions. In GT’s motion seeking post-judgment relief, it raised the issue again, but the trial court reiterated the prior ruling stating: [the Yungs] told the l.R.S. they had been advised by Katz Teller that the Grant Thornton opinion would protect them from penalties. The information provided to the l.R.S. was also provided to Grant Thornton. This advice was merely that the Yungs could use the fact they had an opinion from Grant Thornton to demonstrate their exercise of entering into the transaction and supporting their defense that they had acted with reasonable cause 40 Katz Teller attorneys Andrew Berger and William Russo testified that they performed no independent research or assessment of the GT Opinion Letter’s substance. Indeed, neither attorney had ever given a “second opinion” on tax treatment of a proposed transaction for any client. Katz Teller’s role was explaining the transaction to the Yungs, reviewing documents for legal form and evaluating the weight the l.R.S. would give a tax opinion from GT. 50 and in good faith; it was not advice as to the substance of the Grant Thornton opinion or a second opinion as to the merits of the Lev301. (Emphasis supplied.) The Court of Appeals affirmed. With the Yungs’justifiable reliance on GT’s opinion being at issue, GT claims that the existence of any other advice about the LevSO 1 tax and penalty risks on which the Yungs relied is relevant. GT contends the Yungs waived the attorney-client privilege both by putting their knowledge of the risks of the LevSO 1 transaction at issue (resulting in an implied waiver referred to as an “at issue” or “offensive use” waiver) and by voluntarily disclosing the Katz Teller advice to the I.R.S. and other third parties. In 3M Co. V. Engle, we stated: The client may waive the [attorney-client] privilege by taking positions that place the substance of the communications in issue. ... A position that seems often to bring implied waiver into play is [the] clients’ claim that they acted or refrained from acting on advice of counsel. . . . With this and other similar positions, the inquiry for the trial court “is whether allowing the privilege to protect against disclosure of the information would be manifestly unfair to the opposing party.” 328 S.W.3d 184, 188-89 (Ky. 2010) (citing Robert G. Lawson, The Kentucky Evidence Law Handbook, § 5.05 [10], at 363-64 (4th ed. 2003 & 2010 Supp.) (quoting Home Indem. Co. v. Lane Powell Moss & Miller, 43 F.3d 1322, 1326 (9th Cir. 1995))). Home Indemnity Co. relies on the standard set out in Hearn v. Rhay, 68 F.R.D. 574, 581 (E.D. Wash. 1975), for determining when an implied waiver of the attorney-client privilege occurs. Under Hearn, an implied waiver occurs when: 51 (1) assertion of the privilege was a result of some affirmative act, such as filing suit, by the asserting party; (2) through this affirmative act, the asserting party put the protected information at issue by making it relevant to the case; and (3) application of the privilege would have denied the opposing party access to information vital to his defense. Id. GT cites Christenbury v. Locke Lord Bissell & Liddell, LLP, 285 F.R.D. 675, 679 (N.D. Ga. 2012) in support of its position that the Yungs waived their attorney-client privilege with Katz Teller. In Christenbury, the plaintiffs alleged the Locke Lord law firm negligently supplied false information in a tax opinion. Locke Lord argued that the plaintiffs’ lawsuit put at issue whether the plaintiffs were receiving and relying on independent advice from Lustig, another attorney, when deciding to participate in the transaction. Locke Lord sought attorney-client privileged information from the plaintiffs’ contact with Lustig. After acknowledging that the Hearn v. Rhay standard is “widely seen as the majority view,” id. at 682, the Christenbury court decided, in a case of first impression under Georgia law, that the implied waiver may extend to a nonparty’s legal advice. However, two crucial factors distinguish Christenbury. First, the plaintiffs had sued Lustig in Texas for legal malpractice regarding the same transactions. The Christenbury court stated: Plaintiffs can accuse more than one attorney of malpractice. But in doing so here they put at issue whether and to what extent each attorney's advice actually caused their loss and/or are responsible for a share of damages. Plaintiffs' affirmative act placing the information at issue satisfies the first and second prongs of Hearn. As for the third prong, the 52 Court also finds that applying the privilege would deny the Defendants vital discovery. Defendants' entire theory of causation and comparative negligence turns on Plaintiffs' communications with the other professionals Plaintiffs were consulting. Id. at 684. Here, the Yungs have not sued Katz Teller and placed at issue the advice received (which as noted was not a second opinion) and comparative fault is not a consideration. Second, rather than disfavoring waiver of the attorney-client privilege, Georgia law “confine[s] the attorney-client privilege to its narrowest permissible limit.” Id. at 683. Kentucky has no such rule. We are not persuaded that the facts of this case warrant an extension of the implied waiver to a non-party’s legal advice. An implicit waiver of the attorney-client privilege is not triggered by whether or not the communications are relevant to the issue asserted. United States v. Zollin, 491 U.S. 554, 562-63, 109 S.Ct. 2619, 105 L.Ed.2d 469 (1989), for the implicit waiver rule to become applicable, a party must affirmatively use privileged communications to defend itself or attack its opponent in the action. Dawson v. New York Life Insurance Co., 901 F. Supp. 1362 (N.D. 111. 1995). Hodak V. Madison Capital Mgmt., LLC, 2008 WL 2355798, at *3 (E.D. Ky. June 5, 2008); accord Transamerica Life Ins. Co. v. Moore, 274 F.R.D. 602, 607 (E.D. Ky. 2011) (no implicit waiver where plaintiff has not used that attorney’s advice to defend in the litigation or to attack defendants); State Farm Fire 85 Cas. Co. v. Griggs, 419 P.3d 572, 575 (Colo. 2018) (implied waiver not applicable where plaintiff has not asserted a claim or defense that depends on privileged communications with his attorney); Robert W. Baird & Co. Inc. v. Whitten, 418 P.3d 894, 900-01 (Ariz. App. 2017) (underwriters’professional negligence action against bond counsel did not cause underwriters to assert privilege with 53 other legal counsel, did not put that other counsel’s advice at issue and did not deny bond counsel access to vital information). The main issue here is whether the Yungs justifiably relied on GT in making their decision to utilize the Lev3Ol strategy. The Yungs did not affirmatively use privileged communications with Katz Teller to defend themselves or to attack GT in this action. Perhaps most significantly, as the trial court found, Katz Teller did not even opine on the validity of GT’s opinion, i.e., it did not address the substantive issue but simply opined that reliance on advice from GT, a national accounting firm, would protect the Yungs in the event of an I.R.S. audit. As the Yungs point out, the I.R.S. rejected their advice from Katz Teller as a good faith effort precisely because the I.R.S. found the Katz Teller advice did not constitute a second opinion regarding the Lev301, GT’s opinion, or the tax aspects of the transaction. GT asserts, alternatively, that the Yungs waived the entire attorney-client privilege by expressly disclosing part of Katz Teller’s advice to the I.R.S. A client “waives the privilege if he . . . voluntarily discloses or consents to disclosure of any significant part of the privilege matter.” Kentucky Rule of Evidence (KRE) 509. The “partial waiver” doctrine, however, “permits a client who has disclosed a portion of privileged communication to continue asserting the privilege as to the remaining portions of the same communications.” Jones V. Illinois Cent. R. Co., 617 F.3d 843, 856 (6th Cir. 2010) (citing In re Columbia/HCA Healthcare Corp. Billing Practices Litig. v. Columbia/HCA Healthcare Corp., 293 F.3d 289 (6th Cir. 2002) (internal quotation marks 54 omitted)). We agree with the trial court that the partial waiver doctrine applies to the Yungs’ limited disclosure to the I.R.S. that Katz Teller had advised the Yungs that an opinion from a reputable firm like GT could be relied upon for penalty protection. Accordingly, we affirm the Court Appeals’ conclusion that the trial court did not err in denying discovery as to other attorney-client communications between the Yungs and Katz Teller. B. Compensatory Damages 1. Taxes and Interest as Recoverable Damages The trial court concluded that but for the Yungs’ participation in the Lev3Ol transactions, they would not have paid the taxes and interest in connection with the I.R.S. settlement agreement. The trial court accordingly awarded the Yungs $11,837,860 in taxes; $5,021,494 in interest; and $1,555,873 in penalties, the full amounts the Yungs paid to the I.R.S. The Court of Appeals affirmed the trial court’s decision. GT argues that, as a matter of law, taxes and I.R.S. interest are not recoverable as compensatory damages. Conversely, the Yungs claim that tax liabilities may be recovered when they flow from fraudulent and negligent conduct. This is an issue of first impression in Kentucky. Jurisdictions addressing whether taxes and I.R.S. interest are recoverable as damages have taken different approaches. See, e.g., O’Bryan v. Ashland, 17 N.W.3d 632, 636-37 (S.D. 2006). The approach GT advocates holds that taxes, interest, and penalties are not recoverable as consequential damages because tax liabilities on income stem from Internal Revenue Code 55 requirements and not from an aecountant’s wrongful conduct.41 See DCD Programs, Ltd. v. Leighton, 90 F.Sd 1442, 1447, 1449 (9th Cir. 1996). Aligned with this view, GT primarily directs this Court to Alpert v. Shea Gould Climenko & Casey, 160 A.D.2d 67 (N.Y. App. Div. 1990), a tax shelter case. In Alpert, the court found back taxes and interest were not recoverable as damages. That New York court reasoned that tax recovery would place the plaintiffs in a better position than if they never produced taxable income,42 id. at 71-72, and that interest was not damages as it was money owed to the government by the defrauded taxpayer for the use of money during a period of time he was not entitled to it, id. at 72. As a matter of equity, the Alpert court held the taxpayer should not have the windfall benefit resulting from both having the tax monies and also recovering the interest on those funds. Id. at 72. Other courts have found that taxes are recoverable damages as a liability caused by the accountant’s conduct. See, e.g., Eckert Cold Storage, Inc. v. Behl, 943 F. Supp. 1230, 1234 (E.D. Cal. 1996) ([PJlaintiffs may be entitled to damages due to the tax liability, if they can prove that this liability was caused As GT phrases it, “taxes paid on income received by the plaintiff are not 41 recoverable damages because tax liability results from the Tax Code’s directive that income is taxable, not an accountant’s failure to assist the taxpayer in evading that tax liability.” Of course, the trial court found that the Yungs would not have had this particular tax liability but for GT’s soliciting their business and selling them on the Lev301 strategy. 42 Significantly Alpert, like the plaintiffs in Shalam and Salt Aire, was facing a substantial income tax liability as a result of unrelated matters and he was looking for a tax shelter to offset those liabilities on which he would otherwise owe taxes. The Yungs were not in that position. Their particular tax liability arose because GT sold them on the Lev301. 56 by [the accountant’s] negligent or fraudulent advice and would not otherwise have been incurred.). Courts using this approach disfavor a blanket prohibition of recovery of taxes and interest as being against the overriding principle of placing the plaintiff in the position the plaintiff would have occupied “but for” the actions of the tortfeasor. These jurisdictions leave it to the fact-finder to determine whether under the circumstances of the case a plaintiff has been damaged. O’Bryan, 717 N.W.2d at 637-38. Illustrative of this approach, the Yungs cite Maese v. Garrett, 329 P.3d 713 (N.M. App. 2014), which recognizes that although a plaintiff cannot recover an otherwise valid tax liability as damages, a tax liability which the plaintiff would not have incurred but for the defendant’s negligent advice or wrongful conduct is recoverable. The Yungs assert that as the substantial evidence showed, and the trial court expressly found, but for GT’s fraud and gross negligence, they would have enjoyed the use of their monies overseas without any tax liability. Consequently, the award of taxes, interest, and penalties placed the Yungs in the position they would have occupied if they had not been fraudulently induced by GT to enter the Lev301 transaction. GT, not surprisingly, counters that awarding all taxes and interest owed under the settlement with the I.R.S. puts the Yungs in a better position than if they had never undertaken the transaction.43 GT notes that although the 43 GT states that if justifiable reliance is established and liability exists, the compensatory damages should be limited to the $1.5 million in I.R.S. penalties paid by the Yungs. 57 Yungs may not have distributed the money but for GT’s conduct, they also would not have had a $30 million tax-free dividend available for investment in the U.S. As compensatory damages are meant to make an injured party whole and nothing more, GT takes the position that tax recovery puts the taxpayer “in a better position than he would have been had the wrong not been done.” Ky. Cent. Ins. Co. v. Schneider, 15 S.W.3d 373, 374-75 (Ky. 2000). Noting that interest is a taxpayer’s payment to the I.R.S. for use of the money during the period of time when the taxpayer was not entitled to it, GT argues further that by allowing the Yungs to recover interest as damages, the Yungs are receiving tax-free, interest-free income, despite a statutory obligation to pay taxes. With a tax-free $30 million dividend on top of the profit gained from using the money, GT insists the Yungs are not back in the position prior to the Lev301 — a position where the Yungs could not invest the Cayman corporations’ funds in the U.S. or use the funds personally. Initially, we are compelled to observe that returning the Yungs to the exact position they occupied prior to their involvement with the Lev301 is simply not possible. The trial court’s award of all taxes, interest and penalties will not erase the fact of the I.R.S. audit and its business repercussions for the Yungs nor will it undo these long years of litigation. As for the $30 million from the Cayman corporations, it was brought to the U.S., for good or ill, and will not be returned there. GT characterizes this outcome as the trial court placing the Yungs in a better position but, in fact, it is simply a slightly different position — a position that flowed directly from business transactions 58 the Yungs would not have undertaken but for GT’s conduct — and the position that most closely restores the Yungs to the position they occupied pre-Lev301. As noted, the recovery of taxes and l.R.S. interest has not been specifically addressed by an appellate court in Kentucky, so we begin with generally applicable tort damage recovery rules. For recovery in Kentucky under a theory of fraud or negligence, the plaintiff must establish that the defendant proximately caused the damages. See Ohio Cas. Ins. Co. v. Com., Dept. of Highways, 479 S.W.2d 603, 605 (Ky. 1972); Flegles, 289 S.W.3d at 553. As recently described in Patton v. Bickford, 529 S.W.3d 717, 730 (Ky. 2016), causation consists of two distinct components: “but-for” causation, also referred to as causation in fact, and proximate causation.44 Thus, a plaintiff must show that the harm would not have occurred but for the defendant’s conduct, and the plaintiff must also show that the defendant’s negligence (or fraudulent conduct) was legally sufficient to result in liability. Causation in fact is established if the occurrence would not have happened “but for” the conduct of the defendant. “But-for causation requires the existence of a direct, distinct, and identifiable nexus between the defendant’s breach of duty (negligence) and the plaintiffs damages such that the event would not have 44 Although not employing the “but-for” terminology, several Kentucky cases also describe the causation in fact component as an aspect of proximate cause. See, e.g., Gerebenics u. Gaillard, 338 S.W.2d 216, 219 (Ky. 1960) (“To constitute proximate cause, an act must be such that it induced the accident and without which the accident would not have occurred.”) (citing Mahan v. Able, 251 S.W.2d 994 (Ky. 1952)); Morris V. Combs’ A dm’r., 200 S.W.2d 281, 283 (Ky. 1947). 59 occurred ‘but for’ the defendant's negligent or wrongful conduct in breach of a duty.” Id. The measure of damages for these causes of action is generally limited to “actual pecuniary compensation for the injury sustained,” or “out-of-pocket” damages as the plaintiff “is not to be placed in a better position than he would have been in had the wrong not been done.” See Western Union Telegraph Co. V. Guard, 139 S.W.2d 722, 727 (Ky. 1940); Sanford Const. Co. v. S & H Contractors, Inc., 443 S.W.2d 227, 239 (Ky. 1969). As stated in Sanford, 443 S.W.2d at 239 (quoting 37 Am. Jur. 2d 458, Fraud and Deceit, § 342): The measure of damages for fraud is, as a general rule, the actual pecuniary loss sustained. Thus, as a general rule, one injured by the commission of fraud is entitled to recover such damages in a tort action as will compensate him for the loss or injury actually sustained and place him in the same position that he would have occupied had he not been defrauded. Given Kentucky’s precedent, we join those jurisdictions that deem the issue of whether the plaintiff has actually been damaged by incurring tax and interest liability a question of fact. Consequently, we refuse to adopt the blanket “matter of law” rule advocated by GT forbidding tax and I.R.S. interest recovery in accounting fraud and negligence actions. Under our traditional tort damage principles, if the taxpayer has been injured, recovery should be allowed if the taxpayer meets the burden of proving causation and damages. Therefore, if the tax liability [i.e., taxes and interest paid to the I.R.S.) is a direct result of an accountant’s fraudulent or negligent conduct, a plaintiffs out-of-pocket damages are recoverable. Ultimately, the issue depends on the circumstances 60 of each case, and in this case, we conclude the trial court properly awarded taxes and interest as compensatory damages. 2. The Trial Court’s “But-For” Finding GT also argues that the damage award for taxes should be vacated because the trial court clearly erred in its factual finding that the Yungs would not have distributed the Cayman corporations’ funds and consequently would not have incurred income tax and interest but for GT’s advice. GT points to evidence that seemingly contradicts William Yung’s testimony that the Yungs “never” received distributions from the overseas corporations unless the distributions could be made without incurring tax liability. GT notes that in 2000, Yung caused the Cayman corporations to pay almost $15 million to shareholders in addition to the $30 million Lev301 distribution. Further, Yung told the l.R.S. during the audit that he expected to draw on the equity and credit of all his companies in order to finance his acquisition of Lodgian, Inc. Challenging the trial court’s factual finding, GT insists these facts show the Yungs needed more than $30 million in 2000. The trial court rejected this specific argument in a post-judgment CR 59.05 motion. The trial court found that the testimony was not that the Yungs never made any distributions that were taxable, but they would not have transferred the $30 million into the U.S. at that time without being convinced by GT that the Lev301 would legitimately allow the transfer tax-free. The evidence as a whole further supported a finding that the Yungs generally chose strategies which minimized their tax liability when such legal methods could be 61 found. As for the Lodgian, Inc., acquisition, the trial court found it was not a motivation for the Lev3Ol transaction. GT counters that even if that were true, the Yungs would have paid taxes on retained earnings “whenever they were dividended (sic),” which defeats the finding that GT’s conduct was the “but-for” cause of paying taxes on any distribution. The trial court’s “[f]indings of fact, shall not be set aside unless clearly erroneous, and due regard shall be given to the opportunity of the trial court to judge the credibility of the witnesses.” CR 52.01. When evidence is conflicting, this Court cannot substitute its judgment unless the findings of fact are manifestly against the weight of evidence. Wells v. Wells, 412 S.W.2d 568, 571 (Ky. 1967) (internal quotation and citations omitted). Here, the trial court’s findings were not manifestly against the evidence. Ample evidence supported the findings that (1) the Yungs would not have distributed the $30 million but for GT’s promotion of the Lev301; (2) earlier distributions of Cayman corporation funds were effected when “tax efficient” and; (3) the Lodgian, Inc., acquisition did not motivate the Lev301 transaction. Finally, GT’s retained earnings distribution argument presumes the Cayman corporation funds would eventually be brought back to the United States, but that eventuality was not inevitable as the Yungs could have used the funds overseas; alternatively, the funds could have been distributed at a different time when the Yungs’ overall tax situation would have produced more favorable treatment. Having considered the facts GT cites in support of different factual 62 findings, we cannot conclude that the trial court’s actual findings of fact were clearly erroneous. 3. Mitigation of Damages: Offsetting Benefits GT lastly contends that the Yungs’ award for taxes ($11,837,860) and interest ($5,021,494) should be offset by the profits (between $18-21 million by 2006) they realized by using the distributed funds from 2001 until settlement with the I.R.S. Otherwise, the Yungs have received a pure windfall by having in hand the $30 million tax-free distribution and the profits from its use. More specifically, GT asserts, citing Streber v. Hunter, 221 F.3d 701, 734 (5th Cir. 2000), that the Yungs’ interest recovery should be vacated because the Yungs did not prove that the interest paid to the I.R.S. exceeded additional income earned from investing the dividends. In Streber, the court, taking a middle ground approach as to whether interest is a recoverable damage, upheld the “interest differential,” the difference between the interest earned by the taxpayer from the $1.7 million while she had control of it unlawfully and the interest charged by the I.R.S. for such possession, as a recoverable form of damages. Id. at 734. GT espouses the rule expressed in Restatement Second of Torts § 920 (1979): When the defendant’s tortious conduct has caused harm to the plaintiff or to his property and in so doing has conferred a special benefit to the interest of the plaintiff that was harmed, the value of the benefit conferred is considered in mitigation of damages, to the extent that this is equitable. 63 “Damages resulting from an invasion of one interest are not diminished by showing that another interest has been benefited” under this rule. Id., cmt. b. The Yungs counter GT’s argument by noting the record contains no evidence that the Yungs would have earned a lesser return had they kept the money overseas. As the Yungs argue, the appropriate comparison for a windfall determination is between the return on the use of the distribution in the U.S. and the return on the money had it been kept overseas, such that a benefit may have been shown if the Yungs earned a greater return through U.S. investments. We agree. To the extent that profit from use of the funds in the U.S., as compared to the l.R.S. interest charge, is the properly considered interest harmed and benefitted, id., GT had the burden to show the mitigation of damages by profits earned. Without evidence the Yungs would have earned a lesser return had they kept the money overseas, GT’s profit windfall argument fails. IL PUNITIVE DAMAGES A. Remittitur of an Excessive Punitive Damage Award is Appropriate Before addressing the punitive damage award, we take a moment to address Kentucky appellate courts’ authority to reduce a punitive damage award. The Court of Appeals noted in its opinion that this Court has never explicitly endorsed its authority to order remittitur of punitive damages, although that court has used such authority in Ragland v. DiGiuro, 352 S.W.3d 908 (Ky. App. 2010) and McDonald’s Corp. v. Ogbom, 309 S.W.3d 274 (Ky. App. 64 2009). We take this opportunity to clarify that remittitur of a punitive damage award is proper in a case where the facts justify it. Remittitur is the process used to reduce a damage award when an error is discovered. See Black’s Law Dictionary (10th ed. 2014). It may involve “an order awarding a new trial, or a damages amount lower than that awarded by the juiy and requiring the plaintiff to choose between those alternatives.” Id. Although this doctrine has now been accepted in Kentucky, our predecessor court observed that remittitur is generally prohibited. For example, in Chesapeake & O. Ry. Co. v. Meyers, 151 S.W. 19, 21 (Ky. 1912), the Court explained: [Tjrial courts have no power to remit any portion of a judgment, ... if the judgment is excessive, they must award the defendant a new trial. While this is the general rule [Brown v. Morris, 3 Bush, 81-, L. & N. R. R. Co. V. Earl’s Adm’x, 94 Ky. 370, 22 S. W. 607, 15 Ky. Law Rep. 184), yet where the items constituting the damages recovered are separable, so that the court may eliminate those not properly recoverable from those recoverable, the court has power to require the plaintiff to remit, or may, on plaintiffs motion, remit, so much of the damages as represents the items which are not properly recoverable [Johnson’s Adm’r v. Johnson, 104 Ky. 714, 47 S. W. 883, 20 Ky. Law Rep. 890; Masterson v. Hagan, 17 B. Mon. 325). Louisville & Nashville Railroad Co. v. Complete Auto Transit explained the limits placed on Kentucky courts’ power to correct an excessive award by remittitur likewise: “It is only where the items constituting the damages recovered are separable so that the court may eliminate those not properly recoverable from those which are recoverable that a remittitur may be ordered.” (citations omitted). The facts in that case serve as an example of when 65 remittitur is improper. The jury instructions properly allowed three items of compensatory damage, but the instruction defining the damage limit was in error. The jury awarded an excessive recovery, but because there was no separate itemization for the damages improperly recovered, remittitur was not proper. Id. Columbia Amusement Co. v. Rye, 155 S.W.2d 727 (Ky. 1941), is an example of the Court using remittitur authority to reduce the judgment rather than ordering a new trial. The Court acknowledged that “a successful litigant [will not often be compelled] to remit a portion of an excessive verdict for compensatory damages as an alternative to a reversal,” but observed that the only ground meriting reversal was the erroneous inclusion of a special damage item in the jury instruction. In accordance with other precedent “eliminat[ing] from judgments improper items of recovery where the exact amount thereof was ascertainable from their language or by calculation,” the Columbia Amusement court reduced the judgment, stating: [W]e must either subject the litigants and the Commonwealth to the expense of a new trial because of the commission of a comparatively minor error, or protect the rights of the appellant and conserve the true interests of the appellee by directing that the amount of the item be deducted from the total of the judgment. Justice delayed is too often justice denied, and we are convinced that every practical consideration demands our adoption of the latter course. Id. at 731. Of course, the foregoing cases did not address remittitur of punitive damages. When deciding in Hanson v. American National Bank & Trust Co., 66 865 S.W.2d at 310, whether the Court of Appeals properly reversed a punitive damage award and remanded the case to the trial court for either remittitur of its excess or the grant of a new trial on the award, this Court stated; “[We are] unaware of any authority in this jurisdiction for court-ordered remittitur of punitive damages or for the fixing of the amount of such damages, and it will not here establish such precedent.” Hanson, a lender liability action, was reconsidered on remand from the U.S. Supreme Court in light of TXO Production Corp. v. Alliance Resources Corp., 509 U.S. 443 (1993), and its outcome remained unchanged. This Court affirmed the punitive damage award of $5,775,000 as not unconstitutionally excessive, as well as compensatory damages of $1,065,000. In consideration of the duty to perform a de novo constitutional review as required by Cooper Industries, Inc. v. Leatherman Tool Group, Inc., 532 U.S. 424, 436 (2001), this Court concluded in Sand Hill Energy, Inc. v. Ford Motor Co., 83 S.W.3d 483, 496 (Ky. 2002), that $5 million of a $20 million punitive damage award should be set aside. But, following a grant of certiorari, the United States Supreme Court vacated and remanded Sand Hill for reconsideration in light of State Farm Mutual Automobile Insurance Co. v. Campbell, 538 U.S. 408 (2003). Ford Motor Co. v. Smith, 538 U.S. 1028 (2003). This Court then vacated the punitive damage award and remanded the case to the trial court for a new punitive damage award because the trial court’s jury instructions failed to include a limiting instruction concerning extraterritorial punishment. Sand Hill Energy, Inc. v. Smith, 142 S.W.3d 153, 156 (Ky. 2004). 67 Since that time, our Court of Appeals has reduced punitive damage awards in Ragland, 352 S.W.3d at 924 (reducing $60 million punitive damage award to $30 million), and McDonald’s, 309 S.W.3d at 301 (reducing $1 million punitive damage award to $400,000). In this case, the Court of Appeals correctly noted that it is required to perform a de novo review when considering whether a punitive damage award satisfies the due process requirements elaborated upon in State Farm, 538 U.S. at 408, and BMW of North America, Inc. v. Gore, 517 U.S. 559 (1996). With that constitutional review being squarely within the appellate court’s authority, it necessarily followed that the court has the authority and responsibility to reduce a constitutionally excessive punitive damage award. We agree. The Due Process Clause of the Fourteenth Amendment constrains a state court award of punitive damages, see State Farm, 538 U.S. at 416-18; BMW, 517 U.S. at 568, and, consequently, the courts have the responsibility for determining whether a particular damage award has exceeded constitutional limits. If it has, remittitur is appropriate. We thus turn to whether the trial court made a constitutionally excessive punitive damage award, and if so, whether the Court of Appeals’ reduction to a 1:1 punitive/compensatory damage award was proper. B. The Total Punitive Damage Award to Total Compensatory Damage Award Comparison Is Proper As an initial matter, when analyzing the constitutionality of the damage award, the trial court considered the ratio of the total punitive damage award 68 to the total compensatory damage award rather than the ratios for the respective awards to William and Martha Yung (11.7:1; $55 million to $4.68 million) and the Family Trust (1.7:1; $25 million to $14.6 million). Citing McDonald’s, 309 S.W.3d at 299-300, and Planned Parenthood of Columbia/Willamette Inc. v. American Coalition of Life Activists, 422 F.3d 949, 961 (9th Cir. 2005) ,45 GT argues that each plaintiffs award should be evaluated individually because the due process analysis involves not only GT’s conduct, but also the vulnerability of and degree of harm to each plaintiff. As the Yungs point out, precedent does not compel an individualized analysis for each plaintiff. For example, when the United States Supreme Court considered State Farm, 538 U.S. 408, the $145 million punitive damage award was compared to the two plaintiffs’ $1 million total compensatory damage award (Mr. and Mrs. Campbell were awarded $600,000 and $400,000, respectively). Id. at 425. We agree with the Court of Appeals that although the Yungs and the Family Trust had legally distinct interests and damages, GT owed the same The Planned Parenthood court stated: 45 [I]n a multi-plaintiff, multi-defendant action, an approach that compares each plaintiff s individual compensatory damages with the punitive damages awards against each defendant more accurately reflects the true relationship between the harm for which a particular defendant is responsible, and the punitive damages assessed against that defendant. This approach is preferable to that urged by physicians and adopted by the district court for several reasons. Merging the physicians’ damages against a particular defendant as the district court did, rather than considering them on a plaintiff-by-plaintiff, defendant-by-defendant basis, has the distorting effect of making some ratios appear closer to a constitutional level than they truly are, while making others appear further from it than they really are. 69 duties to them and GT’s misconduct is not really distinguishable across the Yungs and the Family Trust. The fraud and gross negligence was perpetrated simultaneously against all of them as part of an integrated scheme. Consequently, under the circumstances, it is proper to compare the total punitive damages awarded to the total compensatory damages. C. On De Novo Review, the Trial Court’s Award is Sustainable The trial court viewed GT’s conduct as egregious and determined that a punitive damage award four times that of the compensatory damage award was needed to punish and deter like future behavior. Finding the cases GT cited from the United States Court of Appeals for the Sixth Circuit to have similar factual circumstances, our Court of Appeals agreed that punitive damages were warranted but concluded that punishment and deterrence would be satisfied with a punitive damage award equal to the compensatory damage award, i.e., a 1:1 ratio. The Yungs counter that the trial court’s award was well within constitutional limits and in light of the trial court’s findings as to GT’s reprehensible behavior the reduction was not warranted simply because the compensatory damage award was substantial. GT contests the trial court’s reprehensibility finding and emphasizes the very substantial, approximately $20 million compensatory damage award. As often noted, punitive damages exist to punish and discourage certain types of bad behavior. KRS 411.184; Hensley v. Paul Miller Ford, Inc., 508 S.W.2d 759, 762 (Ky. 1974). A jury uses a punitive damage award to punish a defendant, deter future wrongdoing, and express its moral condemnation. 70 Cooper Industries, 532 U.S. at 432; Hensley, 508 S.W.2d at 762. “The precise award in any case, of course, must be based upon the facts and circumstances of the defendant’s conduct and the harm to the plaintiff.” State Farm, 538 U.S. at 425. Simply put, the amount of an award should embody the fact-finder’s determination as to the degree of reprehensibility reflected in the defendant’s actions. Pursuant to KRS 411.184(2), punitive damages are available if a plaintiff proves by clear and convincing evidence that a defendant acted with fraud, oppression, or malice. Punitive damages are also available if gross negligence is shown. See Williams v. Wilson, 972 S.W.2d 260, 262-65 (Ky. 1998). Statutory guidelines for assessing the culpability of the defendant and the amount needed to deter or to make such behavior unprofitable are: (a) The likelihood at the relevant time that serious harm would arise from the defendant’s misconduct; (b) The degree of the defendant’s awareness of that likelihood; (c) The profitability of the misconduct to the defendant; (d) The duration of the misconduct and any concealment of it by the defendant; and (e) Any actions by the defendant to remedy the misconduct once it became known to the defendant. KRS 411.186(2); Sand Hill Energy, Inc. v. Smith, 142 S.W.3d at 167. The Due Process Clause prohibits imposing “grossly excessive” punishment on a defendant. State Farm, 538 U.S. at 416. Excessive punitive damage awards offend the Constitution because “[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. 71 but also of the severity of the penalty that a State may impose.” Id. (quoting BMW, 517 U.S. at 574). “To the extent an award is grossly excessive, it furthers no legitimate purpose and constitutes an arbitrary deprivation of property.” Id. (citing Pac. Mut. Life Ins. Co. v. Haslip, 499 U.S. 1, 54, (O’Connor, J., dissenting)). When exemplary damages are warranted, they should reflect — but not be grossly out of proportion with — the enormity of the offense. BMW, 517 U.S. at 575-76 (citations omitted). The touchstone for determining whether a punitive damage award is constitutional is whether the award is reasonable based upon the facts of the case. TXO, 509 U.S. at 458. Due process requires appellate courts to perform a de novo review of the constitutionality of punitive damage awards. See Sand Hill, 83 S.W.Sd at 493 (citing Cooper Indus., 532 U.S. at 424); St Joseph Healthcare, Inc. v. Thomas, 487 S.W.Sd 864, 878 (Ky. 2016). To determine whether an award is grossly excessive or unreasonable, the U.S. Supreme Court has instructed courts to look at three guideposts: (1) the degree of reprehensibility of the defendant’s conduct, (2) the disparity or the ratio between the harm or potential harm suffered by the plaintiffs and the punitive damage award, and (3) the difference between the punitive damage award and the penalties imposed for similar misconduct. BMW, 517 U.S. at 559. Application of these guideposts ensures “that an award of punitive damages is based upon an ‘application of law, rather than a decisionmaker’s caprice.”’ State Farm, 538 U.S. at 418 (quoting Cooper Indus., 532 U.S. at 436). However, “(sjtates necessarily have considerable flexibility in determining 72 the level of punitive damages that they will allow ... in any particular case.” BMW, 517 U.S. at 568. 1. Reprehensibility “[T]the most important indicium of the reasonableness of a punitive damages award is the degree of reprehensibility of the defendant’s conduct.” BMW, 517 U.S. at 575. Reprehensibility is assessed by considering whether: 1) the harm caused was physical as opposed to economic; 2) the tortious conduct evinced an indifference to or a reckless disregard of the health or safety of others; 3) the target of the conduct had financial vulnerability; 4) the conduct involved repeated actions or was an isolated incident; and 5) the harm was the result of intentional malice, trickery, or deceit, or mere accident. State Farm, 538 U.S. at 419 (citing BMW, 517 U.S. at 576-77). The calculus for reprehensibility considers, but does not require, all five factors: 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. It should be presumed a plaintiff has been made whole for his injuries by compensatory damages, so punitive damages should only be awarded if the defendant's culpability, after having paid compensatory damages, is so reprehensible as to warrant the imposition of further sanctions to achieve punishment or deterrence. Id. With this approach, the factual findings that demonstrate reprehensibility take on special importance. Although our review of a lower court’s due process “grossly excessive” analysis is de novo, we defer to the trial court’s findings of fact unless they are clearly erroneous. See Cooper Indus., 532 U.S. at 440, n.l4 (“While we have determined that the Court of Appeals 73 must review the District Court's application of the [BMIV] test de novo, it of course remains true that the Court of Appeals should defer to the District Court's findings of fact unless they are clearly erroneous.”). Here, GT’s conduct was sufficiently reprehensible to warrant punitive damages. None of the first three BMW factors is present in this case. The harm was economic only, the health or safety of others was not in danger, and the Yungs do not argue they were financially vulnerable. The primary considerations left to be addressed are where GT’s “repeated actions” and “intentional deceit” place them on the reprehensibility spectrum. State Farm, 538 U.S. at 419. The trial court’s findings of fact provide a timeline from when GT first approached the Yungs to consider Lev301 as a strategy to transfer Cayman cash into the U.S. without tax liability to the point GT and the Yungs settled with the l.R.S. Knowing the Yungs’ aversion to tax risks personally and professionally, GT made false representations and omitted material facts imperative to the Yungs’ fully-informed decisionmaking as to Lev301. This includes not only at the initial implementation of the Lev301 transaction but also at subsequent points when no consideration was given to unwinding the transaction, and later when GT addressed the l.R.S.’s scrutiny of the transaction and the Yungs’ tax returns. Knowing from the outset that the Yungs did not want to be GT’s “guinea pigs,” GT falsely represented that other corporations had successfully relied on the strategy. While communicating to the Yungs that GT would reach and did reach a “more likely than not” opinion 74 that the Lev301 as structured for the Yungs would withstand l.R.S. scrutiny, internal Firm communications disclosed otherwise. Over the course of time, despite multiple l.R.S. notices and regulations, professional articles, opinions of outside legal counsel, and internal confusion alerting GT that the LevSOl was likely an abusive tax shelter and l.R.S. regulations likely would apply retroactively to the Yungs’ detriment, GT never once disclosed to the Yungs the problems with the LevSO 1 concept in general nor specifically, e.g., the use of a recourse bank loan and the need for a stated business purpose for the transaction. When the Yungs discovered on their own that the l.R.S. could possibly disallow the Lev3Ol tax benefits and communicated that concern to GT, GT misrepresented its confidence in the product. Although GT stopped selling Lev3Ol multiple times in response to l.R.S. notices and new regulations, the Yungs were not told even once about the cessation of sales of an increasingly dubious product. At one point, the Yungs’ Lev3Ol use was described as a successful sale to GT’s staff for promotional purposes and the staff was also told, despite it not being so, that the Lev3Ol was vetted and approved by outside counsel. Furthermore, although the Yungs were never informed of the problems associated with their particular transaction, the knowledge of those problems (e.g., recourse loan, business purpose) was reflected in numerous internal communications within GT and personnel sought to avoid like circumstances with other sales. When GT was subject to an l.R.S. examination because of the Lev3Ol, GT did not 75 inform the Yungs; instead these “trusting” clients learned of the scrutiny from a tax publication. These various misrepresentations and nondisclosures were made to save the $900,000 deal and to cover GT’s negligent and fraudulent acts that accumulated over time. This summary of GT’s grossly negligent and fraudulent behavior does not fully reflect GT’s reprehensible behavior in the marketing and sale of the Lev3Ol to the Yungs. In our view, these individual and cumulative acts place GT’s behavior toward their clients at the high end of professional reprehensibility. Although the Yungs may not have been financially vulnerable, the reprehensibility of GT’s orchestrated, on-going deceit is not lessened or mitigated by the fact GT defrauded people of wealth, rather than the financially vulnerable. We noted above in the discussion of justifiable reliance that a plaintiffs wealth and business experience cannot preclude a finding of reliance on that plaintiffs trusted accounting and tax advisors nor should it preclude a punitive damage award where the advisors’ conduct is reprehensible. In sum, like the trial court, we believe the fourth and fifth factors weigh heavily against GT. GT challenges the trial court’s reprehensibility assessment as erroneous, contending the repeated conduct factor does not apply. We do not agree. Citing Willow Inn, Inc. v. Public Service Mutual Insurance Co., 399 F.3d 224, 232 (3d Cir. 2005), and Kentucky Farm Bureau Mutual Insurance Co. v. Rodgers, 179 S.W.3d 815 (Ky. 2005), GT presents two inter-related arguments. 76 First, a “repeated conduct” finding cannot be based merely on the various misrepresentations and omissions allegedly made just to the Yungs as opposed to them and others. Second, the trial court did not make findings that GT’s behavior toward other LevSO 1 clients was sufficiently similar to the conduct in this case. In Willow Inn, the Third Circuit found that the District Court improperly considered the various stonewalling tactics employed by the insurance company in processing Willow Inn’s claim to satisfy the “repeated conduct” reprehensibility factor. This finding was based upon BMW and State Farm, wherein the improper transaction handling, the repeated conduct, went beyond a single claimant. Id. at 232.46 Although the trial court considered GT’s behavior toward other Lev3Ol clients, GT contends that the trial court’s findings were not sufficient. We cannot agree that the trial court’s findings were lacking as to GT’s conduct toward other clients. In State Farm, the Supreme Court stated, “[a] defendant’s dissimilar acts, independent from the acts upon which liability was premised, may not serve as the basis for punitive damages.” 538 U.S. at 422. In Rodgers, this Court stated. Nevertheless, the Willow Inn court subsequently stated that the repeated 46 conduct element applied, although with less force, because “insofar as the series of actions and inaction by [the insurer] which delayed settlement of the claim until more than two years after the windstorm implied a concerted effort to lessen [the insurer’s] expected payment on the claim,” the pattern of delay suggested the actions were designed to achieve a fiscally beneficial result for the insurer, which is against the dictate that an insurer must act with the “utmost good faith” toward its insured. Id. at 232-33. 77 The requirement in [State Farm] that the present conduct “replicates” the prior transgressions, mirrors our requirement that to be admissible under KRE 404(b) to prove, e.g., motive, intent, plan or identity, by modus operandi, the prior bad act must have been so strikingly similar to the present act as to constitute a “signature crime.” Of course, as noted in [State Farm], “strikingly similar” does not necessarily mean “identical.” 179 S.W.3d at 819 (internal citations omitted). We fail to see how the trial court’s description of Lev301’s creation; the marketing by GT to other clients; the championing of the product to others when it was removed from GT’s Client Matrix; all LevSOl purchasers being subject, without their knowledge, to GT’s maintenance list; nearly all clients being subsequently audited by the l.R.S. for participation in the strategy; and the Wytec and Casuarina opinions serving as the model opinion for other LevSO 1 clients is an insufficient finding of similarity between the Yungs and other GT clients. Consequently, we do not find that the trial court erred when using evidence of GT’s similar conduct with other clients as a basis for finding GT engaged in repeated misconduct. GT also argues that reprehensibility cannot be enhanced here by consideration of “potential harm.” That position stems in part from the Yungs’ argument that relevant to reprehensibility is the fact that the Yungs’ actual harm was significantly less than the potential harm that GT knew its fraudulent scheme risked inflicting; it also stems from the Court of Appeals’ agreement that the trial court could properly consider GT’s knowledge of those 78 harms, such as that flowing from the Yungs’ personal reputational loss,47 as a factor in determining the reprehensibility of the conduct. We are unpersuaded by GT’s argument that this is not the type of potential harm which may be considered under State Farm’s reprehensibility analysis. Although GT asserts that it is of no consequence that the punitive damage award resulted from a bench trial, it is clear here that the fact-finder was astute and engaged as she heard testimony from witnesses, observed their demeanor and assessed their credibility. The 162 pages of findings of fact reflect a thorough grasp of the evidence as well as numerous specific credibility determinations. As Willow Inn noted, when an appellate court must perform a de novo review, awareness that a fully informed judge was the fact-finder may temper the concerns that passion, prejudice or bias impacted the punitive damage award. “In such a case, the potential for an ill-founded or inflated reprehensibility determination is less.” 399 F.3d at 231. 2. Ratio Between Harm and Punitive Damages Award “The second and perhaps most commonly cited indicium of an unreasonable or excessive punitive damages award is its ratio to the actual harm inflicted on the plaintiff” BMW, 517 U.S. at 580. It is well-recognized that “a mathematical bright line [cannot be drawn] between the constitutionally acceptable and the constitutionally unacceptable” award. Id. at 560. The Supreme Court has provided, however, some general guiding principles to aid As noted, the Yungs’ extensive involvement in the casino business, a highly47 regulated industry, made them particularly sensitive to protection of their personal and business reputations. 79 the courts in the assessment of the reasonableness of an award. First, “few awards exceeding a single-digit ratio between punitive and compensatory damages, to a significant degree, will satisfy due process.” State Farm, 538 U.S. at 425. Second, “an award of more than four times the amount of compensatory damages might be close to the line of constitutional impropriety.” Id. (citing Haslip, 499 U.S. at 23-24). Third, “[w]hen compensatory damages are substantial, then a lesser ratio, perhaps only equal to compensatory damages, can reach the outermost limit of the due process guarantee.” Id. GT, not surprisingly, emphasizes this third point. Here, the ratio is 4:1. Across the three principles, this one — the ratio — principle is the one which primarily brings due process scrutiny. A $20 million compensatory damage award is indisputably substantial, and the ratio principle requires us to consider whether an award of punitive damages equal to that substantial compensatory award is the outermost limit allowed without violating due process. The Yungs argue that the trial court weighed the evidence, made findings of fact supported by substantial evidence and then considered the constitutionality of the punitive damage award under state and federal law. GT maintains that, even if this Court accepts the trial court’s fraud findings, GT’s intentional trickery or deceit did not merit a punitive damage award beyond one equal to the compensatory damage award — the conclusion reached by two members of the Court of Appeals’ panel.48 48 One member of the appellate panel would have affirmed the $80 million punitive damage award. 80 Both parties cite many cases from a variety of jurisdictions to support a decision in their favor.49 For example, GT cites Morgan v. New York Life Insurance Co., 559 F.3d 425 (6th Cir. 2009), as a case supporting a 1:1 or lower punitive/compensatory damage ratio as the absolute constitutional limit in this case because Morgan, under like circumstances — large compensatory damage award, harm solely economic, and the plaintiffs not financially vulnerable — imposed a 1:1 ratio. In that age discrimination action, the Sixth Circuit held that a $10 million punitive damage award was excessive in comparison to a $6 million compensatory damage award when the plaintiff was not financially vulnerable and the repeated misconduct of some company officials (conscious disregard ... [of] Morgan’s right to be free from age discrimination) did not appear so reprehensible as to justify a high punitive damage award. 559 F.3d at 442. GT also cites Boemer v. Brown & Williamson Tobacco Co., 394 F.3d 594 (8th Cir. 2005), a product liability case, to show that even where reprehensible behavior is substantial, some courts still limit the ratio to 1:1. Although the tobacco company’s behavior in that case (“callous disregard for the health consequences of smoking”) was found to be highly reprehensible, the Boemer court lowered the punitive damage award to follow its precedent. The court 49 While we often look to prior decisions for guidance, TXO, 509 U.S. at 457, reminds us that this comparative approach is not part of the test for assessing the constitutionality of a punitive damage award. Furthermore, as is often noted, comparisons of awards and ratios are often difficult because the circumstances in the cases vary significantly from those in the case at bar and among themselves. See Ragland, 352 S.W.3d at 923. 81 noted that unlike another of its decisions, Eden Electrical, Ltd. v. Amana Co., 370 F.3d 824, 829 (8th Cir. 2004), in which a 4.5:1 punitive/compensatory damage award was affirmed in a business fraud case, there was no evidence that the tobacco company intended to victimize its customers. Neither Morgan nor Boemer captures the essence of the culpability in the instant case — repeated fraudulent conduct in the form of misrepresentations and omissions by a trusted professional advisor. Eden on the other hand, referenced in Boemer and cited by the Yungs, illustrates that purely economic injury in a scheme to defraud can justify a punitive damage award beyond a 1:1 ratio where the defendant’s conduct is highly reprehensible. In that case, after a jury verdict awarding Eden Electrical, Ltd. $2.1 million in compensatory damages and $17,875 million in punitive damages against Amana Appliances, Inc., the district court reduced the punitive damage award to $10 million. Amana appealed the award as excessive, but upon review of the following facts, the Eighth Circuit affirmed the roughly 5:1 punitive/compensatory damage award. In Eden, Leon Adam owned and controlled Amana’s Israeli distributorship from which Eden purchased Amana refrigerators to sell in its twenty-five appliance stores throughout Israel. Because of Adam’s legal and financial problems, Amana approached Adam to find another entity to be its Israeli distributor and Adam, in turn, approached Eden. Id. at 826. Eden’s representatives traveled to Iowa, met with Amana executives, signed the distributorship agreement, and delivered to Amana's executives a check for $1.2 million and letter of credit in 82 the same amount. During the negotiations, Amana executives, including a territory manager, the international credit manager, and a vice president, made a variety of assurances to Eden's representatives about Amana’s good faith, its hope of having a long-term business relationship with Eden, and its willingness to have a direct business relationship with Eden as its exclusive distributor in Israel. Seventy-seven days after the agreement was reached and payment was made, Amana terminated the distributorship contract without any explanation. Eden’s attempts to make contact with Amana were met with no response. Unbeknownst to Eden, which believed it was embarking on a long-term relationship as Amana’s exclusive distributor, Amana had, following the execution of the agreement, continued selling refrigerators to other entities for the Israeli market and had represented to others that it was still looking for a long-term distributor for Israel. Adam testified that he had numerous conversations with Amana’s vice president Montross about the “Eden project,” “the only purpose of which was to get rid of the [$2.4 million worth of] inventory which commonly we called . . . junk.” Amana’s leadership directed Adam to find someone in Israel “who can take the junk.” He was instructed to offer the exclusive distributorship as a way of selling “the junk” to someone who had the requisite cash available. . . . Amana entered into the agreement with Eden, then ignored all communications from Eden (including an attempt to place additional orders for Amana appliances) and failed even to deliver to Eden the $2.4 million worth of inventory it had purchased. Eden’s was the briefest exclusive distributorship in Amana's history. Even after Eden received the faxed two-page termination letter, Amana refused to discuss the situation. Id. at 826-28. Because of the extraordinarily reprehensible scheme to defraud, the Eighth Circuit affirmed the punitive damage award even though it was a commercial case and Eden’s compensatory damage recovery was significant. Id. at 829. The circuit court noted that “Amana's actions were purposefully designed to maliciously victimize another company, all the while giving Eden the impression that it was entering into a long-lasting and mutually profitable 83 relationship;” and “[t]he scheme to defraud Eden involved various members of Amana’s leadership team, including a vice president, the international credit manager, and a territory manager.” Although we cannot say the LevSO 1 was purposefully designed to victimize the Yungs, it was a product that GT knew very early on would likely implode with the I.R.S., causing serious financial and business consequences for the Yungs, and yet GT urged them to complete the transaction and never advised them to unwind it. When the Yungs inquired about the effect of new regulations, GT “doubled down,” falsely insisting the regulations actually helped their “more likely than not” conclusion, a conclusion that did not exist at that time and was never truly reached by the Firm. Despite numerous opportunities to alter course and mitigate the damage, GT continually urged the Yungs on, concealing numerous important facts right up to and including that the I.R.S. was examining the Firm and its LevSOl product. Throughout this course of repeated conduct, the Yungs were GT’s client — and they relied on GT’s professional advice expecting the Firm to act in their best interests. GT points out that no Kentucky appellate court has considered a punitive damage award of this size for purely economic damages and argues that an $80 million punitive damage award in this case does not fit within Kentucky’s precedent for far more reprehensible actions. GT notes that, even before State Farm, in Sand. Hill, 83 S.W.Sd 483, this Court reduced a punitive damage award from $20 million to $15 million under BMW when a truck owner was crushed to death because of a defective transmission. Also, in Ragland, 84 352 S.W.3d 908, the Court of Appeals reduced punitive damages from $60 million to $30 million in a wrongful death case arising from the murder of a young college student. It is noteworthy that in Sand Hill and Ragland, these reductions resulted in punitive/compensatory damage award ratios of 5:1 and 9:1, respectively. While a punitive damage award may be large, the value of the award is not considered in a vacuum, but as part of the three guideposts. When the Supreme Court noted in State Farm that a 1:1 ratio may violate due process, the Court did not suggest that the interplay with the reprehensibility of the defendant’s conduct (or the third “penalties” guidepost) should no longer be considered. Indeed, the Supreme Court has identified wrongful conduct which is “the result of intentional malice, trickery, or deceit,” as “particularly reprehensible conduct.” BMW, 517 U.S. at 576. Because the amount of an award is a reflection of the fact-finder’s measure of a defendant’s reprehensible conduct, the ratios must be assessed on a case-by-case basis. Considering GT’s highly reprehensible conduct, we do not find that, in the context of this case, the 4:1 ratio reflects an overly severe punishment. As noted in BMW, “[i]n most cases, the ratio will be within a constitutionally acceptable range, and remittitur will not be justified on this basis. When the ratio is a breathtaking 500 to 1, however, the award must surely ‘raise a suspicious judicial eyebrow.”’ Id. at 583 (citing TXO, 509 U.S. at 481). In sum, we cannot disagree with the trial court that GT’s conduct was egregious and highly reprehensible, and a substantial punitive damage award is warranted. We do not view the $80 million punitive damage award to be 85 disproportionate to the harm suffered by the Yungs. As noted by the trial court, the 4:1 punitive/compensatory damage award is significantly lower when taking into consideration potential harm. We now turn to the third guidepost. 3. Disparity Between Punitive Damage Award and Civil Penalties “Comparing the punitive damage award and the civil or criminal penalties that could be imposed for comparable misconduct provides a third indicium of excessiveness.” BMW, 517 U.S. at 583. These penalties provide notice of possible sanctions to potential violators and provide guidance concerning appropriate sanctions for the conduct at issue. Id. at 583-84. According to testimony, GT was aware of the overall environment putting accounting firms (like KPMG) and law firms at risk, including Congress’s perception of tax shelters, the l.R.S.’s actions in contesting tax shelters, and the implosion of at least two law firms. The trial court concluded that penalties in the hundreds of millions of dollars, such as those levied against KPMG, were sanctions that the l.R.S. could have imposed on GT. Furthermore, several GT partners could have had their law and/or accounting licenses suspended for the fraudulent acts involved in the Yungs’ transaction. Although GT suggests otherwise, we do not find the trial court’s findings on this third indicium to be clearly erroneous. In general, a 4:1 punitive/compensatory damage ratio is not an outlier in Kentucky for economic cases, either before or after State Farm. For example, in Hanson (applying TXO), a lender liability/fraud case, the Court affirmed an 86 approximately 5:1 punitive/compensatory damage ratio. In Farmland Mutual Insurance Co. v. Johnson, 36 S.W.3d 368 (Ky. 2000), an insurance bad faith case not involving bodily injury, the Court affirmed a 7:1 ratio. In Kentucky Kingdom Amusement Co. v. Belo Kentucky, Inc., 179 S.W.3d 785, 788 (Ky. 2005), a defamation/lost profits case, the Court upheld an approximately 5:1 ratio. And in Craig Bishop, Inc. v. Piles, 247 S.W.3d 897 (Ky. 2008), a consumer protection/fraud case, the Court upheld a 6:1 ratio. Unlike the Court of Appeals, to the extent it did so, we do not find the penalties actually imposed by the l.R.S. through a settlement agreement with GT to reflect the measure the Supreme Court instructed courts to use in assessing the disparity between a punitive damage award and other potential penalties. Under BMW, we conclude that the trial court correctly considered the full range of penalties which could have been imposed on GT. GT was on sufficient notice of the possible legal consequences it could face for its highly reprehensible fraudulent and grossly negligent professional conduct. There is no indication that the $80 million award is excessive compared to the potential penalties which could have been imposed on GT. CONCLUSION In consideration of the five factors in KRS 411.186(2), the trial court found that the facts of this case support an $80 million punitive damage assessment and that an award of that magnitude passed constitutional muster under federal due process guidelines. Performing a de novo review of the punitive damage award, we also conclude that the $80 million award is not 87 grossly excessive and is constitutionally acceptable. Consequently, we reverse the Court of Appeals’ remittitur and reinstate the trial court’s $80 million punitive damage award. Otherwise, we affirm the Court of Appeals’ decision. All sitting. Minton. C.J.; Cunningham, Keller, VanMeter, and Wright, JJ., concur. Venters, J., concurs in part and dissents in part by separate opinion. VENTERS, J., CONCURRING IN PART AND DISSENTING IN PART: I respectfully disagree with the Majority’s decision, after de novo review, to reinstate the trial court’s punitive damage award of $80 million dollars. I believe the trial court placed too much dependence upon the use of a 4:1 ratio and not enough emphasis upon the absolute amount of the punitive damages it awarded. I would affirm the Court of Appeals remitter of a $20 million-dollar award of punitive damages. Based upon the relevant factors involved in our de novo review, I conclude that the outrageous conduct of the Appellee/ Cross- Appellant is adequately punished with an exemplary award of $20 million dollars. The five factors to be examined in a review of the degree of reprehensibility of Grant Thornton’s conduct point toward a lesser amount than the sum imposed by the trial court. The harm inflicted was not physical injury; it did not reflect disregard for the personal safety and health of others; and the target of Appellee’s reprehensible conduct was far from vulnerable but was instead a powerful interest in its own right searching for a means to bring millions of dollars in overseas profits into this country without paying taxes. 88 A 4:1 ratio between the harm inflicted and a punitive damage award is not per se excessive, but the ratio cannot be properly evaluated without regard for its effect in absolute terms. A 4:1 ratio applied to a small compensatory loss of only a $1 million yields a reasonably punitive sum of $4,000,000. The punitive damage award exceeds the compensatory damages by $3 million. Geometrically increasing the penalty by a multiplier of four quickly becomes excessive when the compensatory damages increase by $19 million but punitive damage assessment expands exponentially by the sum of $76,000,000. Consequently, I would affirm the Court of Appeals with respect to the award of punitive damages. I have no other disagreement with the Majority’s otherwise fine analysis. COUNSEL FOR APPELLANTS/CROSS-APPELLEES, WILLIAM J. YUNG AND MARTHA A. YUNG: Michael D. Risley Bethany A. Breetz Matthew W. Breetz Stites & Harbison, PLLC COUNSEL FOR APPELLANT/CROSS APPELLEE, THE 1994 WILLIAM J. YUNG FAMILY TRUST: Gerald F. Busing Adams, Stepner, Woltermann & Busing, PLLC 89 COUNSEL FOR APPELLEE/CROSS-APPELLANT, GRANT THORNTON, LLP: Sheryl G. Snyder Griffin Terry Sumner Theresa Agnes Canaday Jason Patrick Renzelmann Frost Brown Todd, LLC AMICUS CURIAE: KENTUCKY DEFENSE COUNSEL, INC.; KENTUCKY CHAMBER OF COMMERCE; AND KENTUCKY BANKERS ASSOCIATION: Aaron John Silletto Goldberg Simpson, LLC KENTUCKY SOCIETY OF CERTIFIED PUBLIC ACCOUNTANTS: Mark Allen Loyd, Jr. Bailey Roese Bingham Greenebaum Doll, LLP Christopher R. Healy Kevin M. Dinan Patricia L. Maher Ashley C. Parrish King & Spalding, LLP KENTUCKY JUSTICE ASSOCIATION: Kevin Crosby Burke Jamie Kristin Neal Burke Neal PLLC 90
Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You should read the full case before relying on it for legal research purposes.
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
