Ballard v. Commissioner

522 F.3d 1229, 101 A.F.T.R.2d (RIA) 1577, 2008 U.S. App. LEXIS 7373

• Court: Court of Appeals for the Eleventh Circuit
• Decided: April 7, 2008
• Docket: 01-17249, 01-17251, 01-17253, 01-17255 to 01-17257, 07-12866 to 07-12871
• Status: Published

Opinion

FAY, Circuit Judge:

The question presented in this appeal is simply whether the Tax Court Judge reviewing the report of the Special Trial Judge gave due regard to the factfindings of the trial judge. 1 That is the only thing simple about this case.

I. Overview

Claude M. Ballard and his wife, Mary B. Ballard, 2 appeal the judgment of the U.S. Tax Court that they fraudulently failed to declare and pay income tax on approximately $3.2 million earned through a kickback scheme. More specifically, the Tax Court held that Ballard and an acquaintance sold their influence at Prudential Life Insurance Company of America (“Prudential”) for money and then hid this income from the Internal Revenue Service (“IRS”). In reaching the conclusion that the Ballards were liable for tax deficiencies and penalties for fraud, the Tax Court Judge rejected the recommendation of the Special Trial Judge, who presided over the Ballards’s trial and concluded that the record did not demonstrate that they earned personal income through a kickback scheme or committed fraud in reporting their income. The Tax Court Judge found that the Special Trial Judge’s fact findings and credibility determinations were incomplete or manifestly unreasonable. On appeal, the Ballards argue that the record supported the Special Trial Judge’s conelu- sions and that the Tax Court Judge failed to give the Special Trial Judge’s report the deference due a finder of fact. We agree. Accordingly, for the reasons discussed more fully below, we vacate the Tax Court’s judgment as to the Ballards and remand with instructions to issue a final order approving and adopting the Special Trial Judge’s report. 3

The Commissioner of Internal Revenue (“Commissioner”) alleged that five separate businessmen or corporations (“the Five”) paid Burton W. Kanter, an expert tax attorney with extensive ties to the business world, to generate business on their behalf. It is alleged that Kanter, in turn, paid Ballard and Robert W. Lisle, both of whom were high-level employees of Prudential, to use their influence at the insurance company to direct business to the Five. 4 The government argues that Kanter arranged for his, Ballard’s, and Lisle’s shares of the payments to be tunneled through corporations and partnerships disguised as corporate income. It urges that Kanter, Ballard, and Lisle, however, each ultimately received and made personal use of their portion of the money. While the corporations and partnerships paid income taxes on the money, Kanter, Ballard, and Lisle did not declare this money on their individual tax returns. Accordingly, it is contended they were responsible for tax deficiencies, having failed to pay income tax on the money, and penalties for fraud, having disguised and routed the money through corporations and partnerships.

In contending that the payments from the Five were earned by, and taxable to, Ballard, the Commissioner relied on two main theories: (1) Ballard attempted to improperly assign his income to the corporation through which the payments were routed and was personally responsible for this money as its true owner; and (2) a corporation to which Ballard’s alleged portion of this money was transferred was Ballard’s alter ego, such that Ballard was personally responsible for this money.

After hearing evidence in support of these allegations, Special Trial Judge D. Irwin Couvillion found that the evidence was insufficient to show the existence of a kickback scheme, such that he could not conclude that the parties earned income in the manner alleged or acted fraudulently in neglecting to report such income. After reviewing Judge Couvillion’s report, the Tax Court Judge, Harry A. Haines, disagreed. Judge Haines found that Judge Couvillion’s recitation of the facts was incomplete or manifestly unreasonable. Judge Haines concluded that the evidence demonstrated that the parties had earned income from the Five and had fraudulently disguised it as corporate income to avoid the personal income tax requirements of the Internal Revenue Code. 5

On appeal, the parties do not dispute that the payments from the Five repre sented income or that the entities through which Kanter routed the money declared it as income on their tax returns. However, the Ballards maintain that the money was corporate income only, such that they were not obligated to report it on their personal tax returns. The Commissioner maintains that the Ballards earned and made personal use of the money and committed fraud by intentionally shielding it from the Internal Revenue Service (“IRS”). The Bal-lards argue that, in siding with the Commissioner, Judge Haines did not give due deference to Judge Couvillion’s findings and conclusions. The Ballards argue that these findings and conclusions should control, as they were supported by the record. Thus, the issue now before us is whether Judge Haines’s review was appropriately deferential.

II. Procedural History

As set forth in our earlier opinion, this case comes to us with this background:

Petitioners-Appellants received Notices of Deficiency from the IRS pertaining to years 1975 through 1982, 1984, and 1987 alleging that they owed additional taxes. As to each deficiency asserted by the IRS, the Ballards filed petitions for redetermi-nation in the Tax Court. Pursuant to I.R.C. § 7443A and Rules 180, 181 and 183, the Chief Judge of the Tax- Court assigned the consolidated case to Special Trial Judge D. Irwin Couvillion for trial.

At the conclusion of the five-week trial during the summer of 1994, Special Trial Judge Couvillion, in accordance with Rule 183(b), prepared and submitted a written report containing his findings of facts and opinions to the Chief Judge for subsequent review by a Tax Court Judge. In accordance with Rule 183, none of the litigants received a copy of Special Trial Judge Couvillion’s report at that time. Thereafter, pursuant to Rule 183(b), the Chief Judge assigned the case to Tax Court Judge H.A. Dawson, Jr. for his review and final disposition. On December 15, 1999, Judge Dawson issued the opinion of the Tax Court in which the Tax Court both approved of and adopted Special Trial Judge Couvillion’s report (T.C. Memo 1999-407, see Investment Research Assoc. Inc. v. Commissioner, 78 T.C.M. (CCH) 951 (1999)), a copy of which was provided to the parties. On July 24, 2001, Judge Dawson entered the final order of the Tax Court against Petitioners-Appellants, assessing tax deficiencies of $1,318,648. Of that amount, $422,812 is penalties against Ballard pursuant to I.R.C. § 6653(b).

On April 20, 2000, prior to the Tax Court’s final order of assessment, the Bal-lards, joined by the other petitioners, filed a motion requesting access to “all reports, draft opinions or similar documents, prepared and delivered to the [Tax] Court pursuant to Rule 183(b),” or, in the alternative, that the Tax Court either certify the issue for interlocutory appeal pursuant to Rule 193 or make the initial findings part of the record for subsequent appeal to the circuit court. On April 26, 2000, Judge Dawson issued an order denying the motion. In the order, Judge Dawson noted that “[he] gave due regard to the fact that Special Trial Judge Couvillion evaluated the credibility of witnesses ... and treated the findings of fact recommended by the Special Trial Judge as being presumptively correct.” On May 26, 2000, the Ballards along with the other petitioners, filed a second motion with the Tax Court. The second motion requested that Special Trial Judge Couvillion’s original report or other documentation be placed under seal and made part of the record for subsequent appellate review. That motion was denied on May 30, 2000.

On August 22, 2000, the Ballards once again joined by the other petitioners, filed a motion requesting that the Tax Court reconsider its denial of access to Special Trial Judge Couvillion’s original report or, alternatively, that the Tax Court grant the petitioners a new trial. In support of this motion, an affidavit from Randall G. Dick (“Dick”), attorney for IRA and for Kanter, was filed. In the affidavit, Dick indicated that two unidentified Tax Court Judges approached him and stated that in the original report submitted to the Chief Judge in accordance with Rule 183(b), Special Trial Judge Couvillion concluded that payments made by “the Five” were not taxable to the individual petitioners and that the fraud penalty was not applicable. Furthermore, Dick indicated that the two unidentified Tax Court Judges expressed that “substantial sections of the opinion were not written by Judge Couvillion, and that those sections containing findings related to the credibility of witnesses and findings related to fraud were wholly contrary to the findings made by Judge Cou-villion in his report.” According to Dick, the two Tax Court Judges stated that the changes to Special Trial Judge Couvillion’s findings relating to credibility and fraud were made by Judge Dawson. Finally, Dick indicated that he confirmed what he was told by the two unidentified Tax Court Judges with yet another unidentified Tax Court Judge. Apparently, the third unidentified Tax Court Judge confirmed that Special Trial Judge Couvillion’s opinion had been “changed.” On August 30, 2000, the Tax Court issued an order signed by Special Trial Judge Couvillion, Judge Dawson and the Chief Judge of the Tax Court denying the motion and confirming that, contrary to the contents of the affidavit, the underlying report adopted by the Tax Court is, in fact, Special Trial Judge Cou-villion’s report.

Subsequently, the Ballards petitioned this court for a writ of mandamus seeking an order directing the Tax Court to provide the Ballards with a copy of the original Special Trial Judge Couvillion report, or, alternatively, seeking an order requiring that the Tax Court provide any changes made by Judge Dawson to the original Special Trial Judge Couvillion report. The petition was denied on October 23, 2000. Ballard v. Comm’r of Internal Revenue, 321 F.3d at 1037, 1040-41 (11th Cir.2003).

Thereafter, we affirmed the ruling of the Tax Court. Ballard, 321 F.3d 1037 (11th Cir.2003). Following this, the Supreme Court granted certiorari and reversed. Ballard v. Comm’r of Internal Revenue, 544 U.S. 40, 125 S.Ct. 1270, 161 L.Ed.2d 227 (2005). The Court disapproved the procedure used by the Tax Court in refusing to make available the original report of the Special Trial Judge, disapproved of the procedure used to create the collaborative report adopted by the Tax Court, vacated the earlier judgments and remanded with instructions.

Subsequently, we remanded the matter to the Tax Court with instructions to: (1) Strike the collaborative report; (2) Reinstate the report of the Special Trial Judge; (3) Assign the case to a regular Tax Court Judge with no prior involvement; (4) Review the case in accord with its revised Rules 182 and 183. Ballard v. Comm’r of Internal Revenue, 429 F.3d 1026 (11th Cir.2005). In that opinion we discussed the deference due the Special Trial Judge’s findings of fact. Id. at 1029-32. We instructed that those findings should not be ignored or changed unless “manifestly unreasonable.” Id. at 1031. This was in accord with the language of the Supreme Court which noted: “The officer who hears witnesses and sifts through evidence in the first instance will have a comprehensive view of the case that cannot be conveyed full strength by a paper record.” It continued: “Fraud cases, in particular, may involve critical credibility assessments, rendering the appraisals of the judge who presided at trial vital to the Tax Court’s ultimate determinations. This case is illustrative. The Tax Court’s decision repeatedly draws outcome — influencing conclusions regarding the credibility of Ballard, Ranter, and several other witnesses.” Id. at 43.

Following our remand the case was assigned to regular Tax Court Judge Harry A. Haines who entered the current opinion of the Tax Court. After extensive review of the record and the opinion of Judge Haines, we conclude that the current opinion must be stricken and the original report of Special Trial Judge Couvillion reinstated as the ruling of the Tax Court.

III. Standard of Review

The Commissioner’s determination of a tax deficiency is entitled to a presumption of correctness, and the taxpayers have the burden of rebutting this presumption by a preponderance of the evidence. Bone v. Comm’r of Internal Revenue, 324 F.3d 1289, 1293 (11th Cir.2003). The Commissioner has the burden of proving allegations of fraud by clear and convincing evidence. Henson v. Comm’r of Internal Revenue, 887 F.2d 1520, 1525 (11th Cir.1989).

When the Commissioner and taxpayer present their evidence on these matters to a Special Trial Judge, Tax Court Rule 183 governs. See Tax Court Rule 183. Rule 183 provides that, after hearing the evidence and considering the parties’ arguments, the Special Trial Judge should file recommended findings of fact and conclusions of law with the Tax Court. Tax Court Rule 183(b). The Tax Court then assigns the report to a Tax Court Judge for final review. See Tax Court Rule 183(d). Rule 183 mandates that, in conducting this review, the Tax Court Judge “[shall give] [d]ue regard ... to the circumstance that the Special Trial Judge had the opportunity to evaluate the credibility of witnesses” and must “[presume to be correct] the findings of fact recommended by the Special Trial Judge.” Tax Court Rule 183(d). Ultimately, the Tax Court Judge may adopt, modify, or reject in whole or in part the report, or may request additional briefs, evidence or oral argument. Id.

Our sister circuits have explained that the deference mandated by Rule 183 requires the Tax Court Judge to review the Special Trial Judge’s report for clear error. See Stone v. Comm’r of Internal Revenue, 865 F.2d 342, 345-47 (D.C.Cir.1989). Likewise, we have explained that the Tax Court Judge may disturb the Special Trial Judge’s findings of fact and credibility determinations only if they are “manifestly unreasonable.” Ballard II, 429 F.3d at 1031, 1032. To further define the deference due the Special Trial Judge’s determinations, we find the Supreme Court’s explanation of the clear error standard, in the context of a Court of Appeals reviewing the district court’s findings of fact in a case tried without a jury, helpful. See Anderson v. City of Bessemer City, N.C., 470 U.S. 564, 573-74, 105 S.Ct. 1504, 1511-12, 84 L.Ed.2d 518 (1985). The Supreme Court explained:

[The clear error] standard plainly does not entitle a reviewing court to reverse the finding of the trier of fact simply because it is convinced that it would have decided the case differently. The reviewing court oversteps the bounds of its duty ... if it undertakes to duplicate the role of the lower court. In applying the clearly erroneous standard to the findings of a district court sitting without a jury, appellate courts must constantly have in mind that their function is not to decide factual issues de novo. If the district court’s account of the evidence is plausible in light of the record viewed in its entirety, the court of appeals may not reverse it even though convinced that had it been sitting as the trier of fact, it would have weighed the evidence differently. Where there are two permissible views of the evidence, the factfinder’s choice between them cannot be clearly erroneous.

Id. (internal quotation and citation omitted). Specifically regarding credibility determinations, the Supreme Court directed that

when a trial judge’s finding is based on his decision to credit the testimony of one of two or more witnesses, each of whom has told a coherent and facially plausible story that is not contradicted by extrinsic evidence, that finding, if not internally inconsistent, can virtually never be clear error.

Id. at 575, 105 S.Ct. at 1512. The Supreme Court explained that credibility determinations are due this extra deference because “only the trial judge can be aware of the variations in demeanor and tone of voice that bear so heavily on the listener’s understanding of and belief in what is said.” Id.

After the Tax Court Judge has conducted his review and issued a final opinion, we generally review this opinion for clear error. See Stone, 865 F.2d at 344; 26 U.S.C. § 7482(a)(1) (instructing Courts of Appeals to review Tax Court decisions in the same manner as district court decisions on civil cases tried without a jury); Fed.R.Civ.P. 52(a)(6) (instructing that Courts of Appeals can set aside the district court’s findings on cases tried without a jury if they are clearly erroneous). However, when the Tax Court Judge rejects the Special Trial Judge’s findings and we are faced with a conflicting report and opinion, we determine whether the Tax Court Judge committed clear error by rejecting the Special Trial Judge’s findings as clearly erroneous by reviewing the Special Trial Judge’s report to determine if his findings were, indeed, without record support. 6 Matter of Multiponics, Inc., 622 F.2d 709, 712-13, 722 (5th Cir.1980) 7 (applying Rule 52(a) to bankruptcy proceedings, wherein the district court rejected the Special Master’s findings, and holding that “we must review the findings of the Special Master and may affirm the [district [cjourt’s reversal only if we also deem the Master’s findings clearly erroneous”).

IV. Facts

We will begin with a summary of the case. We then will divide the specific facts of the case into sections. For each section, we will (1) state the relevant findings, determinations, and reasoning from Judge Couvillion’s report; (2) where necessary, explain the grounds on which Judge Haines concluded these findings incomplete or manifestly unreasonable; and (3) provide any facts from the record necessary to determine whether Judge Haines correctly concluded that Judge Couvillion’s determinations were manifestly unreasonable.

A. The Parties

Judge Couvillion found that Ballard was a senior vice president in the real estate department of Prudential’s corporate headquarters. His position entailed purchasing, developing, managing, and selling Prudential properties. In 1982, Ballard left Prudential and began working at Goldman Sachs. Separate from his employment at Prudential and Goldman Sachs, Ballard managed TMT, Incorporated (“TMT”). TMT’s preferred stock was owned by the Orient Trust, which benefited Ballard’s family. TMT’s common stock was owned by Investment Research Associates, Limited (“IRA”), a subchapter C corporation.

Lisle was a vice president in the real estate department at Prudential’s corporate headquarters and president of a real estate subsidiary of Prudential. In 1982, Lisle left Prudential and began working for Travelers Insurance Company (“Travelers”). Separate from his employment at Prudential and Travelers, Lisle managed Carleo, Incorporated (“Carleo”). Carlco’s preferred stock was owned by the Christie Trust, which benefitted Lisle’s family. Carlco’s common stock was owned by IRA.

Aside from being a tax expert and businessman, Kanter managed BWK, Incorporated (“BWK”). BWK’s preferred stock was owned by the BK Children’s Trust, which benefitted Kanter’s family. BWK’s common stock was owned by IRA. Generally, TMT, Carleo, and BWK were included in IRA’s consolidated group of corporations for tax purposes.

Ballard and Lisle met Kanter in connection with Prudential’s development of the Houston Hyatt Hotel between 1968 and 1970. Kanter was an employee and business acquaintance of the Pritzker family, who were majority owners of the Hyatt Corporation (“Hyatt”), and a personal friend of A.N. Pritzker, the head of the family.

B. Payments By The Five i. Hyatt

Judge Couvillion found that, in the early 1970s, Prudential was involved in building the Embarcadero Hotel in San Francisco. Lisle and Ballard were chai-ged with finding a company to manage the hotel. Two companies, Del Webb and Intercontinental Company (“Intercontinental”), were interested in the management contract. While Pritzker also was interested in the contract, Lisle initially would not accept a bid from Hyatt. However, J.D. Weaver, a high-level employee of a company that worked with Prudential in developing the Houston Hyatt Hotel, convinced Lisle to accept a bid from Hyatt.

On the day in which the three management companies were to enter their bids, Intercontinental did not appear, and Del Webb’s executive refused to enter a bid, stating that he had already been promised the contract but refusing further elaboration. Accordingly, as Hyatt had entered the only bid, it was awarded the hotel management contract.

Thereafter, Hyatt entered into an agreement with Weaver whereby Hyatt paid Weaver’s corporation, KWJ Corporation, 10% of its fees from managing the Embar-cadero Hotel. Sometime after 1972, Pritz-ker informed Ballard that Hyatt was paying Weaver this “finder’s fee.” Ballard informed Pritzker that if Hyatt could afford to pay a finder’s fee, Prudential likely would reduce its payments under any future hotel management contracts. Pritz-ker assured Ballard that the finder’s fee to Weaver was a unique occurrence. In 1973, Pritzker also informed Kanter of the agreement, asking him to review its terms.

Sometime before March of 1976, Kanter and Weaver agreed that Weaver would transfer to IRA an option to purchase all of KWJ Corporation’s shares, and the at tendant right to the 10% finder’s fee, in exchange for $150,000 and Weaver’s continued right to receive 30% of the Hyatt payments. At the time, KWJ Corporation also had $149,000 in accumulated assets. Approximately three years later, in 1979, IRA exercised its option.

In 1983, KWJ Corporation was liquidated and IRA’s subsidiaries, TMT, Carl-eo, and BWK, received its assets. These subsidiaries formed a partnership, KWJ Company, to which they contributed the assets. TMT and Carleo each had a 45% interest in KWJ Company, and BWK had a 10% interest.

In finding that the Hyatt transaction was not part of a kickback scheme, Judge Couvillion reasoned that Hyatt’s payments to Weaver could not have been kickbacks to Ballard and Lisle for allowing, and accepting, Hyatt’s bid because Kanter, Ballard, and Lisle did not know of the payments until months after the Hyatt/Weaver fee-sharing agreement was made. Moreover, Ballard’s displeasure upon learning of the payments, and Pritz-ker’s assurance that he would not make similar arrangements in the future, “dispelled] the notion that there was collusion between” the parties.

Judge Haines stated that Judge Couvil-lion’s account of the transaction was incomplete and added the following. At trial, Ballard denied ever meeting Weaver. Pritzker did not inform any other Hyatt executives, save his two sons, of his fee-sharing agreement with Weaver.

Regarding the negotiation of IRA’s purchase of KWJ Corporation, Kanter testified that Weaver agreed to sell for the low amount of $150,00 because he needed the money. However, Weaver did not receive any money from the sale until four years later, in the mid-1980s. Once the sale was complete, Weaver and Kanter did not inform Hyatt of the change in ownership. Rather, Weaver simply forwarded future Hyatt payments to Kanter, and Kanter returned 30% of the fee amount to Weaver. Along with one of the checks that Weaver forwarded to Kanter, he sent a letter asking Kanter to “please deposit and issue appropriate checks to the participants.”

Again, when KWJ Corporation was liquidated and KWJ Company formed, no one informed Hyatt of the change. Weaver continued to forward the checks to Kan-ter, and Kanter forwarded the money to TMT, Carleo, and BWK. On one occasion, Kanter paid TMT’s and Carlco’s portions in checks made out to Ballard and Lisle. Later ledger entries indicated that these checks were voided.

In concluding that the Hyatt transaction was indicative of a kickback scheme, Judge Haines reasoned that Judge Couvillion clearly erred in finding that the record did not demonstrate how Weaver persuaded Lisle to allow a Hyatt bid. Specifically, Judge Haines “infer[red]” that Lisle allowed the bid, and that he and Ballard ultimately accepted the bid, because Weaver informed them of Pritzker’s promise to share 10% of Hyatt’s management fees and promised them a portion of these fees if they awarded Hyatt the contract. Later, Weaver also promised Kanter a portion of the fees if he made arrangements to hide the kickback payments. To do so, Kanter devised a plan whereby Weaver would sell to IRA KWJ Corporation and the right to a portion of the Hyatt payments for a nominal fee and IRA would then liquidate KWJ Corporation and form KWJ Company, with Ballard’s TMT, Lisle’s Carleo, and Kanter’s BWK as partners. Judge Haines stated that Kanter’s issuance of checks to Ballard and Lisle demonstrated that the Hyatt payments were earned by Ballard and Lisle personally. Judge Haines also stated that Weaver’s letter referring to the “participants” demonstrated that IRA was not the sole recipient of the Hyatt payments.

Judge Haines further reasoned that Judge Couvillion erred in crediting Ballard’s and Ranter’s testimony. Specifically, Judge Haines stated that Ballard’s testimony that he was displeased upon learning of the fee-sharing agreement was not credible, as it was “self-serving on its face and uncorroborated by other witnesses” and as Pritzker wished to keep the payments secret and would not have volunteered their existence to Ballard. Judge Haines also stated that “Ballard’s indignation [that Pritzker would share his fees] was feigned.” Likewise, Judge Haines pointed out that Ballard’s denial of meeting Weaver cast doubt on his credibility, as even Judge Couvillion acknowledged that Ballard and Weaver met during the Houston Hyatt Hotel project. Regarding Ranter, Judge Haines stated that the fact that Weaver received no money from his sale of RWJ Corporation for four years “cast serious doubt on” Ranter’s testimony that Weaver sold RWJ Corporation because he needed money.

Contrary to Judge Haines’s account of what led Lisle to allow Hyatt’s bid, Ballard testified at trial that, although Lisle initially did not want to accept a Hyatt bid because he had heard that Pritzker was planning to build another hotel in the San Francisco area, he changed his mind when Weaver suggested that Pritzker might not build this other competing hotel if he could bid on and potentially win the Embarcade-ro Hotel contract. Also, the portion of the record cited by Judge Haines to illustrate Ballard’s lack of credibility with regard to knowing Weaver shows that when an attorney asked Ballard if he knew whether Ranter and Weaver were acquaintances, Ballard responded in the negative. As an aside, Ballard then asked the attorney if he had ever met Weaver, to which the questioning attorney replied “No.” Ballard did not deny meeting Weaver, it was the attorney making this denial. Furthermore, contrary to Judge Haines’s account of Ranter’s testimony on Weaver’s sale of RWJ Corporation, Ranter testified that Weaver wanted to sell RWJ Corporation because he needed money and was unhappy with the outcome of a recent dispute with Pritzker regarding the fee-sharing agreement. Judge Couvillion found this testimony credible.

ii. Frey

Judge Couvillion found the following. At some point in 1978, Bruce Frey formed a partnership, BJF Development, to convert apartment complexes into condominiums for sale to individual purchasers. Frey expected to receive development and management fees from these conversion projects. He met with Ranter for tax advice on the projects.

In the course of their meeting, Ranter indicated that “Ranter and/or entities associated with Ranter” could help raise the capital necessary for Frey’s projects. In return, Ranter expected a share of any development and management fees earned.

Frey became involved with Prudential in 1979, when he approached an executive in Prudential’s Miami regional office and offered to purchase for $20 million an apartment complex owned by a pension fund that Prudential managed. As Prudential already had considered selling the complex and $20 million was more than its appraised market value, the regional executive consulted Ballard. Ballard advised selling, noting especially that refusing the offer might violate Prudential’s fiduciary duty to the pension fund. Having purchased the complex, Frey set about converting it. Ranter arranged for several entities to invest in the project, including Zeus Ventures, Incorporated (“Zeus”), a subsidiary of IRA.

Given the success of this project, the regional executive asked Frey to work with Prudential in converting several other apartment complexes owned by Prudential. Ranter again arranged for several entities to invest in these projects, including Zeus.

In 1981, Frey and Ranter’s fee-sharing agreement was formalized in a written agreement between BJF Development and Zeus. Per the agreement, Frey would pay Zeus 5% of its development fees and 20% of its profits from Prudential projects. In keeping with this agreement, Frey’s partnership made payments to Zeus for several of its projects.

In concluding that the payments from Frey to Zeus were not part of a kickback scheme, Judge Couvillion found that Prudential initially agreed to do business with Frey because he offered more than the market value of the complex in question. Judge Couvillion further found that any future dealings between Frey and Prudential were the doing of an un-involved Prudential regional executive. It was a sound business arrangement for both parties.

Judge Haines stated that Judge Couvil-lion’s findings were incomplete and added that Frey paid Ranter’s share of the fees to Zeus only because Ranter instructed him to do so. On one occasion, Frey made a check payable to Ranter, but later voided the check and reissued it to Zeus.

In concluding that the Frey payments were evidence of a kickback scheme, Judge Haines reasoned that Judge Couvillion clearly erred in finding that Ranter did not personally earn the Frey payments. Specifically, Judge Haines stated that Judge Couvillion’s finding that Ranter told Frey that “Ranter and/or entities associated with him” could help raise the necessary capital was manifestly unreasonable. Judge Haines stated that Frey was relying solely on Ranter to help raise capital and that, aside from the investments made by Zeus, no one acting on behalf of a Ranter-related entity assisted Frey in any manner. It was only on Ranter’s instructions that Frey made the payments to Ranter-related entities. Judge Haines further stated that the fact that Frey made a check payable to Ranter, rather than Zeus, indicated that it was Ranter who earned the money.

Judge Haines also reasoned that, while there was no direct documentary evidence that Ranter agreed to share this money with Ballard and Lisle, he could “infer” from the circumstances that Ranter told Ballard and Lisle that if they could influence Prudential to do business with Frey, he would share with them a portion of the fees paid by Frey. The circumstances that Judge Haines pointed to were Ballard’s role in approving the sale of the pension-fund-owned apartment complex and Prudential’s extensive future dealings with Frey.

Such a conclusion simply ignores the facts that were found by Judge Couvillion and supported in the record that the Frey deals with Prudential originated with a Prudential executive located in Miami, Florida, were sound business deals for all concerned, and all monies paid were fully reported on the appropriate tax returns.

in. Schaffel

Judge Couvillion found the following. William Schaffel was a mortgage broker. In 1979, Ranter invited Schaffel, Ballard, and Lisle to dinner. He indicated that he had a business opportunity to discuss with Schaffel. Schaffel, eager for potential business and to meet two senior Prudential real estate executives, accepted the invitation.

At the dinner, the foursome did not discuss Prudential business. Rather, Ranter told Schaffel of a proposed casino in need of financing. Kanter told Schaffel that he “and/or another entity associated with” him could help Schaffel obtain the contract to arrange the casino’s financing in return for 50% of SchaffePs earned fees. Sometime after the dinner meeting, in a letter addressed to IRA, Schaffel agreed to pay IRA 50% of his fees earned from any business deals that IRA or its associates were instrumental in setting up. The casino project ultimately fell through.

In the year that followed, however, Schaffel was able to arrange financing with Prudential for several of his clients. Citing the aforementioned letter, Schaffel shared with IRA a portion of his broker’s fees from each of these projects. Some of the projects occurred after Ballard and Lisle quit working at Prudential.

Later, after Lisle left Prudential and began working at Travelers, Schaffel had substantial business dealings with Travelers. Initially, Schaffel did not share with Kanter his fees earned from these projects, taking the position that their fee-sharing agreement lapsed when Ballard and Lisle left Prudential. Kanter refuted this position and, in a letter to Schaffel, stated that he felt the agreement extended to Travelers projects because of the continuity “of the very personnel to whom [Schaffel was] first introduced.” Ultimately, Schaffel and Kanter agreed that the fee-sharing arrangement extended to Travelers projects.

In concluding that the Schaffel payments were not kickbacks to Ballard and Lisle, Judge Couvillion found that the payments to IRA were simply Schaffel’s way of repaying Kanter for having introduced him to Ballard and Lisle. Judge Couvil-lion acknowledged that SchaffePs letter to IRA indicated that the payments were conditioned upon IRA associates helping to arrange deals and that Ballard and Lisle both knew that Schaffel and Kanter had a fee-sharing agreement, but found that this evidence was insufficient to prove that Ballard and Lisle directed financing to Schaf-fePs clients in return for money.

Judge Haines stated that Judge Couvil-lion’s findings were incomplete and added the following. Lisle testified that he understood that Kanter arranged the dinner in part to see whether Schaffel might be able to do business with Prudential. When Schaffel agreed to split with Kanter a percentage of his earned fees, Schaffel insisted that Kanter’s portion be paid to an individual or entity with a real estate broker’s license. Schaffel was concerned with the legality of sharing broker’s fees with a non-broker.

Shortly after the dinner meeting, a representative of IBM, Incorporated (“IBM”), approached Schaffel about selling IBM’s headquarters to Prudential. Schaffel arranged a meeting between himself, the representative, and Ballard. After this meeting, Ballard referred the matter to a Prudential field office which did complete the purchase. Schaffel split part of his fee for arranging the sale with Kanter.

Judge Haines concluded that the evidence showed that Kanter agreed with Ballard and Lisle to share any fees he might receive from Schaffel if Ballard and Lisle influenced Prudential to direct contracts to SchaffePs clients. Judge Haines reasoned that Judge Couvillion clearly erred in thinking that IRA, rather than Kanter, earned the payments. Specifically, Judge Haines stated that Schaffel expected Kanter alone to help him obtain contracts and that no one at IRA helped arrange such contracts. Judge Haines stated that the only reason that the payments were made to IRA was SchaffePs fear that directly paying Kanter, a non-broker, was illegal.

Judge Haines likewise reasoned that Judge Couvillion clearly erred in finding that there was insufficient evidence to show that Kanter agreed to share the fees with Ballard and Lisle in exchange for their influence. While Judge Haines admitted that there was no direct evidence of this, he stated that a number of facts, namely the joint dinner meeting, Ballard’s “immediate role” in Prudential’s purchase of IBM’s headquarters, Ballard’s and Lisle’s abilities to exert influence at Prudential, and Kanter’s position during his dispute with Schaffel that the fee-sharing agreement was inextricably linked to Lisle’s assistance, provided “compelling circumstantial evidence” of a kickback arrangement. Furthermore, although Judge Haines acknowledged that some of Schaf-fel’s projects were initiated after Ballard and Lisle left Prudential, he stated that “the inference may fairly be drawn” that those projects were an “outgrowth” of Ballard’s and Lisle’s earlier decisions to do business with Schaffel, which gave Schaffel a favorable reputation at Prudential.

Judge Haines draws inferences from the relationships of Kanter, Ballard, Lisle and Schaffel with no hard facts to support his conclusions. Obviously, Kanter could act on behalf of IRA with or without knowledge on the part of Ballard and Lisle. Credibility was the key factor in the facts found by Judge Couvillion and he alone heard and saw the witnesses. The findings drawn by Judge Couvillion are just as plausible as those of Judge Haines and thus they are not clearly erroneous.

iv. Property Management Systems, Incorporated (“PMS”)

Judge Couvillion found the following. Kenneth Schnitzer was a real estate developer. In the course of his development business, he met Ballard. In 1974, he acquired for $1.3 million a small real estate management company, PMS. He immediately set about expanding PMS’s management business.

At the time, PMS managed a relatively small number of Prudential’s properties. In an effort to increase his Prudential business, Schnitzer approached Ballard and offered to give Prudential a 50% interest in PMS, believing that Prudential would award PMS additional management contracts if it had a stake in PMS’s profitability. Ballard’s superiors seriously considered, but ultimately rejected, Schnit-zer’s offer.

In the course of Schnitzer’s talks with Prudential executives, they told him of their interest in standardizing the operating reports prepared by management companies on properties. Although his proposal had been rejected, Schnitzer applied this knowledge and standardized his operating reports for the Prudential properties he already managed. Thereafter, his business with Prudential increased substantially-

In 1977, Kanter told Schnitzer that he and/or IRA could help PMS obtain more management business,- possibly with the Pritzker family. In order to obtain Kan-ter’s and IRA’s assistance, Schnitzer sold IRA a 47.5% interest in PMS for $150,000.

By 1979, however, Schnitzer concluded that Kanter and/or IRA had failed to produce the additional business promised. Accordingly, Schnitzer purchased from IRA the 47.5% interest in PMS for $3.1 million, to be paid over a 10-year period with interest.

In concluding that the PMS transactions were not demonstrative of a kickback scheme, Judge Couvillion found that Schnitzer did not need Kanter’s, Ballard’s, or Lisle’s help obtaining Prudential business, as his contracts with Prudential had increased substantially after he standardized his operating reports. Judge Couvil-lion found that Schnitzer sold a portion of PMS to IRA because he needed help get ting business from sources other than Prudential.

Judge Haines stated that Judge Couvil-lion’s findings were incomplete and added that Schnitzer conferred with Ballard before selling an interest in PMS to IRA. Judge Haines also noted that, after IRA acquired an interest in PMS, PMS’s business with Prudential continued to grow and its gross income increased substantially-

In concluding that Judge Couvillion erred in finding that the PMS deal was not part of a kickback scheme, Judge Haines reasoned that Judge Couvillion’s finding that PMS contracted for Ranter’s and IRA’s services was manifestly unreasonable. Specifically, Judge Haines stated that Schnitzer was relying solely on Ranter and that “there [was] no credible evidence” that anyone other than Ranter brought business to PMS.

Judge Haines also reasoned that Ranter, Ballard, and Lisle realized that they could earn an easy profit by acquiring PMS stock and then using Ballard’s and Lisle’s influence at Prudential to direct Prudential management contracts to PMS. The substantial appreciation between their $150,000 purchase price and $3.1 million sale price represented income to Ballard, Ranter, and Lisle, routed through IRA. While Judge Haines acknowledged that there was no direct evidence of such an arrangement, he argued that the “surrounding circumstances strongly supported] an inference” of such. Specifically, Judge Haines pointed to the facts that Ballard and Lisle knew that Schnitzer was so anxious to expand PMS business that he was willing to part with nearly half of PMS’s stock for nothing, Schnitzer conferred with Ballard before selling part of PMS to IRA, and Ballard and Lisle were in positions to increase PMS’ portfolio of Prudential contracts.

In keeping with Judge Couvillion’s finding that Schnitzer did not need Ranter’s assistance in bringing in more Prudential business, Schnitzer testified at trial that he did not have any interest in Ranter bringing in more Prudential business, as PMS “already had that relationship.” Rather, Schnitzer hoped that Ranter could bring in Hyatt business. Schnitzer was willing to sell an interest in PMS for “something north of nothing” to bring in this extra business.

Again, credibility is the key factor. Judge Couvillion heard Schnitzer’s testimony, evaluated it and found accordingly.

v. Essex Partnership

Judge Couvillion found the following. John Eulich was a real estate developer and business acquaintance of Ballard, Lisle, and Ranter. In 1975, he started Motor Hotels Management, Incorporated (“MHM”). MHM began by operating small hotels, but Eulich also wished to acquire management contracts for larger hotels. He believed that Ranter could be of service in this endeavor because of his extensive contacts in the hotel industry, namely with the Pritzker family.

John Connolly was employed by a separate hotel management company as the on-site manager of Prudential’s Gateway Hilton Hotel (“Gateway”). Connolly had been instrumental in revamping the once-seedy Gateway. In 1981, he informed Prudential that he was considering leaving his position at the Gateway. Not wishing to lose Connolly, Prudential terminated its contract with Connolly’s employer and aimed to award the management contract to Connolly himself. Prudential advised Connolly, however, that he establish a hotel management company of his own. Accordingly, Connolly established Gateway Hotel Management Corporation (“GHM”). Prudential awarded GHM the manage ment contracts for the Gateway and Midland Hilton Hotel (“Midland”).

Eulich and Kanter came to realize that Connolly could use assistance running GHM, as it would be uneconomical for Connolly to employ the necessary full-time financial manager and accounting staff when he managed only two hotels. They also realized that Eulich’s MHM could provide the needed services. Therefore, Eu-lich, Kanter, and Connolly formed the Essex Partnership to provide consulting and liaison services to its partners. MHM had a 47.5% interest in the partnership, and IRA had a 26.125% interest. The Holding Company, Incorporated (“THC”), of which Kanter’s family trusts owned substantial stock, had a 21.375% interest. Connolly had a 5% interest.

The partnership received income by way of fee-sharing agreements with MHM and GHM, whereby each contributed a certain percentage of the fees earned on its management contracts. These fee-sharing agreements were continually adjusted so that MHM and GHM contributed roughly the same amounts. Also, the total fees that MHM contributed roughly equaled its distributive share of the partnership’s income.

As planned, employees of MHM provided most of the help Connolly needed to run GHM. MHM was not directly paid for these services. Moreover, while IRA and THC received a combined 47.5% of the partnership’s income, they provided no similar assistance to Connolly. However, Eulich considered his participation in the Essex Partnership to be a tool by which MHM might obtain contracts for large hotels through Kanter and gain experience and expertise in operating a large hotel through aiding Connolly.

In 1986, Eulich sold MHM to Aircoa, an unrelated company. Despite this change in ownership, MHM remained a participant in the Essex Partnership arrangements until approximately 1990.

Judge Couvillion found that the Essex Partnership arrangements were not indicative of a kickback scheme. He found that Eulich and MHM participated in the partnership solely in hopes of receiving Kan-ter’s aid in obtaining contracts for large hotels. He also noted that an unrelated company continued to contribute a percentage of MHM’s fees to the Essex Partnership until approximately 1990.

Judge Haines stated that Judge Couvil-lion’s account of the facts was incomplete or inaccurate and added the following. Ballard and Lisle were instrumental in Prudential’s decision to offer Connolly the Gateway management contract. When Prudential relayed this offer, but conditioned it on Connolly establishing his own management company, Ballard realized that Connolly would need assistance. Accordingly, Ballard introduced Eulich to Connolly as someone who could provide the support services Connolly needed. Thereafter, Eulich, most likely with Kan-ter’s assistance, organized GHM for Connolly. Ballard and Lisle also were instrumental in Prudential’s awarding GHM the Midland management contract.

At approximately the same time as establishing GHM for Connolly, Eulich and Kanter formed the Essex Partnership. Although Connolly was a partner, he was not involved in its formation and remained largely ignorant of its operations and functions. At trial, he did not know the identity of the partners, believed he was offered a 5% interest in the partnership in exchange for referring to MHM any contracts that GHM could not handle, and did not remember that a portion of GHM’s fees were paid to the partnership.

In the years that followed, MHM provided most of the services needed to run the Gateway and Midland. Connolly’s partic ipation was akin to that of ah on-site manager. Connolly even received a salary from the Gateway. Indeed, Eulich considered the Gateway a part of MHM’s management portfolio, even though the contract for this hotel technically was awarded to Connolly and GHM.

In 1984, IRA transferred to Carleo, TMT, and BWK its 26.125% interest in the Essex Partnership. Carleo and TMT each received an 11.75% interest, and BWK received a 2.6125% interest.

In concluding that the aforementioned arrangements were part of a kickback scheme, Judge Haines reasoned that the Essex Partnership was a conduit through which Ballard, Lisle, and Kanter received portions of the management fees paid by Prudential for the Gateway and Midland. Judge Haines explained the following. Ballard and Lisle arranged for Prudential to award Connolly the Gateway and Midland management contracts and for Eulich to set up GHM to “make it appear as if Connolly was the owner of a hotel management company.” GHM, however, was a sham corporation and Connolly merely a pawn. MHM employees did all the work, which suited Eulich because he wanted the contacts and experience. And while MHM’s contribution to the partnership equaled its distributive share and Connolly had only a minor distributive share, Kan-ter’s entities took nearly half of the partnership’s income. Finally, Kanter later transferred part of his interest to Ballard’s TMT, Lisle’s Carleo, and his own BWK, so that Ballard and Lisle could receive part of the fees.

Contrary to Judge Haines’s account of the matter, Eulich testified at trial that the Essex Partnership was his and Kanter’s idea. Kanter wanted to get involved with hotel management, and Eulich wanted to develop a relationship with Kanter because Kanter knew the owners of several major hotels. They invited Connolly to become involved because he had the expertise to manage big hotels, which MHM lacked. Eulich believed that Kanter helped arrange for the Essex Partnership partners to get the Gateway Hotel management contract. Eulich described the Gateway Hotel as the biggest hotel with which Essex Partnership was involved.

Such testimony, if believed, is certainly plausible. Again, Judge Couvillion is the one who saw and heard the witness and found the testimony credible.

to. General Reasoning

Judge Couvillion generally reasoned that the Commissioner had not proved the existence of a kickback scheme because none of the witnesses at trial stated that such a scheme existed and various witnesses associated with the Five expressly denied paying kickbacks. Moreover, Ballard and Kanter had testified that they were not engaged in a kickback scheme.

Judge Haines concluded that Judge Couvillion’s reliance on the Five’s denials of paying kickbacks to Ballard, Lisle, and Kanter was misplaced. Judge Haines reasoned that the Commissioner had argued that Ballard, Lisle, and Kanter did not disclose their scheme to the Five, such that their denials of its existence was not dispositive of the matter. Judge Haines stated, therefore, that Judge Couvillion clearly erred by giving too much weight to these denials in reaching his ultimate conclusion. How much weight was to be given the testimony of witnesses, however, was not up to Judge Haines. That was the function of Judge Couvillion.

C. Distribution of Payments from the Five to the Parties

i. Loans and Consulting Fees

Judge Couvillion noted the Commissioner’s argument that Kanter used loans and consulting payments to distribute to Ballard and Lisle the funds that the Five paid to IRA. Regarding loans, Judge Couvil-lion found the following. Ballard and Lisle each established two grantor trusts for the purpose of investing as limited partners in movie shelter partnerships. Each trust financed the acquisition of its partnership interest with a loan from IRA. Each trust gave IRA a promissory note in return. IRA transferred these notes to International Films, Incorporated (“IFI”), a corporation of which IRA was the majority shareholder. IFI gave IRA promissory notes in return. Although Kanter later claimed that he helped the trusts obtain their partnership interests because the investments seemed promising, the movie ventures ultimately proved unsuccessful. Because Ballard and Lisle were not personally liable for the trusts’ loans, and because the trusts had no other assets, the notes from the loans were non-collectible in light of the venture’s failure.

At some point, IFI loaned Ballard and Lisle money as individuals. Ballard and Lisle gave IFI personal promissory notes in return.

In 1987, IRA attempted to collect on IFI’s promissory notes. IFI lacked sufficient money to pay the notes, and IRA agreed to accept all of IFI’s assets instead. These assets mainly consisted of IFI’s stock, the trusts’ promissory notes, and Ballard’s and Lisle’s personal promissory notes. IRA sold the trusts’ notes to an unrelated corporation for $1 each. That corporation’s president later testified that he was a longtime friend of Ranter’s and agreed to purchase the trust’s notes, which were essentially worthless because the movie ventures had failed, as a “favor” to Kanter. IRA forgave Ballard’s and Lisle’s personal notes. IRA sold IFI’s stock to an individual for $1 per share. Then, on its tax return, IRA claimed a loss with respect to its sale of the trusts’ notes to the unrelated corporation, deductions with respect to Ballard’s and Lisle’s personal notes, and a deduction with respect to IFI’s stock.

After Kanter had become IRA’s president in 1989, and the IRS had begun investigating Ballard, Lisle, and Kanter, Kanter sought to collect on Ballard’s and Lisle’s personal promissory notes. Ballard’s notes totaled approximately $196,000. In 1992, Ballard agreed to pay $120,000 in settlement of this debt.

Regarding consulting fees, Judge Cou-villion found that, beginning in 1983, KWJ Corporation and, later, KWJ Company, paid monthly “consulting fees” of $1,000 each to Ballard’s two daughters and Lisle’s son and daughter. In 1988, KWJ Company terminated the fee arrangement as to one of Ballard’s daughters. After Kanter had become IRA’s president in 1989, and the IRS had begun investigating Ballard, Lisle, and Kanter, Kanter terminated the fee arrangement as to the remaining children. He sent them letters explaining that the arrangements had been made by a different IRA president, Lawrence Freeman, at a time when KWJ Corporation existed as IRA’s subsidiary, and should have been terminated long ago, as the children had done nothing to earn the fees for many years. Judge Couvillion also noted that Ballard and Lisle did not receive salaries for managing the assets of TMT and Carleo, respectively, until 1990, when Kanter arranged for TMT to pay Ballard a salary and Carleo to pay Lisle a salary.

Judge Couvillion found that the evidence presented at trial did not prove that the loans and consulting payments were a disguised means of transferring to Ballard and Lisle their portions of the Five’s payments. Judge Couvillion acknowledged that the loans to the trusts were “exceptionally favorable.” Judge Couvillion also stated that he did not accept Ranter’s testimony that IRA made these loans be cause the movie ventures seemed promising, reasoning that an arm’s-length lender would have at least required that Ballard and Lisle personally guarantee the loans. Judge Couvillion found, however, that Kanter considered Ballard and Lisle extremely valuable business contacts and may have arranged the favorable loans in recognition of this advantage and their friendship. Judge Couvillion reasoned that the consulting payments also may have been arranged by Kanter as favors to Ballard and Lisle or as compensation to Ballard and Lisle in lieu of salaries.

Judge Haines concluded that Judge Couvillion’s discussion of the loans and consulting fees was incomplete. Indeed, Judge Haines stated that “there is little indication [that] any consideration was given to [the Commissioner’s] argument that these loans were shams.” Accordingly, Judge Haines added the following. Regarding loans, neither Ballard, Lisle, nor their grantor trusts were required to pay interest to IFI or IRA on their loans. Also, in 1987, Ballard reported more than $2.4 million in income on his tax return, such that he had the resources to repay either IFI or IRA for his personal loans and the loans given his trusts. Lisle likewise had the resources to repay his and his grantor trusts’ debts. Regarding consulting fees, the money that KWJ Company paid to Ballard’s and Lisle’s children came from loans from IRA totaling $250,000. Furthermore, in 1982, Ballard received from IRA a $12,500 director’s fee, even though, as Ballard had testified at trial, he never served as a director of IRA.

Judge Haines stated that these loans and payments were one of Kanter’s methods of getting the money earned by Ballard, Lisle, and Kanter to those parties. Judge Haines reasoned that the facts that Ballard and Lisle were not charged interest on their loans or ever forced to repay them, even though they had the financial ability to do so, demonstrated that the loans were shams. Judge Haines reasoned that Kanter’s attempts to collect the loans and letters terminating the consulting fees were merely post-investigation “window dressing.” Regarding Haines’s suggestion that Ballard received a portion of the kickback money through a director’s fee, Ballard testified at trial that he received the $12,500 payment for consulting services, rather than as a director’s fee.

Once again, there is a clear difference in the roles of Judge Couvillion and Judge Haines. Judge Couvillion made certain findings based upon the testimony of live witnesses; Judge Haines drew a different conclusion based upon a cold record. Accepting the witnesses statements regarding “consulting fees” and the loans at issue, there is nothing manifestly unreasonable about the findings of Judge Couvillion.

ii Transfers of IRA’s Income

Aside from the above discussion of loans and consulting fees paid to Ballard, Lisle, their trusts, and their children, Judge Cou-villion did not address the Commissioner’s so-called “flow of funds” argument. In this argument, the Commissioner analyzed IRA’s distribution of its income and transfers of its interests in KWJ Corporation and Essex Partnership and argued that, by way of these distributions and transfers, the money from the Five made its way to Ballard, Lisle, and Kanter.

Judge Haines concluded that it was error not to address this argument. Accordingly, Judge Haines added the following facts. By 1983, IRA had accumulated $5,056,929 in payments from the Five. 8 IRA reported this income on its consolidated tax returns.

In 1984, Kanter recommended that TMT, Carleo, and BWK be removed from IRA’s consolidated group of corporations. Kanter testified that he was concerned that Carleo’s investments might interfere with certain deductions normally claimed by IRA. Kanter also testified that he wanted to shelter Ballard and Lisle from any “second-guessing” by IRA officers.

Also in 1984, Kanter directed that IRA’s “free cashflow,” including its accumulated assets and future income, be distributed in the ratio of 45% each to TMT and Carleo and 10% to BWK. Accordingly, approximately $4.2 million of the more than $5 million that IRA had accumulated was transferred to TMT, Carleo, and BWK. These subsidiaries called the money “capital contributions” and reported it on their tax returns. During this period, IRA also held large amounts of cash that were invested in certificates of deposit, rather than allocated to TMT, Carleo, and Kan-ter.

At trial, Kanter testified that the allocation of IRA’s free cashflow would allow for diversification of IRA’s investments. Specifically, he testified that Ballard was to invest TMT’s money primarily in real estate, Lisle was to invest Carlco’s money primarily in municipal bonds, and Kanter was to make assorted investments with BWK’s money.

Along with this $4.2 million influx of money, TMT, Carleo, and BWK each also received a portion of the future payments made by the Five. This income totaled more than $4 million. 9

Once the money was in TMT substantial portions were invested in certificates of deposit, savings and money market accounts, and a brokerage account. The Ballards had signatory authority over one of these savings accounts.

TMT did make loans to, and investments that benefitted, Ballard and his family.

From 1984 to 1989, TMT made loans to Ballard equaling $146,943 and to Mary Ballard equaling $160,000. While the Bal-lards agreed in 1986 to repay the $160,000 loaned to Mary Ballard, there is some question concerning whether or not it was repaid in accordance with the agreement. During these years, TMT also loaned $135,155 to Seabright Trust, which benefit-ted Ballard’s children, and $41,520 to Sea-bright Corporation, which was owned by a trust that also benefitted Ballard’s family. As of trial, these loans to Seabright Trust and Corporation had not been repaid.

In 1986, TMT also loaned $4,000 to Fi-com International, Incorporated (“Ficom”), a corporation organized by Melinda Ballard. Later, in 1988, TMT invested $15,000 in Ficom. TMT later claimed a loss on its tax return for the $15,000 investment.

TMT likewise transferred to Ballard shares of stock in Fairfield Planting Company (“Fairfield”). In return for these shares, Ballard executed a promissory note to TMT for $100,000 and the right to receive 90% of Fairfield’s dividends. Ballard secured this note by pledging 1,000 shares of Fairfield common stock. There is some question as to whether or not TMT received any payments with respect to this note.

After Ballard began working at Goldman Sachs, he was required to submit disclosure statements informing the firm of outside business involvement. In a 1988 disclosure statement, Ballard indicated that he was involved in farming operations individually and through two corporations owned by family trusts. At trial, Ballard testified that the corporations to which he referred were TMT and Fairfield. Ballard also indicated that he served as director for a friend’s unrelated corporation.

Judge Haines concluded that the transfers of IRA’s cash flow to TMT, Carleo, and BWK were another of Kanter’s methods of getting the payments earned by Ballard, Lisle, and Kanter to those parties for their personal use and enjoyment. In reaching this conclusion, Judge Haines reasoned that Kanter removed TMT, Carl-eo, and BWK from IRA’s consolidated group of corporations to “reflect the reality that those corporations were owned and controlled by” Ballard, Lisle, and Kanter. Judge Haines stated that Ballard’s use of TMT money to make loans to Ballard and his family, trusts, and corporations, and subsequent failure to repay these loans, demonstrated self-dealing and Ballard’s treatment of TMT as a personal pocketbook, or alter ego. Judge Haines also stated that Ballard’s failure to disclose to Goldman Sachs that he managed TMT demonstrated that he was not managing TMT for IRA, but rather was completely in control of TMT.

Judge Haines also reasoned that Kan-ter’s testimony on these matters was not credible. Specifically, Judge Haines stated that Kanter’s explanation that he de-consolidated IRA because certain of Carlco’s investments might imperil IRA’s deductions was not credible because it did not apply to TMT and BWK. Judge Haines noted that there was no indication that TMT’s or BWK’s investments imperiled IRA’s deductions. Also, Judge Haines stated that Kanter’s explanation that he deconsolidated IRA because he wished to save Ballard, Lisle, and himself from second-guessing by IRA officers was not credible because TMT, Carleo, and BWK remained subject to these officers’ directions as IRA subsidiaries.

Judge Haines further stated that Kan-ter’s explanation that the 45%-45%-10% split was an effort at diversification was not credible because the only assets transferred to TMT, Carleo, and BWK were traceable to the Five and IRA held other large sums of money that it invested in certificates of deposit rather than distribute. Judge Haines reasoned that the 45%-45%-10% transfer to TMT, Carleo, and BWK was part of preexisting agreement between Kanter, Ballard, and Lisle.

Judge Haines further reasoned that Ballard’s testimony on these matters was not credible. Specifically, Judge Haines stated that Ballard’s testimony that Fairfield was a family-trust owned corporation through which he was involved in farming operations lacked credibility, as Ballard, rather than a family trust, owned Fairfield.

With regard to Judge Haines’s statements that Ballard had not repaid his Fairfield loan, Ballard testified at trial that he borrowed approximately $200,000 from TMT to purchase Fairfield. He acknowledged that the loan remained outstanding, but stated that Kanter refused to accept repayment. Ballard explained that the loan was worth more to TMT so long as it remained unpaid, given that it was accompanied by a right to receive 90% of Ballard’s distributions from Fairfield. Indeed, Ballard stated that the deal turned out bad for him, but good for TMT.

Only one familiar with the use of corporations, partnerships and trusts, as well as the complexities of our tax laws could evaluate the transactions involved. Judge Couvillion did just that knowing the background of Kanter and the legality or illegality of these moves. While either finding may be plausible, we conclude nothing is clearly erroneous about the findings made by Judge Couvillion.

D. Judge Couvillion’s and Judge Haines’s Conclusions

Judge Couvillion found that Ballard had not earned money from the Five through a kickback scheme and had not attempted to skirt the IRS’s personal income tax paying requirements by assigning this income to TMT. Judge Couvillion likewise found that Ballard was not liable for additions to tax as a result of fraud. To this end, Judge Couvillion especially noted that taxes were paid on all of the income in question and that none of the examining IRS agents had recommended this addition to tax, despite having conducted a lengthy examination of the case with full access to all of the evidence later presented at trial.

Judge Haines concluded that Ballard had used his position at Prudential to earn kickbacks from the Five. Judge Haines also concluded that Ballard attempted to assign this income to either IRA or TMT. Judge Haines further concluded that TMT was Ballard’s alter ego, as Ballard had made personal use of its assets, so that Ballard was responsible for the income tax on this money. Judge Haines finally concluded that Ballard was liable for additions to tax for fraud because he failed to report this money, despite knowing that it was personal income from the Five, and agreed with Kanter to disguise the income by funneling it through sham corporations as loans and other payments.

V Final Analysis

Again, the issue now before us is whether Judge Haines’s gave the appropriate deference to Judge Couvillion’s report or clearly erred in concluding that Judge Couvillion’s findings were manifestly unreasonable. After carefully reviewing both Judge Couvillion’s report and Judge Haines’s opinion, we conclude that Judge Couvillion’s findings were not clearly erroneous. Rather, Judge Haines simply failed to give Judge Couvillion’s findings of fact and credibility determinations the deference required under Rule 188 and our instructions on remand. See Matter of Multiponics, Inc., 622 F.2d at 712-13, 722; Tax Court Rule 183(d); Ballard II, 429 F.3d at 1031, 1032.

A. Kickback Schemes

We conclude that Judge Couvillion’s findings that Ballard did not earn money by giving the Five business in exchange for kickbacks and, therefore, did not attempt to fraudulently assign his income to IRA or TMT, was supported by the evidence and not manifestly unreasonable. Specifically, Judge Couvillion’s finding that the Hyatt payments were not indicative of a kickback scheme was plausible given the fact that Hyatt received the Embarcadero Hotel contract because Pritzker made the only bid and, therefore, did not need Ballard’s and Lisle’s help. For Judge Couvil-lion’s finding to be clearly erroneous, the record would need to show that Ballard and Lisle somehow convinced Intercontinental and Del Webb not to bid. The Commissioner did not argue this and Judge Haines did not include it in his findings and the evidence does not suggest it. Rather, Del Webb’s executive stated at the bidding that he would not participate because someone prematurely promised him the contract. It seems unlikely that he would attend the bidding and then concoct this story, when he simply could have declined to show like Intercontinental’s representatives. Also, Judge Couvillion’s finding is plausible given that Ballard testified at trial that Lisle allowed Hyatt’s bid, not because of some promise to receive a kickback, but because Weaver explained to him the advantage of doing so, namely that having the Embarcadero Hotel might dissuade Hyatt from building a competing hotel nearby.

The fact that Weaver sent Kanter one of Hyatt’s payments with a letter referring to “the participants” does not persuade us that Judge Couvillion was required to find that Ballard and Lisle always were meant to receive part of the payments. The record shows that, by this point, IRA owned KWJ Corporation and remitted a portion of the payments to Weaver, so that “the participants” could have been KWJ Corporation and Weaver. Weaver may not have named these participants outright because he and Kanter had not disclosed to Hyatt that Weaver sold KWJ Corporation. Likewise, the fact that Kanter sent a KWJ Company distribution to Ballard and Lisle, rather than TMT and Carleo, does not persuade us that Ballard and Lisle earned the Hyatt payments or that TMT and Carleo were their alter egos. Kanter viewed Ballard and Lisle as TMT’s and Carlco’s managers. As such, they would have been the gatekeepers of TMT’s and Carlco’s money. Also, Kanter apparently corrected the error by voiding those checks.

Furthermore, Judge Haines’s rejections of Ballard’s and Lisle’s testimony on this matter were either beyond his power as a reviewer, rather than fact-finder, or were based on inaccurate facts. For instance, Judge Haines stated that Ballard’s testimony was not credible because it was self-serving, uncorroborated, or simply “feigned.” However, defendant testimony is most often self-serving and uncorroborated and, without having heard it, Judge Haines could not say that any testimony was “feigned,” especially when the factfin-der reached an opposite finding. Judge Haines also stated that Ballard’s testimony was not credible because he gave contradictory accounts of whether he had met Weaver. The record shows, however, that Ballard never denied meeting Weaver and that it was the questioning attorney who did so. Judge Haines likewise stated that Kanter’s testimony was not credible because Kanter claimed that Weaver sold KWJ Corporation for the money when Weaver actually did not receive any money from the sale for four years. The record shows, however, that Kanter did not indicate that Weaver needed money immediately and explained that Weaver’s other impetus for selling was a dispute with Pritzker.

Judge Couvillion’s finding that the Frey payments were not indicative of a kickback scheme was plausible and, therefore, not clearly erroneous, given that Frey’s offer to purchase the first apartment complex in question was accepted by Prudential because he offered more than its appraised value. Ballard’s only involvement with the sale was advising Prudential to accept the offer when Prudential’s fiduciary duty already arguably required accepting the offer. Moreover, this sale, like all of Frey’s future dealings, was done with an unrelated Miami Prudential employee, rather than Ballard or Lisle.

In reaching the conclusion that the transaction was indicative of a kickback scheme, Judge Haines merely cited Ballard’s advice to accept the deal and Frey’s and Prudential’s future relationship. Judge Haines’s reasoning is insufficient to persuade us that Judge Couvillion committed clear error. Basically, Judge Haines stated that because Ballard did his job and Frey continued to do business with one of the largest financiers in the country, Ballard and Lisle must have agreed with Kan-ter to direct business to Frey in exchange for kickbacks. Judge Haines considered the same facts as Judge Couvillion, but reached an entirely different outcome, without demonstrating why Judge Couvil-lion’s finding must have been wrong. This was beyond Judge Haines’s power when reviewing the findings of a Special Trial Judge.

Judge Couvillion’s finding that the Sehaffel payments were not indicative of a kickback scheme was not clearly erroneous. Nothing in the record shows that Schaffel’s and Kanter’s fee-sharing agreement was anything beyond Sehaffel paying Kanter for introducing him to Ballard and Lisle, which gave Sehaffel an entree to obtaining business with Prudential. Also, nothing in the record suggests that Schaf-fel’s entree was anything beyond that which normally accompanies an introduction.

The fact that Ballard met with Sehaffel and an IBM representative shortly after this introduction and then advised Prudential to purchase IBM’s headquarters does not persuade us that Judge Couvillion’s finding was manifestly unreasonable. Ballard simply could have been doing his job. Likewise, the fact that Ballard and Lisle had the power to influence Prudential is insufficient to persuade us that Judge Cou-villion was required to find that they did, in fact, exercise this power. Furthermore, although Kanter suggested in his letter to Sehaffel that their fee-sharing agreement extended to Travelers deals because it was inextricably linked to Lisle, these arguments are insufficient to prove something untoward. If Sehaffel was paying Kanter as thanks for having made the introduction, then the fee-sharing agreement understandably might extend to Lisle’s future employer.

Moreover, Judge Haines’s reasoning that Schaffel’s projects with Prudential after Ballard and Lisle left the company were an “outgrowth” of the earlier introduction actually supports Judge Couvil-lion’s finding. If the payments were kickbacks for Ballard and Lisle directing Prudential business to Sehaffel, then they would have stopped when Ballard and Lisle could no longer do so.

Judge Couvillion’s finding that the PMS transactions were not indicative of a kickback scheme is supported by the record. Schnitzer did not need help obtaining Prudential conversion business, but sold 47.5% of PMS to IRA because he needed help obtaining business from other companies. The record upholds Judge Couvillion’s reasoning. Schnizter testified at trial that he had no interest in Kanter bringing in more Prudential business because he already had developed a relationship with Prudential. That Schnitzer developed this relationship by preparing standardized operating reports, which he knew would please Prudential, demonstrates that Schnitzer’s Prudential business did not involve Ballard’s or Lisle’s maneuvering.

The fact that Schnitzer’s business with Prudential continued to increase after IRA purchased a portion of PMS is insufficient to persuade us that Judge Couvillion’s account of the matter was erroneous. This continued growth could have been an extension of the previous growth. Also, the fact that Schnitzer conferred with Ballard before selling a portion of PMS to IRA is insufficient to show that Ballard agreed to give PMS more business in exchange for kickbacks. Schnitzer testified that he conferred with Ballard only to determine whether Kanter could deliver on his promise of bringing in more business. Judge Couvillion evidently found this testimony credible, and the testimony is neither contradicted by extrinsic evidence nor internally inconsistent. See Anderson, 470 U.S. at 573-74, 105 S.Ct. at 1511-12.

Finally, Judge Couvillion’s finding that the Essex Partnership transactions were not indicative of a kickback scheme was not clearly erroneous. A review of the facts suggests that both Judge Couvillion’s and Judge Haines’s accounts were plausible. As Judge Couvillion concluded, Kan-ter may have become involved with the Essex Partnership simply as a hotel management venture. Eulich may have become involved with the partnership, and been willing to do most of the work, for the proximity to Kanter and the experience. Indeed, Eulich testified that he got involved because he wanted to gain Kan-ter’s influence in the hotel industry. In light of this, the fact that Kanter’s organizations contributed little to the partnership and that Eulich’s MHM realized little profit from the partnership does not necessarily show that it was a way of disguising kickbacks. Lastly, Connolly simply may have been happy to receive the assistance in running GHM.

As Judge Haines concluded, the partnership may have been designed by Ballard, Lisle, and Kanter as a way of pocketing Prudential’s management fees on the Gateway and Midland. Ballard and Lisle may have directed Prudential contracts to Connolly, whom Prudential already liked and who had experience working at large hotels, and then used Eulich to do all of the work. However, when both scenarios are plausible, the scenario described by the factfinder deserves deference. See id. at 575, 105 S.Ct. at 1512. Accordingly, although we find Judge Haines’s conclusion on this matter possible, this is insufficient to show that Judge Couvillion’s finding was manifestly unreasonable.

In general, we find that Judge Haines overstepped the bounds of his review in concluding that Judge Couvillion gave too much weight to the Five’s testimony that they were not paying kickbacks. Judge Haines did not state, and the record does not illustrate, that this testimony was not credible. The weight given to credible testimony was entirely within Judge Cou-villion’s discretion as the factfinder. See id.

Also, we generally note that Judge Haines rarely could point to concrete evidence proving his account of the transactions and frequently stated that his conclusions were based on only inferences from the circumstances. For Judge Haines to have properly found that Judge Couvil-lion’s findings were clearly erroneous, he would have needed to cite, and the record would have needed to provide, evidence that these findings were not plausible. Inferences, based largely on the same facts considered by Judge Couvillion but entire ly opposite of Judge Couvillion’s findings, were insufficient.

We likewise note Judge Haines’s conflict with Judge Couvillion’s use of phrases such as “Kanter and/or entities associated with him” when describing with whom the Five agreed to do business. Judge Haines repeatedly emphasized that the Five agreed to do business solely with Kanter and that no one at IRA did anything. Judge Haines used this finding to argue that Kanter, Ballard, and Lisle were the true earners of the money and fraudulently attempted to assign their income to IRA and the other entities. We acknowledge that Judge Haines’s finding that Kanter did the work may demonstrate that Kanter earned, and attempted to assign, the money. However, this finding does not prove the same of Ballard, absent evidence that Ballard helped Kanter do the work to earn the money. 10 Since Judge Couvillion’s finding that Ballard did not sell his influence for kickbacks was not clearly erroneous, as discussed above, we cannot reach this conclusion.

B. Flow of Funds

We find that Judge Couvillion did not clearly err in finding that the loans and consulting fees paid to Ballard and his family were not a means of disguising income. It is plausible that IRA lent Ballard’s and Lisle’s trusts the necessary funds to invest in movie shelter partnerships under such favorable terms simply because Kanter highly valued, and wished to maintain, his friendship with Ballard and Lisle. This friendship had brought IRA a significant monetary advantage. For example, Schaffel paid IRA a percentage of his mortgage broker fees simply because Kanter had introduced him to Ballard and Lisle.

The fact that Ballard’s and Lisle’s trusts were not charged interest on these loans, or that Ballard and Lisle were not charged interest on their personal loans, is insufficient to persuade us that Judge Couvillion was required to find that the loans were shams. Rather, this fact may have been one of the favorable conditions that Ballard’s and Lisle’s valuable influence brought them. Likewise, the fact that Ballard and Lisle were never forced to repay their trust’s loans does not persuade us that Judge Couvillion clearly erred, as Ballard and Lisle had not personally guaranteed the loans. Regarding Ballard’s and Lisle’s personal loans, Ballard later was asked by Kanter to repay these loans. Judge Haines’s conclusion that this request was “window dressing” was unsubstantiated.

Also, it is plausible that KWJ Corporation, and later, KWJ Company, paid Ballard’s and Lisle’s children consulting fees as another favor, or in lieu of salaries, to Ballard and Lisle. Nothing in the record contradicts this finding. It also is plausible that the children originally did work to earn these payments and that the payments continued after this work lapsed simply as an oversight by Freeman, as Kanter’s termination letters suggested. Judge Haines’s conclusion that these letters were merely “window dressing” also was unsubstantiated.

We also conclude that Judge Couvillion did not clearly err in declining to further analyze the Commissioner’s flow-of-funds argument. Because he determined that Ballard had not earned income by accepting kickbacks, and this finding was not manifestly unreasonable, Judge Couvillion had no reason to determine whether the alleged income made its way to Ballard.

We nonetheless note that Judge Haines’s thorough treatment of the flow-of-funds argument does not convince us that Judge Couvillion’s finding concerning the scheme was manifestly unreasonable. The fact that Ranter transferred to TMT, Carleo, and BWR money received from the Five does not necessarily show that Ballard and Lisle earned that money. A finding that Ballard was the true earner of 45% of the payments because his corporation ultimately received 45% of the payments is circular. Likewise, Judge Haines’s attack on Ranter’s explanation for the allocation involves leaps of logic. Although Judge Haines accurately stated that Ranter’s reasoning that Carlco’s investment might imperil IRA’s deductions did not apply to TMT or BWR, this does not necessitate a conclusion that TMT, Carleo, and BWR were the parties’ alter egos. Rather, Ranter may have wished for TMT, Carleo, and BWR to maintain similar tax status. He was an acknowledged expert in our tax laws. Also, although Judge Haines accurately noted that all of the money allocated per the 45%-45%-10% split was traceable to the Five and that IRA had other money that was not allocated, this does not necessitate a finding that the allocation was a method of getting kickback income to anyone. Rather, Ranter simply may have chosen one of its sources of income to fund the allocation. In sum, Judge Haines attributed too much to these facts.

Also, Ballard’s treatment of the money allocated to TMT does not necessarily show that he was making personal use and enjoyment of money gained through a kickback scheme. While TMT made loans to Ballard and his family that may not have been repaid, this only demonstrates that Ballard may have been an irresponsible manager. It does not show that Judge Couvillion clearly erred in finding that Ballard did not earn the money or commit fraud. Moreover, Ballard testified that he attempted to repay the loan made to purchase Fairfield. Also, while Ballard failed to disclose his position as TMT’s manager to Goldman Sachs, this only demonstrates that he committed an oversight. It does not demonstrate that Judge Couvillion’s finding that TMT was not Ballard’s alter ego was clearly erroneous. Finally, Judge Haines’s statement that Ballard’s testimony was not credible because he testified that Fairfield was owned by a family trust when he actually owned it personally again assigns too much significance to a potentially innocuous mistake.

VI. Conclusion

For the reasons discussed above, we conclude that Judge Couvillion’s findings of fact and credibility determinations were supported by the record and that his finding that Ballard was not responsible for a deficiency and had not committed fraud were not manifestly unreasonable. We conclude that, in finding otherwise, Judge Haines did not presume Judge Couvillion’s findings to be correct or give Judge Cou-villion’s credibility determinations their due deference. See Tax Court Rule 183. Rather, Judge Haines conducted a nearly de novo review of the facts. This review violated Rule 183’s and our instructions. See Anderson, 470 U.S. at 573-74, 105 S.Ct. at 1511-12. It is clear that this case is a “close call.” Had Judge Haines been the original trial judge, his rulings would probably be entitled to an affirmance. However, he was not the trial judge and did not see or hear the witnesses. Judge Couvillion did and found them credible. It is no surprise that a knowledgeable tax attorney would use numerous legal entities to accomplish different objectives. This does not make them illegitimate. Unfortunately such “maneuvering” is apparently encouraged by our present tax laws and code. The Ballards were charged with fraud. Judge Couvillion heard the wit nesses and reviewed the exhibits and concluded the government simply failed to prove such by clear and convincing evidence. The record fully supports his findings of fact and conclusions. Accordingly, we remand to the Tax Court with instructions to vacate Judge Haines’ opinion and enter an order approving and adopting Judge Couvillion’s original report as the opinion of the Tax Court. 11

REVERSED, VACATED and REMANDED with instructions.

1 . Rule 183(c) of the Tax Court provides in relevant part that "due regard shall be given to the circumstances that the Special Trial Judge had the opportunity to evaluate the credibility of witnesses, and the findings of fact recommended by the Special Trial Judge shall be presumed to be correct.”

2 . Mary B. Ballard is a participant in the dispute only because she filed joint tax returns with her spouse. Accordingly, unless otherwise specified, references to "Ballard” denote Claude M. Ballard. Also, the Ballards are deceased and proceed through their estates.

3 . The Ballards also argue that the Tax Court Judge violated their due process rights by rejecting the Special Trial Judge's findings and determinations on a "cold” record. Given our conclusion that the Tax Court failed to give the Special Trial Judge's report the appropriate deference, we find it unnecessary to address the due process argument.

4 . Although Ballard was joined in the scheme by others, only the case as it pertains to the Ballards is before this Court.

5 .For a complete account of this case’s complex procedural history, we direct the reader to our previous published opinions, Ballard v. Comm'r of Internal Revenue, 321 F.3d 1037, 1043-44 (11th Cir.2003), and Ballard v. Comm'r of Internal Revenue, 429 F.3d 1026, 1030 (11th Cir.2005) (“Ballard II”).

6 . The government seems to argue that we are only reviewing the Tax Court opinion (Judge Haines) for clear error as opposed to reviewing the Special Trial Judge’s (Judge Couvil-lion) recommendations for clear error. We agree to the extent that we are now reviewing the ruling of the Tax Court (Judge Haines) for clear error. Our conclusion in reviewing the Tax Court opinion is that Judge Haines committed clear error when he failed to accord the original findings of fact, credibility of witnesses and conclusions of Judge Couvillion the due deference required under the law.

7 . See Bonner v. Prichard, 661 F.2d 1206, 1209 (11th Cir.1981) (en banc) (adopting as binding precedent in the Eleventh Circuit, all decisions of the former Fifth Circuit announced prior to October 1, 1981).

8 . Specifically, Hyatt’s payments to IRA’s subsidiary, KWJ Corporation, equaled $670,168 after 30% was remitted to Weaver. Frey’s payments to Zeus equaled $882,042. This money was transferred to IRA from Zeus by way of $51,000 paid to IRA allegedly in re payment of an earlier loan, $774,000 paid to IRA in exchange for a note receivable from THC that never was paid, and $13,000 paid to IRA in exchange for another note receivable from THC that never was repaid. Schaffel’s payments to IRA equaled $1,398,626. Schnitzer’s payments towards its $3.1 million purchase price for PMS equaled $1,941,505. Finally, IRA’s distributive share from Essex Partnership equaled $164,588. A review of the exhibits in the record shows that these numbers are correct.

9 . Specifically, by this time, KWJ Corporation had been liquidated and KWJ Company formed, with TMT and Carleo having 45% interests and BWK having a 10% interest, so that these corporations split Hyatt's post-1983 payments according to this ratio. By 1989, the amount split between the corporations equaled $1,103,721. TMT, Carleo, and BWK reported this money on their individual tax returns. Frey continued to pay Zeus as before. This amount totaled $232,263. Zeus's ledgers indicated that rather than transferring this money to IRA as before, it used the money to purchase the stock of an unrelated corporation. IRA reported this income from Frey on its tax return. Because Ballard and Lisle left Prudential at approximately this time, Schaffel made one final payment of $213,750 to IRA. It is not clear whether IRA transferred this money to TMT, Carleo, and BWK according to the ratio. IRA reported this money on its tax return. Schnitzer continued to make payments to IRA. By 1989, these totaled $2,287,191. IRA transferred this money to TMT, Carleo, and BWK according to the ratio. IRA reported this money on its tax return. Finally, by this time, IRA had transferred its interest in Essex Partnership to TMT, Carleo, and BWK, with Carleo and TMT each receiving an 11.75% interest, and BWK receiving a 2.6125% interest. Thus, these corporations split IRA's 26% distributive share of the partnership’s income. By 1989, the amount split equaled $623,867. TMT, Carleo, and Ballard reported this money on their individual tax returns. A review of the exhibits in the record shows that these numbers are correct.

10 . We also note that corporations only operate through employees, agents, or representatives. The same can be said for partnerships, trusts and other legal entities.

11 . We note that, in Lisle v. Comm’r of Internal Revenue, 341 F.3d 364, 367-68 (5th Cir.2003), the Fifth Circuit considered whether the original Tax Court Judge's opinion was clearly erroneous as to Lisle. The Fifth Circuit held that the Tax Court had clearly erred in finding that the government proved that Lisle committed fraud by clear and convincing evidence. Id. at 375-84. The Fifth Circuit also held, though, that the Tax Court had not clearly erred in finding that the government had proven that Lisle was responsible for deficiencies by a preponderance of the evidence. Id. at 384. However, as this opinion was issued before Judge Couvillion’s original report came to light, the Fifth Circuit's analysis was not benefitted by the knowledge that Judge Couvillion found that the parties were not responsible for deficiencies or additions to tax for fraud. Accordingly, this analysis does not alter our own.