PW Enterprises, Inc. v. Kaler

• Court: District Court, D. North Dakota
• Decided: June 18, 2020
• Docket: 3:18-cv-00263
• Status: Unknown

Opinion

IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF NORTH DAKOTA

In re Racing Services, Inc., ) ) Debtor. ) ) ORDER PW Enterprises, Inc. and Robert Carlson, ) ) Case No. 3:18-cv-263 Appellants, ) ) Bankruptcy Case No. 04-30236 vs. ) ) Susan Bala and Kip Kaler, as Chapter 7 ) Trustee for Racing Services, Inc., ) ) Appellees. ) ______________________________________________________________________________

Appellants PW Enterprises, Inc. (“PWE”) and Robert Carlson (“Carlson”) appeal from the Bankruptcy Court’s1 Ruling on Claims (the “Order”), which was issued on November 28, 2018. In re Racing Servs., 595 B.R. 334 (Bankr. D.N.D. 2018).2 This appeal involves two claims: PWE’s Amended Claim3 for over $10.8 million in unauthorized taxes assessed on PWE’s account wagering4 activity through Debtor Racing Services, Inc. (“RSI”), and, similarly, Carlson’s Claim

1 The Honorable Thad J. Collins, Chief Bankruptcy Judge for the Northern District of Iowa, sitting by designation. 2 The Order is also found in the District Court docket at Doc. No. 1-1. Citations to docket entries refer to the District Court docket unless otherwise stated. 3 Even though PWE’s Amended Proof of Claim incorporates PWE’s original claim, references to “PWE’s Amended Claim” in this opinion will refer to only the additional amount of money that PWE claims for unauthorized taxes. 4 Account wagering is a system whereby a player sets up an account with a service provider from which the player makes wagers. The service provider debits out of this account for wagers placed and deposits any winnings back into the account. Account wagering saves players time and transaction costs by eliminating the need for cash or bank transfers for every wager and payout. for over $380,000 in unauthorized taxes assessed on his account wagering through RSI.5 PWE and Carlson appeal the denial of their respective claims, arguing that the Bankruptcy Court committed a number of legal and factual errors in concluding that (1) they failed to show that they were entitled to the money under their respective oral rebating agreements with RSI and (2) they failed to show that RSI would be unjustly enriched if it got to retain the money instead of returning

it to them. For the reasons explained below, the Order is affirmed. I. BACKGROUND A. The Parties and Trial Witnesses Appellee Susan Bala (“Bala”) served as the president and chief executive officer of RSI, the debtor in this action. RSI is wholly owned by RSI Holdings, Inc., the latter of which is wholly owned by Bala. Beginning in 1993, RSI possessed the sole license to engage in North Dakota’s pari-mutuel horse wagering industry6 as a service provider.7 PWE and Carlson are both professional players who used RSI’s services to bet in high volumes on horse racing. Peter Wagner (“Wagner”) is the owner of PWE, and William (Bill)

Wass (“Wass”) is the chief operating officer. David Cuscuna (“Cuscuna”) is another professional gambler who used RSI to engage in horse betting. Cuscuna is not a party to this appeal, but he testified at trial in support of Bala.

5 In the Order, the Bankruptcy Court also ruled on claims made by Susan Bala and on PWE’s original claim, which was primarily for funds remaining in PWE’s wagering account. Those portions of the Order are not at issue in this appeal and thus will not be addressed. 6 Pari-mutuel wagering is a form of wagering in which players bet against each other rather than against the “house.” 7 At the time of the wagering involved in this case, North Dakota law defined a “simulcast service provider” as “a person engaged in providing simulcasting services to a simulcast operator and establishing, operating, and maintaining the combined pari-mutuel pool.” N.D. Admin. Code. § 69.5-01-11-01(13) (1990). Simulcasting refers to broadcasting the signal of a horse race to another location for the purposes of off-track betting on the race. Kip Kaler (“Trustee”) has been the Chapter 7 bankruptcy trustee of RSI since June 2004. B. Procedural History In 2004, RSI filed for bankruptcy after both it and Bala had come under federal investigation and indictment for illegal gambling and money laundering.8 PWE obtained derivative standing to bring an adversary proceeding against the State of North Dakota (the

“State”) on behalf of all of RSI’s creditors. PWE challenged whether the State had the authority to tax account wagering and sought the return to the bankruptcy estate of all money collected as taxes as well as the denial of the State’s claim for taxes that the State believed RSI still owed. Ultimately, in 2015, the Eighth Circuit determined that North Dakota law did not authorize taxes on account wagering and remanded to the Bankruptcy Court to calculate the amount of unauthorized taxes the State had to return to the estate. PW Enters., Inc. v. North Dakota (In re Racing Servs.), 779 F.3d 498, 507 (8th Cir. 2015). In short, when the North Dakota Legislature amended N.D. Cent. Code § 53-06.2-10.1 to authorize account wagering, it failed to correspondingly amend the takeout formulas in N.D. Cent. Code § 53-06.2-11 to include taxes for

account wagering. Id. at 502. Upon remand, PWE and the Trustee entered into a settlement with the State for the return of $15.872 million to the bankruptcy estate. The Bankruptcy Court approved the settlement over Bala’s objection.9 With the return of millions of dollars to the estate, PWE amended its original claim and Carlson filed his claim. Bala objected to both claims. The Trustee joined her objection to Carlson’s Claim and partially joined her objection to PWE’s Amended Claim.

8 RSI and Bala were convicted on several counts, but in 2007, the Eighth Circuit reversed their convictions because of insufficient evidence. United States v. Bala, 489 F.3d 334, 343 (8th Cir. 2007). 9 More specifically, Bala argued that the State owed the estate a greater amount of money. In January 2018, PWE’s Amended Claim and Carlson’s Claim came before the Bankruptcy Court for a three-day trial. On November 28, 2018, after post-trial briefing, the Bankruptcy Court issued the Order on appeal here. PWE and Carlson filed timely Notices of Appeal and elected to have their appeals heard by this Court rather than the Bankruptcy Appellate Panel (“BAP”), pursuant to 28 U.S.C. § 158 and Fed. R. Bankr. P. 8005.

C. RSI’s Business Bala first became involved in the horse racing industry in 1988 when she was asked by the State to develop a feasibility study on the horse racing industry in North Dakota. After the state legislature passed legislation allowing off-track betting,10 Bala started a company called Dakota Race Management and obtained the exclusive service provider license in North Dakota. In 1993, she formed RSI, which took over as the licensed service provider. RSI referred to itself not only as a service provider but also as a technology integrator and market developer. The services provided by RSI were wide-ranging and included satellite signals, data links, and betting pool reconciliations.

RSI had two kinds of customers: (1) regular players who would go to a teller and place their bets, and (2) professional players like PWE and Carlson, who bet in significantly higher volumes. Professional players sought—and RSI provided them with—a variety of additional services, such as private betting rooms and tellers, informational support for handicapping and individual wagering, accounting support, technology, and access to a wide menu of tracks. The total amount bet by a player over a specific period through a service provider like RSI is referred to as the handle. RSI would deduct a percentage of the handle—called the takeout or

10 Off-track betting means making a bet at a location other than the racetrack where the race is occurring. retainage—and the remainder of the handle would go into the pari-mutuel pool to be distributed to the winning bettors. The takeout rate is set by statute and varies based on the jurisdiction of the race, the particular racetrack, and the type of bet made. RSI used the takeout to pay a variety of fees and expenses, including track fees,11 tote fees,12 charitable contributions, and taxes. RSI paid money to the State—which the State accepted—for pari-mutuel taxes that RSI believed at the time

that it owed. These taxes varied depending on the pool, but they averaged around 3.7% or 3.8% of the handle. Around 1995, Bala first became aware of rebating in the horse betting industry. Under a rebating agreement, a service provider agrees to pay a portion of its takeout to a player as a bonus incentive on the player’s volume. Rebates provide an incentive for professional players to gamble because whether the bet is a winning or losing one, the player is guaranteed to get some portion of the handle back. Early rebates were fairly unsophisticated, with service providers simply paying flat rates. Flat rate rebates were risky for providers because the takeout percentage for the pools a player bet into would dictate how much of the takeout remained for the provider to pay its

expenses. For example, if a player with a 10% flat rate rebate bet into a pool with a 15% takeout, the provider was only left with 5% with which to operate. If the provider’s expenses were to exceed 5%, the provider would lose money on every bet. Thus, service providers had to develop rebating formulas that enabled them to provide rebates at rates that were competitive in the industry while still keeping enough of the takeout in order to operate in the black.

11 Track fees are fees charged for the ability to place a wager at a specific racetrack. RSI negotiated contracts with tracks annually, which meant the amount of track fees could vary over time. 12 Tote fees are fees paid to a totalizer (or tote) company to lease hardware and software that allows a service provider to communicate wagers to a track. In 1997, RSI entered into its first rebating agreement. That spring, Cuscuna came to RSI seeking a 10% rebate. Cuscuna told Bala that his bets tended to be in jackpot pools that had a higher-than-average takeout of around 24% or 25%. Bala was willing to give Cuscuna a 10% rebate if the takeout on his bets were really that high, but she expressed concern that he would not meet that projected takeout. To allow Cuscuna to achieve his goal while making Bala comfortable

about the takeout for his bets, the two developed what is called the variable rate model. Under the variable rate model, the percentage offered to the player as a rebate depends on the takeout rate of the particular pool that the player bets into. For example, Cuscuna’s agreement with RSI essentially had 13.75% of the takeout going to RSI (before expenses) and the remainder of the takeout going to Cuscuna. If Cuscuna bet into a 24% takeout pool, he would get a 10.25% rebate. If he bet instead into a higher or lower takeout pool, he would receive a higher or lower rebate, respectively. Over time, players could achieve a desired effective rate, the average rebate over all their bets among all the different pools. RSI only offered variable rate rebates. When a player approached RSI seeking a rebating

agreement, RSI would calculate the projected average takeout based on the player’s style of play, ask the player what amount of rebate the player was seeking with RSI or was getting elsewhere, and determine whether it was feasible to meet the desired rebate. Because of constantly changing variables in the industry, RSI’s rebating agreements were all oral with no continuing obligations on either side. The player was not obligated to bet, and RSI was not obligated to accept the player’s bets; however, if a bet was made, RSI was obligated to pay the agreed-upon rebate. D. RSI’s Dealings with PWE Bala met Wagner in early 1999 after the service provider that PWE was playing through terminated PWE’s wagering because of PWE’s use of computer algorithms in determining how to bet. Wagner was looking for a service provider willing to handle PWE’s controversial betting style and willing to give PWE a rebate. Wagner also sought a private betting space, technological support, accounting support, and a large menu of racetracks. PWE’s negotiations with RSI involved Wagner, Bala, and—at least to some extent that is disputed—Raymundo Diaz (“Diaz”), an RSI employee who did not testify at trial.13 Wass was not a party to the negotiations.14

PWE’s prior rebate in St. Kitts averaged out to approximately 11%. According to Bala, Wagner told her about his prior rebate and expressed that he was flexible but was hoping to achieve an 11% rebate again. Bala testified that based on the pools and tracks that Wagner said PWE had been betting on, she projected an average takeout rate of 20.5%. Wagner, meanwhile, testified that he did not indicate a particular percentage goal. He did not remember telling Bala that PWE previously had an 11% rebate but admitted that it was possible that he did so. He believed that he discussed the likely takeout range on PWE’s wagering at least to some extent. Bala testified about her understanding of the variable rebating agreement that was eventually reached. The starting point of the negotiations was the 11% number that Wagner

allegedly gave her along with the projected average takeout of 20.5%. Bala took the 20.5% projected takeout and subtracted the 11% desired rebate, which left RSI with approximately 9.5% of the handle to work with. This 9.5% went to a number of identified expenses—including the assumed pari-mutuel taxes—that varied slightly from week to week, plus a 1% RSI management fee. The 1% fee was part of RSI’s assumptions on its side and a mathematical device to get PWE

13 Bala testified that Diaz was not one of the principals to the negotiations, but because he was in the accounting and finance department, he would have participated in analyzing the numbers involved. Wagner, however, testified that Diaz was involved in the negotiations, perhaps even slightly more than Bala was. 14 Because Wass was not involved in the negotiations between RSI and PWE, his testimony about the rebating agreement was primarily based upon his review of documents that were admitted into evidence. to 11% and RSI to 9.5%. Because the taxes were paid by RSI out of the 9.5%, if they had been eliminated, the extra money would have ended up on RSI’s side of the table. According to Bala, the goal was to get PWE to an effective rebate of approximately 11% over time if PWE bet into the anticipated pools. PWE ultimately determined its rebate percentage based on its betting choices, however, with the rebate amount being driven by the takeout rate.

PWE could achieve a higher (or lower) rebate if its betting deviated from projections. For example, if PWE instead chose to bet into pools averaging a 26% takeout, PWE would achieve a much higher rebate of approximately 16.5% (that is, 26% minus 9.5%). Thus, if PWE’s rebate in a given week or over a longer period were something other than 11%, neither side was being shorted and no reconciliation was required. Wagner contended that Bala’s version of the agreement was inaccurate. According to Wagner, the agreement was that “[s]he got one percent, we paid all expenses and taxes and got what was left.” Wagner repeated some variation of “she got one percent and we got the rest” several times throughout his testimony. He did not, however, explain how and why the two sides

came to this arrangement. Had the agreement been for an 11% effective rebate, PWE would have optimized its computer algorithms to that end; instead, PWE optimized the code based on RSI only getting to keep 1%. Contrary to Bala’s testimony, Wagner asserted that PWE paid the pari-mutuel taxes on its betting, with RSI’s role merely being that of a collection agent passing the money along to the State. In support of this contention, PWE offered into evidence PWE Exhibit 10, a May 2000 email from Wagner to Diaz explaining why PWE left St. Kitts and then The Mirage in Las Vegas to move operations to North Dakota. Twice in the email, Wagner referenced how the move to North Dakota meant PWE now had to pay horsemen’s fees and state taxes. No evidence was presented that anyone at RSI ever responded to this email or otherwise corrected Wagner’s understanding. Both Bala and Wagner agreed that income statements admitted as PWE Exhibit 3 accurately reflect RSI and PWE’s agreement. These statements were prepared weekly by RSI and reflected PWE’s wagering through RSI in North Dakota for the week ending August 5, 2001,

through the week ending August 3, 2003. PWE received a statement for each week that it bet through RSI. Each statement records PWE’s handle, the takeout on the handle, and three categories of expenses—“State and Track Expenses” (i.e., track fees and state pari-mutuel taxes), “TeamMakers Site Expense” (i.e., charitable contributions), and “Racing Operations.” Every week without fail, the statements included a 1% RSI fee in the “Racing Operations” expenses. On each statement, the total expenses were subtracted from the total takeout amount, resulting in a “Net Revenue” line. This “Net Revenue” was PWE’s rebate amount for the week. Bala admitted that the income statements were the only documents that she was aware of that reflected her version of the rebating agreement. She did, however, offer into evidence a

summary spreadsheet, admitted as Bala Exhibit 1, which was created by her counsel for purposes of the bankruptcy proceedings. Over PWE’s 105 weeks of wagering, the effective takeout rate was approximately 20.6%,15 and the total expenses constituted 9.7% of the total handle. The spreadsheet indicates that PWE’s effective rebate over the two years was 11.0%. It appears based on the data from the weekly statements, however, that the actual effective rebate was just short of 11% at 10.924%.

15 This number is not listed on the spreadsheet; however, the percentage is derived from the spreadsheet’s total handle and retainage dollar amounts. According to Bala, the fact that the rebate varied some from week to week but ultimately averaged out to approximately 11% was proof that the model worked. Wagner (and Wass), meanwhile, emphasized the fact that every week RSI got a 1% fee without variation. PWE introduced a number of contemporaneous documents mentioning a 1% RSI management fee to further bolster this point.

It is undisputed that neither Bala nor Wagner contemplated a retroactive change in taxes, let alone discussed what would happen in such event. Wagner testified that RSI and PWE discussed the possibility that taxes would be reduced but admitted that these discussions involved the tax rates going forward. He expected that PWE would benefit if the State reduced taxes in the future. Bala, meanwhile, acknowledged that if taxes went down going forward, RSI would have renegotiated its rebating agreements. According to Bala, if the tax rate went down, RSI would benefit by having more money to work with, which in turn would benefit the players because RSI could offer better rebates. She clarified, however, that RSI would not have given all the benefit of lower taxes to players like PWE because RSI was still a business that needed “to maintain a proper

and competitive share of its revenue.” In 2001, PWE and RSI negotiated a separate rebating agreement at a second wagering location, Nuevo Laredo, just across the international border from Texas. Because the tax rate in Mexico was lower than that in North Dakota, the parties had extra margin to work with. Instead of giving the extra margin entirely to either side, however, they agreed to split it 1/3 to RSI and 2/3 to PWE. According to Wass, PWE agreed to a split because Wagner was willing to assist Bala with additional costs that existed in Nuevo Laredo. On the other hand, Wagner himself apparently made a large initial investment in the Mexican racing operations. E. RSI’s Dealings with Carlson Testimony regarding Carlson’s Claim was much more abbreviated. In 1999, Carlson was looking for a place to get a rebate, so he met with Bala that July. At the meeting, Bala gave Carlson a projection sheet showing his potential rebate at different probable cases of takeout (Carlson Exhibit 2) and a list of takeout rates at different tracks and for different bet types (Carlson Exhibit

3). Bala testified that “virtually, without exception” players would tell her what kind of rebate they were seeking, but she could not remember specifically if Carlson did so. Carlson denied approaching Bala with a specific rebate goal. Carlson and RSI reached an oral variable rate agreement whereby RSI would hold approximately 11% of the takeout and the remainder would go to Carlson. His rebate ranged based on the takeout percentage. For example, if the effective takeout for the week were 24%, Carlson’s rebate was 13%. This agreement is evinced by the projection sheet and weekly income statements that RSI prepared for Carlson (Carlson Exhibit 5). Out of the 11%, RSI would take care of the presumed state pari-mutuel taxes, track fees,

and bookings, and what was left over was for RSI’s fee. According to Carlson, RSI was looking for a fee of approximately 3.5%. None of Carlson’s exhibits show this fee, however, and Carlson admitted that no written evidence existed that RSI agreed to only receive 3.5% as a fee. Carlson believed that he was the one paying the taxes because if he had no handle, there would be no taxes. Carlson testified that the agreement that RSI got to retain 11% was premised on RSI paying the taxes, track fees, and tote fees without a profit and that he would have negotiated a better rebate if there were no tax. Bala denied ever marking up the North Dakota taxes and trying to make a profit off them; rather, the amount listed on the weekly statements was the actual— presumed at least—amount of taxes. Track fees were also a “pass through cost”; what was listed was the actual cost. Sometime in 2001, Carlson complained to Bala about his rebate amount after hearing that his fees were higher than a lot of other players’ fees. Carlson met with Diaz and Bala, who told him that if taxes were lowered, he would receive at least some of the benefit.16 Carlson clarified

on cross-examination that Diaz and Bala made this promise on a prospective basis: If taxes changed in the future, Carlson would realize some type of benefit in the future. This is consistent with Bala’s testimony that she told players who were desiring to renegotiate that “[i]f something changes in the future, we can look at it.” Carlson and RSI never negotiated a retroactive adjustment to his rebate, and the tax rate never changed during Carlson’s wagering through RSI. F. Bankruptcy Order In the Order, the Bankruptcy Court denied PWE’s Amended Claim and Carlson’s Claim for the monies returned by the State as improperly assessed taxes on the players’ account wagering. A summary of the Bankruptcy Court’s findings and conclusions as to each claim follows.

1. PWE’s Amended Claim As an initial matter, the Bankruptcy Court found Bala’s testimony about PWE’s rebating agreement sufficient to rebut the presumption of validity of PWE’s Amended Claim and shift the ultimate burden of persuasion to PWE. Racing Servs., 595 B.R. at 354.17 PWE presented three main arguments to support its claim, which the court addressed in turn: (1) The plain and express terms of the agreement show that RSI was to receive 1% and PWE was to receive the remainder of the takeout after expenses; (2) if the plain and express terms are not clear enough, extrinsic

16 Carlson agreed that it was “just a general statement of you’ll benefit if we get taxes lowered” but understood that to mean that the entire tax savings would go directly to his rebate. 17 The Court will explain the burden-shifting framework later in this opinion. evidence also supports PWE’s claim; and (3) even if PWE’s contract-based arguments fail, PWE is entitled to payment of its claim based on principles of implied or quasi contract and equity. Id. a. Express Contract Terms The court rejected PWE’s express contract argument. Id. at 356. The court found both parties’ interpretations of the agreement “theoretically plausible” but stated that “neither

interpretation offered stands firm on the express terms of the oral agreement alone.” Id. Instead, the court concluded that “[t]he undisputed, existing terms of the oral contract in no way address this situation” of “money paid in taxes years earlier being unexpectedly returned to the RSI bankruptcy estate well after the parties finished their contractual relationship.” Id. The money at issue here “should be viewed as money returned to the potential ‘rebate pool’—which was in fact RSI’s takeout.” Id. The parties based their agreement on how to divide the takeout on the assumption that taxes were a fixed cost of RSI. Id. Ultimately, because RSI and PWE never anticipated the retroactive change in taxes, their agreement did not expressly address—and they never possessed the mutual intention to address—the situation they eventually found themselves

in. Id. at 357. b. Extrinsic Evidence Next, the Bankruptcy Court analyzed the extrinsic evidence under North Dakota’s latent ambiguity doctrine. Id. at 357–58. The court rejected PWE’s latent ambiguity argument, finding that the extrinsic evidence “suggest[ed] that if the parties had actually faced the situation where money otherwise allocated for taxes became available during the course of the contractual relationship, they would have renegotiated their deal.” Id. at 358. Further, RSI and PWE “did not intend to cram new circumstances like this into a formula designed to deal with very different circumstances.” Id. The court based this finding on two facts in the record. Id. First, when RSI and PWE faced different tax circumstances in Nuevo Laredo, they renegotiated and ultimately split the additional margin. Id. According to the court, had the North Dakota agreement been as Wagner testified and any money saved on taxes went to PWE, the parties would have had no reason to make a new deal in Nuevo Laredo. Id. Moreover, “[c]ommon sense suggests that RSI would get a better deal in

North Dakota, where Bala had more time and money invested, and a lesser deal in Nuevo Laredo, where Wagner made a large initial investment and looked to re-coup the start-up money he provided.” Id. at 359. The other fact that influenced the court was Bala’s testimony that the parties would have renegotiated if the taxes were ever declared illegal. Id. The court rejected PWE’s argument that “Wagner’s testimony shows that he intended the agreement to deal with taxes and every other possibility—and PWE should get everything after a 1% fee and RSI’s expenses.” Id. At best, the court wrote, the testimony was evidence of his “secret intention” and was “little more than an attempt ‘to add something to the contract’—which is entirely ineffective extrinsic evidence.” Id. (citations omitted).

The court also rested its conclusions on credibility findings. Id. The court found Wagner’s testimony “mostly not credible,” reasoning that: Wagner did not explain how the terms of the oral agreement were reached, why PWE and RSI structured it the way they did, or why they did not address this specific issue with clarity. He simply repeatedly stated that: “She got 1% and we got the rest.” His testimony did nothing to help the Court understand why the existing agreement should apply in dramatically different circumstances like this where millions of dollars suddenly became available. His testimony seemed to be self-serving and little more. His demeanor, in addition to his actual testimony, further contributed to this finding. Id. On the other hand, the court found Bala’s testimony more credible. Id. The court noted that Bala “provided a full explanation of why things were done as they were, and how they ultimately arrived at their deal.” Id. And unlike Wagner, she “ultimately conceded what [the court concluded] to be the reality of the situation, that the parties most likely would have renegotiated and come to a new deal had they faced this specific issue during the course of their contractual relationship.” Id. The court did find, however, that “Bala’s testimony that the agreement was clear on its face and that RSI should get the money is equally self-serving and lacks some

credibility.” Id. Overall, though, Bala’s testimony was better supported, and her demeanor was more truthful. Id. at 359–60. Having rejected PWE’s express contract and latent ambiguity arguments, the court concluded that PWE failed to meet its burden of proof on its contract-based theories. Id. at 360. c. Implied Contract and Unjust Enrichment The Bankruptcy Court also rejected PWE’s arguments under theories of implied contract, quasi contract, and unjust enrichment. The court quickly disposed of PWE’s argument that PWE had an implied in fact contract with the intent that PWE receive any returned tax money attributable to its bets. Id. The court found that the opposite was true, relying on the parties’ negotiations in

Nuevo Laredo where there were no state pari-mutuel taxes. Id. Next, the Bankruptcy Court jointly addressed PWE’s quasi contract and unjust enrichment theories, noting that the principles of unjust enrichment form the basis for the existence of quasi contracts. Id. at 360–62. The court first rejected PWE’s argument that it was the taxpayer and that RSI simply collected and remitted the taxes on PWE’s behalf. Id. at 362. The court cited a Tax Court case for the principle that in horse betting, taxes and other expenses “discharged from the takeout are expenses imposed upon the track [or here, the provider], not the bettors.” Id. (quoting Lakhani v. C.I.R., 142 T.C. 151, 160 (2014)). The court then addressed PWE’s apparent argument that “even if it was not technically the taxpayer, it essentially supplied the funds to RSI to pay the tax by allowing RSI to claim the tax as an expense in their negotiated agreement.” Id. It followed then, according to PWE, that had the tax not existed, PWE would have gotten more money as part of the deal. Id. The court rejected this argument for several reasons. Id. First, it was not supported by the record. Id. “The parties

did not express their intent for this to happen in the contract or through extrinsic evidence,” and PWE pointed to nothing else that showed it was treated unfairly. Id. Second, PWE could not show that the tax money was a benefit conferred on RSI by PWE. Id. If RSI was enriched at all, the benefit came from an unexpected court decision rather than from PWE. Id. Third, PWE could not show it was impoverished because it got all it bargained for. Id. “RSI is not using a legal technicality to hold on to something that really belongs to PWE.” Id. At most, PWE could show it would have pressed for renegotiations if the tax were eliminated during their business relationship. Id. at 362–63. “Missing out on a theoretically better deal in the past

by the occurrence of unforeseen events in the future does not qualify as an impoverishment under these equity principles.” Id. at 363. Fourth, PWE did not show that “‘as between’ RSI and PWE, ‘it is unjust for [RSI] to retain’ the money instead of PWE.” Id. (alteration in original). Rather, “RSI has a more equitable claim to the money.” Id. RSI operated for years without rebating agreements, and under those circumstances, RSI got its statutory takeout, used it to pay all its expenses, and then kept the rest as “profit.” Id. When professional players like PWE began pressing RSI for rebates, RSI was left with less money to pay expenses, including the taxes everyone assumed the State required. Id. In short, the takeout “was RSI’s money to begin with—until professional gamblers started to demand more of it. [The portion that was paid to the State] is thus RSI’s money when it was returned to the bankruptcy estate years later.” Id. The court then considered whether any formula existed to allocate the returned funds between the parties, ultimately concluding one did not exist. Id. at 364. The court found that PWE failed to show that it would have received the entire benefit of eliminated taxes retroactively. Id.

Instead, all PWE could show is that the parties would have renegotiated prospectively as they did in Nuevo Laredo, where they split the benefit of the reduced taxes. Id. “[T]o rely on equitable principles and divide the money based on some theoretical renegotiated deal,” the court would have had to guess or speculate on the proper division, which it refused to do. Id. The only relevant example to consider was the Nuevo Laredo deal, but that deal “appear[ed] to have additional, very different, elements to consider than the elements at play in . . . North Dakota.” Id. The circumstances in North Dakota likely would have called for a more favorable deal for RSI than the 1/3 that RSI got in Nuevo Laredo. Id. The court also considered the fact that PWE did not claim entitlement to any retroactively

returned funds in its original claim, instead only doing so many years later after the funds were returned. Id. The court found that PWE’s “failure to raise its entitlement earlier—when it was making the claim for the estate—goes to the credibility of PWE’s belated assertion that this was its money. At a minimum, it shows that PWE’s arguments about unjust enrichment are lacking for an addition[al] reason.” Id. at 364–65. Lastly, the court noted that North Dakota unjust enrichment law requires one final element that PWE could not meet: misconduct, fault, or undue advantage taken by RSI. Id. at 365. For all these reasons, the court rejected PWE’s equity-based arguments and denied PWE’s Amended Claim. 2. Carlson’s Claim Like PWE, Carlson made both contract- and equity-based arguments in support of his claim. First, he argued that his agreement with RSI was that he would receive the benefit of any change in North Dakota taxes. Id. at 365. Second, he argued that he was entitled to the returned money under the doctrine of unjust enrichment because he was the taxpayer and these taxes were

analogous to sales taxes. Id. As an initial matter, the Bankruptcy Court found that Bala rebutted the claim’s presumption of validity, shifting the ultimate burden of proof to Carlson. Id. at 366. The court also made credibility findings: Carlson’s testimony “lacked detail and credibility,” and Bala’s testimony was “credible on all points dealing with her agreement with Carlson.” Id. at 350. a. Contract Terms The court began its analysis by addressing the terms of Carlson’s oral rebating agreement with RSI. The court disagreed with Carlson that RSI was limited to a 3.5% fee, stating that there was even less evidence to support Carlson’s version than PWE’s similar claim. Id. at 366. Instead,

Carlson’s rebating agreement was based on the percentage he would receive. Id. No contract terms addressed the specific situation the parties now found themselves in, so Carlson did not have a claim for breach of an express contract provision. Id. The court then turned to Carlson’s argument that RSI agreed he would receive the benefit of reduced taxes. The court found that the parties discussed the possibility that North Dakota would reduce taxes in the future. Id. at 367. However, such discussions “were always statements of intention to renegotiate the oral agreement in the future” and not “a binding future modification of the original oral agreement.” Id. Carlson did not provide any support for his testimony that the discussions contemplated retroactive payment of returned tax money. Id. In fact, Carlson corroborated Bala’s testimony when he testified that she told him, “If something changes in the future, we can look at it.” Id. Because the parties simply agreed to future negotiations if circumstances changed, Carlson did not meet his burden of proof on his contract theory. Id. b. Unjust Enrichment Turning to Carlson’s unjust enrichment theory, the Bankruptcy Court first rejected his

argument that he was the taxpayer for the same reasons that it rejected PWE’s similar argument. Id. at 367–68. In short, RSI, not Carlson, was the taxpayer, and the taxes here were more like the taxes in Lakhani than the sales taxes at issue in Mann v. North Dakota Tax Commissioner, 2005 ND 36, 692 N.W.2d 490 (2005). Racing Servs., 595 B.R. at 368. The court then rejected Carlson’s remaining unjust enrichment arguments. Carlson got what he bargained for with RSI and was not deprived of anything that was rightly his. Id. RSI and Bala’s overturned criminal convictions had nothing to do with RSI being unjustly enriched at Carlson’s expense. Id. Further, the record failed to support a finding that RSI or Bala had unclean hands or continued to collect the tax while knowing it was unauthorized. Id. Lastly, the court

incorporated its reasoning for denying PWE’s unjust enrichment arguments. Id. Because Carlson failed to meet his burden to show the money equitably belonged to him, the court denied his claim for the returned tax money. Id. II. STANDARD OF REVIEW AND BURDEN OF PROOF FOR BANKRUPTCY CLAIMS When a bankruptcy court judgment is appealed to a district court, the district court functions as an appellate court. PW Enters. v. North Dakota (In re Racing Servs.), 504 B.R. 549, 554 (D.N.D. 2014) (citing Fix v. First State Bank of Roscoe, 559 F.3d 803, 808 (8th Cir. 2009)). A bankruptcy court’s findings of fact “shall not be set aside unless clearly erroneous, and due regard shall be given to the opportunity of the bankruptcy court to judge the credibility of witnesses.” Id. (quoting Fed. R. Bankr. P. 8013). “A factual finding is clearly erroneous ‘when although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed.’” Id. (quoting Anderson v. City of Bessemer City, N.C., 470 U.S. 564, 573 (1985)). “All legal determinations made by the bankruptcy court are reviewed de novo,” and so are conclusions involving mixed questions of law

and fact. Id. (citing DeBold v. Case, 452 F.3d 756, 761 (8th Cir. 2006)). The burden of proof is also important to this appeal. The Bankruptcy Code and Bankruptcy Rules provide the standards for burden of proof on bankruptcy claims. When a party properly files a proof of claim, “the claim is deemed allowed and the proof constitutes prima facie evidence of the claim’s validity and amount.” In re Tanner, No. BR 12-01429, 2013 WL 2318848, at *4 (Bankr. N.D. Iowa May 28, 2013) (citing In re Be-Mac Transp. Co., 83 F.3d 1020, 1025 (8th Cir. 1996); 11 U.S.C. § 502(a)). Pursuant to Bankruptcy Rule 3001(f), the burden shifts to the objecting party to produce some evidence to rebut the presumption of validity. Id. (citing In re Dove-Nation, 318 B.R. 147, 152 (B.A.P. 8th Cir. 2004); In re Muller, 479 B.R. 508, 514 (Bankr. W.D. Ark.

2012)). To elaborate on this second step in the burden-shifting analysis: [T]he burden of going forward with the evidence then shifts to the objecting party to produce evidence at least equal in probative force to that offered by the proof of claim and which, if believed, would refute at least one of the allegations that is essential to the claim’s legal sufficiency. This can be done by the objecting party producing specific and detailed allegations that place the claim into dispute, by the presentation of legal arguments based upon the contents of the claim and its supporting documents, or by the presentation of pretrial pleadings, such as a motion for summary judgment, in which evidence is presented to bring the validity of the claim into question. . . . In re Armstrong, 320 B.R. 97, 104 (Bankr. N.D. Tex. 2005) (first alteration in original) (quoting In re Rally Partners, L.P., 306 B.R. 165, 168–69 (Bankr. E.D. Tex. 2003)). If the objecting party meets its burden of production, “the ultimate burden of persuasion as to the allowability of the claim resides with the creditor.” Tanner, 2013 WL 2318848, at *4 (citing Dove-Nation, 318 B.R. at 152; Be-Mac Transp., 83 F.3d at 1025 n.3). At this final step, the claimant must “establish the validity and amount of the claim by a preponderance of the evidence.” Armstrong, 320 B.R. at 104 (quoting Rally Partners, 306 B.R. at 169). III. DISCUSSION As they did below, both PWE and Carlson argue that they are entitled to the returned money

on contractual and equitable grounds. They each assert a number of legal and factual errors committed by the Bankruptcy Court in denying their respective claims. The Court will begin its analysis with Carlson’s appeal and conclude with PWE’s appeal. A. Carlson’s Appeal Carlson presents two broad issues on appeal: (1) whether the Bankruptcy Court erred as a matter of law when it denied his claim on the basis of unjust enrichment and failed to order restitution of amounts paid by him and refunded by the State; and (2) whether, in the alternative, the Bankruptcy Court erred when it denied his claim on the basis of breach of contract. 1. Unjust Enrichment

Carlson dedicates most of his briefing to arguing that the Bankruptcy Court erred as a matter of law when it denied his claim based on a lack of unjust enrichment. Prior to reviewing Carlson’s arguments, a brief overview of the law relative to unjust enrichment will be helpful. To prevail on an unjust enrichment claim under North Dakota law, one must prove the following elements: “(1) an enrichment, (2) an impoverishment, (3) a connection between the enrichment and the impoverishment, (4) the absence of a justification for the enrichment and impoverishment, and (5) the absence of a remedy provided by law.” McDougall v. AgCountry Farm Credit Servs., PCA, 2020 ND 6, ¶ 21, 937 N.W.2d 546 (quoting McColl Farms, LLC v. Pflaum, 2013 ND 169, ¶ 18, 837 N.W.2d 359). The equitable doctrine of unjust enrichment “may be invoked when a person has and retains money or benefits which in justice and equity belong to another.” Id. (quoting McColl Farms, 2013 ND 169, ¶ 18, 837 N.W.2d 359). Put another way, “[t]he person receiving the benefit is liable to pay . . . only if the circumstances of its receipt or retention are such that, as between the two persons, it is unjust for him to retain it.” Jerry Harmon Motors v. Heth, 316 N.W.2d 324, 328 (N.D. 1982) (citation omitted).

“A determination of unjust enrichment is necessarily a conclusion of law because it holds that a certain state of facts is contrary to equity.” In re Zent, 459 N.W.2d 795, 798 (N.D. 1990) (citing Midland Diesel Serv. & Engine Co. v. Sivertson, 307 N.W.2d 555, 557 (N.D. 1981)). “The [bankruptcy] court’s conclusions that unjust enrichment has not occurred and that there is no contract implied in law, are, therefore, fully reviewable.” Id. (citing Midland Diesel, 307 N.W.2d at 557). Any factual findings, however, that support these legal questions are still subject to the usual clear error review. Id. Carlson raises several issues with the Bankruptcy Court’s unjust enrichment analysis, which the Court will address in turn.18 a. Was This a Sales Tax and Who Was the Taxpayer?

Analysis of Carlson’s unjust enrichment arguments primarily boils down to two questions: (1) Was the “tax” paid to the State akin to a sales tax, and relatedly, (2) was Carlson essentially the “taxpayer”? For Carlson’s arguments from the Restatement on Restitution and case law to

18 As an initial matter, in their briefs, the parties argue back and forth about whether there is a difference between unjust enrichment and restitution and whether Carlson should be able to raise restitution arguments based on the Restatement on Restitution. Any distinction is a distinction without a difference for the purpose of this Court’s analysis. Unjust enrichment is the doctrine that provides a basis for relief, and restitution is the remedy. See Apache Corp. v. MDU Res. Grp., Inc., 1999 ND 247, ¶¶ 13–14, 603 N.W.2d 891. Because Carlson is not making a new type of argument for why he should get the returned funds, his arguments based on the Restatement are not waived for failure to raise them below. The Court will therefore consider Carlson’s arguments from both case law and the Restatement in reviewing the Bankruptcy Court’s unjust enrichment analysis. succeed, the answer to one or both questions essentially must be yes. Unfortunately for Carlson, however, this Court finds no error in the Bankruptcy Court’s determinations that the pari-mutuel taxes paid in this case were distinguishable from sales taxes and that RSI, if anyone, should be considered the taxpayer. The taxes improperly collected by the State are simply not analogous to a sales tax. Under

North Dakota law, the consumer is responsible for paying a sales tax on top of the purchase price, and the retailer merely acts as a collection agent for the taxing authority. N.D. Cent. Code § 57- 39.2-08.2(1). Thus, if the consumer buys some good priced at $10 and the sales tax rate is 6%, the consumer must pay a total of $10.60 to the retailer who collects the tax and passes it on to the State. The taxes paid on Carlson’s account wagering worked differently, as they were not an additional expense paid on top of the original purchase price. If Carlson wanted to bet $10, he paid $10, not $10.40.19 If Carlson’s bet was a winning bet, he would get the $10 back, plus his winnings and whatever portion of the takeout he was entitled to under the rebating agreement. The Court is not unaware that if a player bet no money, there would be no handle and thus

no tax; however, that fact does not transform the pari-mutuel tax into a sales tax or Carlson into the taxpayer. The Court finds an analogy made by Bala in her response brief instructive here: It should be evident that, as a logical matter, this simply goes too far. By this logic, Carlson would be “the taxpayer” with respect to whatever marginal income taxes his counsel owes based on the income derived from Carlson hiring him and paying his fees, which all must agree would be absurd. The question here is who has the legal responsibility for paying the tax, not for generating the activity that leads to it. That was unquestionably RSI at the time all parties believed that the taxes were owed under the statute. See [N.D. Cent. Code] § 53-06.2-11 (2001). Doc. No. 36, pp. 17–18.

19 Carlson tended to place exotic wagers, which by statute had a 4% tax rate. Determining who was the “taxpayer” is somewhat muddied by the confusing state of the statutory and regulatory scheme in North Dakota in the early 2000s. What is clear, however, is that Carlson should not be considered the taxpayer. Under North Dakota’s takeout statute, the licensee is entitled to withhold the takeout, and from the takeout “the licensee” shall pay any required taxes. See N.D. Cent. Code. § 53-06.2-11. No argument can be made that Carlson, or

any player, was the licensee; RSI, meanwhile, was at least the licensed service provider—and perhaps also the site operator upon the advent of account wagering.20 It is undisputed that, at all relevant times, RSI withheld the takeout from the players’ handle and the taxes came out of this takeout. This Court also finds no error in the Bankruptcy Court’s citation to a single tax law case, Lakhani, in ruling on the players’ claims even though no taxes were actually authorized on account wagering. The Bankruptcy Court used Lakhani for the limited purpose of supporting authority for how takeout and taxes generally work in pari-mutuel horse wagering: The takeout belongs to the track, and “the taxes, license fees, and other expenses discharged from the takeout are expenses

20 In its appeal, PWE argues that under North Dakota law, the takeout belonged to the “simulcast operator” rather than the “simulcast service provider,” the two roles were mutually exclusive, and RSI was only the “simulcast service provider.” See N.D. Admin. Code §§ 69.5-01-11-04 (“A service provider may not be licensed as a simulcast operator.”) and 69.5-01-11-06 (“The simulcast operator is responsible for the payment of the state takeout . . . .”). Bala, meanwhile, argues that account wagering worked differently, and for the purposes of account wagering, RSI was properly serving as both the service provider and the site operator. See Bala, 489 F.3d at 338-39 (“[T]he 2001 statute that authorized account wagering expanded the exception for simulcast pari-mutuel wagering . . . [granting] the licensed simulcast service provider, RSI, exclusive right to conduct account wagering, without an additional license . . . .”); N.D. Cent. Code § 53-06.2-10.1 (2001) (“An account wager . . . may only be made through the licensed simulcast service provider authorized by the commission to operate the simulcast pari-mutuel wagering system under the certificate system.”). Ultimately, the Court need not delve into any possible distinction between a “simulcast service provider” and a “simulcast operator” because it is undisputed that the players held no license for either role. The site operator’s alleged interest in the returned money would not grant Carlson or PWE an interest in the money or a greater entitlement to it than RSI. imposed upon the track, not the bettors.” Lakhani, 142 T.C. at 160.21 Here, the takeout went to the provider, not the track, but the point remains the same: The takeout does not belong to the player, so absent some agreement to the contrary, the player has no entitlement to any portion of that takeout, including the portion that was used to pay presumably required pari-mutuel taxes. Because these taxes were not akin to sales taxes and Carlson was not the taxpayer, his

comparisons to examples from the Restatement and case law are unpersuasive. For example, Carlson relies heavily upon Acton Construction Co. v. Commissioner of Revenue, 391 N.W.2d 828 (Minn. 1986), but the facts of that case are distinguishable. In Acton, two general contractors factored expected sales tax liability into their winning bids. Id. at 831. It was later determined that the sales tax was erroneously collected and paid to the state. Id. The Commissioner of Revenue refused to return the refunded amount of tax to the contractors unless they would remit it to the customers who paid it. Id. at 831–32. The Minnesota Supreme Court held that the Commissioner had the authority to require the contractors to return the refunded sales tax to their customers as a condition of the refund, explaining:

[T]he contractor is characterized for tax purposes as a vendor whose transfer of the finished pipe product to the customer is taxed as a retail sale. As purchaser, the contract customer has paid the sales tax, while the contractor, as vendor, merely remits the tax to the state. Under the facts of this case, appellants’ contract customers are the purchasers to whom the refund of excess sales tax paid must be remitted. Id. at 832–33. The court further explained that if the contractors did not retain the refunded amount, they would “still receive the expected amount of profit from these contracts”; the contractors would be “deprived only of a windfall profit, an amount both parties mistakenly included in their calculations of a fair contract price.” Id. at 833.

21 That Lakhani involved a different state’s (California) pari-mutuel wagering statute is irrelevant here. As discussed above, under North Dakota’s statute, the takeout similarly does not belong to the player, and taxes similarly come out of the takeout. Acton might work in Carlson’s favor if the taxes paid to the State were akin to sales taxes on his account wagering, but, as explained above, they were not. In Acton, the customers ended up paying for sales taxes on top of the completed product; this cost was borne by the customers, not the contractors, though the latter were responsible for remitting the tax to the State of Minnesota. Here, however, the players did not bear the burden of the tax; RSI paid the tax out of

its takeout, and the players did not have to pay anything extra for the tax. A $1 bet cost $1 regardless of the tax rate. Because RSI was not functioning as a vendor that merely remitted the tax to the State, Acton is inapposite. Carlson’s remaining arguments for why he is entitled to restitution for taxes assessed on his account wagering fail on similar grounds. Carlson ultimately has not demonstrated that the Bankruptcy Court erred in finding that RSI had a greater entitlement to the money. b. Wrongfulness Element Carlson claims three other errors in the Bankruptcy Court’s unjust enrichment analysis. First, he asserts the court erroneously concluded that North Dakota law requires a showing of

wrongfulness before a party can prevail on an unjust enrichment claim. North Dakota law seems somewhat unclear on this issue, as some cases appear to suggest that a showing of misconduct or fault is necessary, but others explicitly state that no such showing is required. Compare Jerry Harmon Motors, 316 N.W.2d at 328 (citation omitted) (“A valid claim for unjust enrichment can be based only on an element of misconduct or fault or undue advantage taken by one party of another, and, in the absence of fraud or bad faith, a person is not entitled to compensation on the ground of unjust enrichment if he received from the other that which it was agreed between them that the other should give in return.”), with Sykeston Twp. v. Wells Cty., 356 N.W.2d 136, 140 (N.D. 1984) (“Fraud or other misconduct on the part of the person alleged to have been unjustly enriched need not be shown.”), and Midland Diesel, 307 N.W.2d at 557 (internal citations omitted) (“A complainant need not show fraud or other misconduct on the part of the recipient to recover. It is sufficient if the latter has, without justification, obtained a benefit at the direct expense of the former, who then has no legal means of retrieving it.”). The Court need not resolve any apparent discrepancy in North Dakota law, however, as

any error in adding a requirement of fault or misconduct is harmless. See Fed. R. Civ. P. 61 (harmless error standard); Fed. R. Bankr. P. 9005 (adopting FRCP 61 in the bankruptcy context). The finding that RSI did not engage in misconduct or fault or take undue advantage of the players was merely given by the Bankruptcy Court as one final reason that unjust enrichment was not proven by either player. Other reasons for denying relief based on unjust enrichment still stand. Most significantly, even if fraud or misconduct is not required, Carlson has failed to show that “as between [him and RSI], it is unjust for [RSI] to retain [the money].” Jerry Harmon Motors, 316 N.W.2d at 328. c. Impoverishment Element

Second, Carlson draws this Court’s attention to one sentence in the Order that he argues shows a fundamental misunderstanding of the nature of equitable relief: In discussing the impoverishment element, the Bankruptcy Court wrote, “Carlson got what he bargained for with RSI, and like PWE was not unfairly deprived of something that was his.” Racing Servs., 595 B.R. at 368. Carlson does have a point that “[r]estitution deals with nonbargained benefits . . . [and] contract law with bargained benefits and harms.” Alts. Unlimited, Inc. v. New Balt. City Bd. of Sch. Comm’rs, 155 Md. App. 415, 453, 843 A.2d 252 (2004) (quoting Saul Levmore, Explaining Restitution, 71 Va. L. Rev. 65, 67 (1985)). Nevertheless, the sentence in the Order about Carlson getting what he bargained for does not constitute an error meriting reversal. The court correctly identified that impoverishment is a requirement under North Dakota law. The court then considered what could constitute an impoverishment in this case. As part of that analysis, the court noted that no impoverishment could be found in the agreement itself, as PWE and Carlson both got the rebates they bargained for. The court also rejected another possibility: that “[m]issing out on a theoretically better deal in the past by the occurrence of unforeseen events in the future” could

qualify as an impoverishment. Id. at 363. Carlson does not posit anything else that could constitute an impoverishment, outside of his sales tax-related arguments that the Court has already rejected. d. Equitable Accounting Lastly, Carlson argues that the Bankruptcy Court’s concern about equitable accounting is without basis under the law and undisputed facts. In short, he asserts that the court was incorrect to determine that there was no formula to account equitably for allocation of the funds because case law and the Restatement make clear that the applicable measure of damages for restitution is the amount of the benefit unjustly received. See KLE Constr., LLC v. Twalker Dev., LLC, 2016 ND 229, ¶ 15, 887 N.W.2d 536 (stating that the proper amount of damages is the amount that the

defendant was unjustly enriched). This Court agrees that the proper amount of restitution would be the total amount of returned taxes assessed on Carlson’s wagering if his sales tax arguments were meritorious. However, the Bankruptcy Court apparently was getting at whether there was a way to equitably account for the likelihood that the parties would have renegotiated had the tax rate changed—if missing out on renegotiation could constitute an impoverishment. Based on the record, the court found that Carlson would not have gotten the full amount of the additional tax margin, so some other formula than 100% to Carlson would be needed.22 The court ultimately

22 This equitable accounting discussion was actually part of the court’s analysis of PWE’s claim, but as mentioned previously, the court incorporated its analysis of PWE’s unjust enrichment theory into its analysis of Carlson’s similar theory. concluded that there was no way to adequately determine how the parties would have split the margin. No error exists here. In sum, none of the errors claimed by Carlson related to the Bankruptcy Court’s unjust enrichment analysis persuade this Court that reversal is warranted. Carlson has not demonstrated a greater entitlement to the returned money than RSI. The taxes were paid out of RSI’s takeout,

which Carlson has no entitlement to outside of any rebating agreement stating otherwise. This Court therefore affirms the Bankruptcy Court’s denial of Carlson’s Claim on unjust enrichment grounds. 2. Breach of Contract In the alternative, Carlson argues that the Bankruptcy Court erred when it denied his claim on contractual grounds. More specifically, he asserts that the Bankruptcy Court erred in ruling that the rebating agreement did not address the issue between the parties because the parties agreed that RSI would not profit from or upcharge the actual taxes that resulted from Carlson’s wagering. He does not challenge on appeal, however, the court’s findings that the parties’ agreement to

renegotiate if the tax rate changed was of a prospective, not retroactive, nature and that the parties based their rebating agreement on the percentage Carlson would receive rather than the percentage RSI would ultimately retain. Again, a brief overview of North Dakota law will be helpful. In North Dakota, an oral contract is interpreted using the same general principles as a written agreement, except that “the terms of an oral contract can be established only through extrinsic evidence” and “[a] determination of these terms, if they are disputed, must therefore be made by the trier of fact.” Tallackson Potato Co. v. MTK Potato Co., 278 N.W.2d 417, 422 (N.D. 1979); see also Curtis Constr. Co. v. Am. Steel Span, Inc., 2005 ND 218, ¶ 13, 707 N.W.2d 68 (citation omitted) (“For disputes involving oral contracts, the trier of fact decides the terms of the contract.”). “A contract must be so interpreted as to give effect to the mutual intention of the parties as it existed at the time of contracting so far as the same is ascertainable and lawful.” N.D. Cent. Code § 9-07-03. “However broad may be the terms of a contract, it extends only to those things concerning which it appears that the parties intended to contract.” N.D. Cent. Code § 9-07-13. Mutual intention

does not exist “unless the parties all agree upon the same thing in the same sense.” Thompson v. Thompson, 391 N.W.2d 608, 610 (N.D. 1986) (quoting N.D. Cent. Code § 9-03-16). “The existence of an oral contract and the extent of its terms are questions of fact subject to the clearly erroneous standard of review . . . .” WFND, LLC v. Fargo Marc, LLC, 2007 ND 67, ¶ 38, 730 N.W.2d 841 (citing Edward H. Schwartz Constr., Inc. v. Driessen, 2006 ND 15, ¶ 6, 709 N.W.2d 733). “After the terms of the oral contract have been determined, the issue of the parties’ intentions, if ascertainable from those terms, involves a question of law for the court to decide, and will be reversed by this court on appeal if erroneous.” Tallackson Potato Co., 278 N.W.2d at 422. Having reviewed the record, the Court is not persuaded that the Bankruptcy Court erred in

denying Carlson’s Claim on contractual grounds. The Bankruptcy Court made specific factual findings regarding the terms of the oral rebating agreement based on Carlson’s and Bala’s testimony that evinces the court’s rejection of Carlson’s argument.23 First, the court found that RSI calculated taxes based on the handle and the presumed statutory tax rate for the particular type of bet. At all times, RSI paid taxes on Carlson’s handle at the rate determined by the types of bets he made—usually 4%, as the vast majority of his wagering consisted of exotic wagers. Second, the court found that the rebating agreement was based on the percentage Carlson would receive

23 The Bankruptcy Court also explicitly rejected Carlson’s argument that “RSI agreed to never profit from or ‘mark up,’ the tax” in its unjust enrichment analysis. Racing Servs., 595 B.R. at 367. rather than the amount RSI would ultimately retain as a fee. In other words, Carlson’s rebate was not directly affected by the amount paid as taxes. Lastly, the court found that none of the contract terms addressed the specific issue involved here: the unexpected return of tax money years later. The parties did not anticipate the retroactive change in taxes and thus lacked the mutual intention to address allocation of the money. Instead, the parties arrived at an agreement that was based on

the circumstances as they believed them to be at the time. Testimony about any additional contract term pertaining to profiting off taxes was sparse at best. Only two potential references to such a term were made by Carlson, who the Bankruptcy Court found to lack credibility. Carlson answered affirmatively when asked “Just for clarity your deal where RSI got to retain 11 percent was premised on them paying the taxes without a profit?” Earlier, he had denied that “there [was] ever to be any profit made by RSI on the taxes that [he] paid.” Meanwhile, Bala’s testimony about marking up or profiting off taxes was limited to the following exchanges on cross- and recross-examination: Q: Did RSI mark up the North Dakota state taxes and try to make a profit on the North Dakota State taxes? A: I don’t know what that means. Q: Well, if – if on – on one of PWE’s sheets, for example, it says that there was 3.8 percent horsemen and state taxes. Was that the actual amount of taxes or did you do some sort of a markup at RSI? A: That would have been the actual. Q: So you never tried to get an extra half-percent or quarter-percent or anything like that? A: Not on those taxes, no. Q: Okay. And the same is true of track fees? A: It was a pass through cost. It was, in other words, not a wholesale onto retail model. It was – that was – that was the cost. . . . . Q: But you didn’t negotiate a deal to make a profit on the taxes. You said that a couple times and it’s still true right now, correct? Yes or no, please. A: The – no, we didn’t make a deal to make a profit on taxes. Based on the parties’ understanding of the tax situation in North Dakota in the early 2000s, RSI did not mark up taxes or try to profit off them. In context, the alleged agreement not to mark up taxes or profit off them—if such an agreement even existed—appears to be just that: an agreement not to profit off taxes by marking them up to a higher percentage than the parties believed at the time that the statute required. Carlson has failed to show that the Bankruptcy Court’s findings regarding the terms of the contract were clearly erroneous, and thus the court’s findings cannot be set aside. The Bankruptcy Court’s denial of Carlson’s Claim is affirmed. B. PWE’s Appeal

The Court now turns to PWE’s appeal. PWE alleges several specific legal and factual errors. PWE also generally challenges the Bankruptcy Court’s conclusions at the second and third steps of the burden-shifting analysis and argues that the Court’s ruling was not supported by substantial evidence. The Court will address each of the specific claims of error before concluding by reviewing the Bankruptcy Court’s burden-shifting analysis. 1. References to RSI’s “Profit” First, PWE asserts that the Bankruptcy Court erred as a matter of law when it referred to RSI’s “profit” because under the takeout statute, takeout can only be used for directly incurred expenses or charitable purposes. PWE argues that because RSI cannot make a profit, the returned portion of the takeout must go either to charity or to the players, and the Bankruptcy Court erred

by instead awarding the money to RSI. In 2001, subsection 5 of the takeout statute read as follows: A licensee may not use any of the portion deducted for expenses under subsections 1 and 2 for expenses not directly incurred by the licensee in conducting parimutuel racing under the certificate system. After paying qualifying expenses, the licensee shall use the remainder of the amount so withheld only for eligible uses allowed to charitable gambling organizations under subsection 2 of section 53-06.1-11. N.D. Cent. Code § 53-06.2-11(5). Given that subsections 1 and 2 technically did not apply on their face to account wagering until 2007, the Constitution of North Dakota is also instructive: The legislative assembly shall not authorize any game of chance, lottery, or gift enterprises, under any pretense, or for any purpose whatever. However, . . . the legislative assembly may authorize by law bona fide nonprofit veterans’, charitable, educational, religious, or fraternal organizations, civic and service clubs, or such other public-spirited organizations as it may recognize, to conduct games of chance when the entire net proceeds of such games of chance are to be devoted to educational, charitable, patriotic, fraternal, religious, or other public-spirited uses. N.D. Const. art. XI, § 25. It may very well be correct that North Dakota law in the early 2000s required the licensee to dedicate its entire takeout from account wagering to certain expenses or charitable purposes; however, the Court need not decide this question, nor does the Court think it proper to do so. “As a general rule, [an appellate court does] not consider arguments or theories on appeal that were not advanced in the proceedings below.” Ames v. Nationwide Mut. Ins. Co., 760 F.3d 763, 770 (8th Cir. 2014) (quoting Wright v. Newman, 735 F.2d 1073, 1076 (8th Cir. 1984)). The post-trial briefing below is devoid of any argument that RSI could not legally keep any of the takeout as profit. Further, the profit issue is currently before the BAP on a separate appeal.24

24 On December 31, 2018, the State filed a new proof of claim, purportedly “as Representative for the benefit of eligible nonprofit organizations.” Doc. No. 1015, Bankr. Case No. 04-30236, p. 1. The State claimed that “certain eligible charitable organizations are entitled to any excess funds in the bankruptcy estate,” arguing that under North Dakota law, net proceeds must be disbursed to charity. Id. at 4. Bala and the Trustee objected. Id. On January 29, 2020, the Bankruptcy Court denied the State’s Claim, holding that (1) the State lacked standing on its own behalf and (2) the State may have standing as a representative of an eligible charitable organization under a Consent and Assignment Agreement, but the charity’s claim was “barred by the doctrine of laches and can no longer be pursued by any party.” Id. at 2. The State and PWE appealed the order to the BAP. The BAP dismissed PWE’s appeal (docketed at 8th Cir. BAP Case No. 20-6004) for lack of standing but has yet to rule on the State’s appeal (docketed at 8th Cir. BAP Case No. 20-6002). Ultimately, whether it was erroneous for the Bankruptcy Court to make limited references to RSI’s “profit” is inconsequential to the instant appeal. Assuming without deciding that RSI is not permitted to retain “profit” or “net proceeds,” RSI’s inability to do so would not logically alter the Bankruptcy Court’s denial of PWE’s Amended Claim and entitle PWE to the money that was improperly collected as taxes. Rather, if PWE is correct, the Trustee would be required to pay the

returned money to charity, not PWE. PWE itself has argued that the portion of the takeout that was returned by the State must either go to other directly incurred expenses or to charitable purposes. PWE does not fall under either of these categories, however. The Bankruptcy Court found that PWE did not meet its burden of proof on the contract portion of its claim; in other words, barring error in that respect, the returned money is not an additional rebate expense that RSI must pay. And as a company engaged in very high-volume betting for profit, PWE is clearly not a charitable organization. Thus, whether RSI can ultimately keep the returned money has no bearing on PWE’s entitlement to the money. Any error in referring to RSI’s “profit” is harmless. 2. References to Returned “Taxes”

Next, PWE argues that the Bankruptcy Court erred when it referred to the money returned to the bankruptcy estate as returned “taxes” because the Eighth Circuit held that North Dakota law did not authorize a tax on account wagering until 2007. This argument amounts to nothing more than semantics. As evinced by the Bankruptcy Court’s recitation of the background of the bankruptcy proceedings, the court understood that no taxes had been authorized and did not conclude otherwise by referring to the money as returned “taxes.” In fact, in the Order, the court at times referred to the money as “money previously paid as taxes,” “illegally collected taxes,” and “unauthorized taxes.” It is clear that the court used the term “taxes” either as a shorthand way to describe the money returned by the State or as a reference to what the parties identified as taxes when negotiating their rebating agreement and when conducting business according to that agreement. Further, as noted previously, the Bankruptcy Court did not rely extensively on tax law, instead citing a single tax case for a limited purpose. All in all, the court’s use of the word “taxes” as shorthand does not constitute legal error. 3. Identity of the “Taxpayer”

Assuming any party can even be considered the “taxpayer” where no taxes were actually authorized, PWE argues that the Bankruptcy Court erred as a matter of law and fact by concluding that RSI, not PWE, was the “taxpayer.” PWE’s argument largely tracks Carlson’s argument that this Court rejected above; however, PWE relies on two additional pieces of evidence that merit the Court’s consideration. First, PWE points to the legislative history behind Senate Bill 2285, a 2003 bill to authorize a new type of wagering in North Dakota. PWE Exhibit 60 shows that Bala and RSI advocated for a reduced tax structure because “[a] substantial portion of the wagering handle in North Dakota [was] at risk to move out of the state without a more competitive tax rate.” Doc. No. 5-25, p. 14.

Bala and RSI also hoped that a lower tax rate would stimulate out-of-state volume. Id. at 12, 14. PWE argues that “[t]he legislative history demonstrates that Bala recognized that the players altered their wagering behavior based upon tax rates, which (logic dictates) players would only do if they were responsible for those taxes.” Doc. No. 19, p. 38. Bala, however, provided a reasonable alternative explanation at trial for why lower tax rates draw players to certain jurisdictions: If the tax rate goes down, the service provider has more money to work with, in turn allowing players to negotiate a better rebate. The Bankruptcy Court recognized as much, finding that “[a]t the time RSI negotiated deals with these professional gamblers, the only money RSI could use was the amount of the takeout minus expenses which appeared to everyone involved to include taxes.” Racing Servs., 595 B.R. at 363. This finding is supported by the record and does not constitute clear error. Second, PWE highlights Wagner’s statement in his May 31, 2000, email (PWE Exhibit 10) that “it became best for [PWE] to leave the Mirage and go to ND where we pay horsemans fees and state taxes.” Doc. No. 5-8. The Court fails to see how this statement demonstrates error in

the Bankruptcy Court’s finding that RSI was the “taxpayer.” At most, the email proves that Wagner believed in 2000 that PWE was the “taxpayer.” Wagner’s subjective understanding does not change the objective reality recognized by the Bankruptcy Court that taxes are paid by a licensee out of its takeout. PWE has ultimately not shown that the Bankruptcy Court committed any legal or factual error in determining the identity of the “taxpayer.” 4. Additional Claimed Errors and General Review Lastly, PWE raises a number of additional errors the Bankruptcy Court purportedly made, argues that the court improperly shifted the burden of proof to PWE, and ultimately asserts that the court’s ruling on PWE’s Amended Claim was not supported by substantial evidence.

a. Failure to Address Contemporaneous Evidence First, PWE argues that the Bankruptcy Court failed to consider seven different pieces of contemporaneous evidence, thereby calling into question the court’s ultimate findings and conclusions. This evidence is as follows: • Wagner’s testimony about a contemporaneous computer model internal to PWE; • PWE Exhibit 10, the May 31, 2000, email from Wagner to Diaz, in which Wagner expresses an understanding that he was the taxpayer in North Dakota; • Wass’s testimony based on his document review that the agreement was for RSI to only receive 1% and for PWE to receive the rest; • PWE Exhibit 5, a July 18, 2001, email from Bala’s assistant to Wagner with an attached business model comparison that mentions a 1% RSI fee; • PWE Exhibit 17, an August 3, 2001, email from Wagner to Bala during the Nuevo Laredo negotiations, mentioning that RSI gets “1.0% of the handle in North Dakota”; • PWE Exhibit 4, an undated email/letter from Wagner to Bala and Diaz during negotiations involving an Idaho casino that mentions a 1% RSI fee; and • PWE Exhibit 50, a spreadsheet prepared by RSI during the North Dakota negotiations that includes a 1% RSI fee in its projections of PWE’s rebate at different levels of takeout. The court’s failure to explicitly address three of these pieces of evidence is rather inconsequential. PWE presented no evidence of its contemporaneous computer model beyond Wagner’s testimony, which the court found to be not credible. Wass was not employed by PWE at the time of the negotiations with RSI, so his testimony about the agreement was primarily based upon his review of documents that speak for themselves. The court instead logically focused on the testimony of the only witnesses who were part of the negotiations: Bala and Wagner. Wagner’s email expressing his subjective understanding of who pays taxes in North Dakota also possesses little to no evidentiary value as explained in the previous section. The failure to explicitly address the other four pieces of evidence merits a more in-depth look by this Court since they provide at least some level of direct support for the crux of PWE’s contract-based argument: that RSI was limited to its 1% fee. As an initial matter, while the Bankruptcy Court did not explicitly address these contemporaneous documents, it is apparent from the Order that the court reviewed and considered them: As noted in the previous section, the oral agreement contained language specifically addressing only the allocation of rebate money based on what the parties knew at the time of their agreement. PWE, at best, is arguing that the agreement becomes ambiguous when the money previously assumed to be for taxes is returned after unanticipated Court rulings holding there was no power to tax. PWE’s latent ambiguity argument would be that this later event makes its oral argument with RSI ambiguous about how extra money would be treated. PWE, in essence, argues that extrinsic evidence explains the ambiguity, how the structure of the rebating agreement meant to address this situation, and that as properly viewed the extrinsic evidence helps show the agreement actually means PWE prevails on its claim. The Court has reviewed the extrinsic evidence offered during the three-day hearing and is not persuaded. The evidence suggests that if the parties had actually faced the situation where money otherwise allocated for taxes became available during the course of the contractual relationship, they would have renegotiated their deal. They did not intend to cram new circumstances like this into a formula designed to deal with very different circumstances. Racing Servs., 595 B.R. at 358. A court is “not required to make specific findings with respect to all of the evidence presented to it.” Talley v. U.S. Postal Serv., 720 F.2d 505, 507 (8th Cir. 1983) (citing Bell v. Bolger, 708 F.2d 1312, 1318 n.8 (8th Cir. 1983)). “Findings are adequate if they afford a reviewing court a clear understanding of the basis of the trial court’s decision.” Bell, 708 F.2d at 1318 n.8 (citations omitted). Though more explicit findings may be desirable, it is sufficient if the court’s findings “adequately outline the basis of its decision.” Id. Although more explicit findings related to these four contemporaneous documents may have been preferable, the lack of such findings does not constitute a basis for reversal here. The Bankruptcy Court’s findings do provide an understanding of the basis of its decision. It is clear that the court (1) found both PWE and Bala’s versions of how the agreement was structured plausible based on the evidence but could not ultimately determine which was correct and (2) found that the parties’ agreement did not contemplate the retroactive return of tax money. On the first point, the court had to reject any idea that the documents established PWE’s version that RSI got 1% and PWE got the rest. It was not erroneous for the court to do so. The documents do support the fact that there was a 1% RSI fee, but even Bala did not dispute that fact. According to Bala, the 1% fee was one component of how the rebate was structured. The real question is whether RSI was limited to only its 1% fee if more money later came into the picture— a question not answered by the documents. On the second point, the documents have no bearing. It is undisputed that the parties did not consider the situation that ultimately occurred. The parties allocated all the takeout that they believed was available at the time, and evidence in the record suggests they would have renegotiated if more money had become available. As such, it was reasonable to find that the parties’ agreement did not address how to allocate the returned tax money. All in all, the court’s

basis for its decision on PWE’s contract-based theory was sufficiently laid out and supported by the record. b. Income Statements Second, PWE argues that the RSI-created income statements in PWE Exhibit 3 demonstrate how the monies belong to PWE. PWE points to purported mathematical errors that exist if Bala’s version of the agreement is correct and calls her explanation of the statements a post hoc rationalization. The Court is not persuaded. Looking at the income statements, this Court agrees with the Bankruptcy Court that both parties’ versions of the agreement are plausible. The income statements do provide some support

for PWE’s version, as they show a 1% RSI fee that never changes from week to week. In addition, PWE’s rebate each week on the statements equals the takeout minus the 1% fee and other expenses. It could thus be true then that RSI got 1% and PWE got the rest, as Wagner repeatedly testified. However, the income statements also comport with Bala’s version. Bala testified, “I took 20.5, I subtracted 11, which was the rebate I was offering to him, and that left, in round numbers, 9.5 on average, effectively, to work with.” Per Bala, the 1% was derived from the 9.5%, i.e., it was a mathematical way to get to that 9.5% based on the other projected expenses. In Bala’s version, (1) Wagner did not get exactly 11% each week because his rebate varied based on the pools and tracks he bet on; (2) the 1% RSI fee never changed every week because it was a component of the deal, just not the driving force; and (3) the expenses varied slightly from week to week, primarily due to different track fees. Ultimately, as demonstrated by the summary spreadsheet in Bala Exhibit 1, the numbers did average out to close to what should be expected if Bala’s testimony about the agreement is truthful: 9.7% of the handle in total expenses and an 11.0% effective rebate (10.924% per the Court’s own calculation).25

In sum, the income statements do provide some level of evidence for PWE’s version of the agreement, but the Bankruptcy Court’s ultimate determination that PWE failed to carry its burden of proof is not upended or shown to be erroneous by the evidence. c. Credibility Determinations Third, PWE argues that the Bankruptcy Court’s credibility determinations—that both Wagner and Bala lacked some credibility but that Bala was more credible—do not bar overturning the ruling on PWE’s Amended Claim. Since this Court is functioning as an appellate court, the standard for reviewing credibility determinations on appeal is as follows: When findings are based on determinations regarding the credibility of witnesses, Rule 52(a) demands even greater deference to the trial court’s findings; for 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. This is not to suggest that the trial judge may insulate his findings from review by denominating them credibility determinations, for factors other than demeanor and inflection go into the decision whether or not to believe a witness. Documents or objective evidence may contradict the witness’ story; or the story itself may be so internally inconsistent or implausible on its face that a reasonable factfinder would not credit it. Where such factors are present, the [appellate court] may well find clear error even in a finding purportedly based on a credibility determination. But 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.

25 In addition to attacking the 11.0% in the total row of the summary spreadsheet, Wagner points out that this summary was prepared for the purposes of litigation. While that is true, the summary simply consists of all the data in the weekly income statements. Thus, the time and purpose for which the summary was created is irrelevant. Bessemer City, 470 U.S. at 575 (internal citations omitted). The Bankruptcy Court’s credibility determinations were based on two primary factors: (1) demeanor and (2) the amount of explanation given by the witnesses for their respective versions of the agreement. This Court cannot assess the witnesses’ demeanor, but it finds the Bankruptcy Court’s reliance on the second factor to be reasonable. Bala gave a detailed explanation of how

and why variable rate rebates came to be and how the parties came to her version of the agreement. Wagner, meanwhile, simply kept repeating “she got 1% and we got the rest” without any explanation whatsoever of how the parties arrived at that agreement. While the contemporaneous evidence can be construed at least partially in Wagner’s favor, the evidence is not sufficient to deem the Bankruptcy Court’s findings that were based on credibility determinations clearly erroneous. d. Reliance on the Nuevo Laredo Agreement Fourth, PWE asserts that the Bankruptcy Court erred in its speculation about the Nuevo Laredo negotiations and the inferences it drew from the agreement that resulted there. On appeal,

PWE’s ability to demonstrate error based on inferences is significantly limited by the standard of review. “In an action tried without a jury, inferences drawn from documents or undisputed facts shall not be set aside unless they are clearly erroneous.” Clark Equip. Co. v. Keller, 570 F.2d 778, 793 n.15 (8th Cir. 1978) (citations omitted). “Moreover, when conflicting inferences may be drawn from the evidence, a reviewing court is not at liberty to disturb a reasonable inference made by the trial court unless it is definitely and firmly convinced that a mistake has been made.” Id. (citation omitted). “Finally, on appeal the evidence, including the inferences reasonably drawn therefrom, must be viewed in the light most favorable to the prevailing party.” Id. (citation omitted). The Bankruptcy Court’s Nuevo Laredo-based inferences find support in the record. It is evident from the exhibits related to Nuevo Laredo that PWE did not get all the money that was saved on taxes in that jurisdiction, only two-thirds of it. Bala testified about how the parties renegotiated in Nuevo Laredo when they were “in a world with a lower tax rate” and how the benefit of the lower taxes did not go entirely to PWE. This is consistent with her admission that

had the taxes been reduced or eliminated in North Dakota, she would have renegotiated with the players and they would have gotten a portion of the benefit. Wass agreed during his testimony that when PWE began wagering in this new jurisdiction with lower taxes, PWE did not get the entire benefit; he did couch this testimony, though, by explaining that Wagner was willing to assist Bala with additional costs in Nuevo Laredo. Perhaps then, other circumstances were different in Nuevo Laredo besides the lower taxes—hence the court’s refusal to use what happened in Nuevo Laredo as a formula to equitably allocate the returned tax money in North Dakota. Nuevo Laredo does nonetheless reasonably stand for the propositions the court used it for: The parties renegotiated when not faced with the North

Dakota taxes, and in that jurisdiction, RSI was not limited to 1% and PWE received some—but not all—of the extra tax margin. The Bankruptcy Court’s inferences from the Nuevo Laredo negotiations are thus reasonable and not clearly erroneous. e. Reliance on Cuscuna’s Testimony Fifth, PWE argues that the court erroneously relied upon Cuscuna’s testimony because Cuscuna was not involved in the negotiations between RSI and PWE and was unfamiliar with their agreement. The following is the entirety of the court’s mention of Cuscuna: David Cuscuna was the first professional player who received a rebate with RSI. Mr. Cuscuna testified at trial about rebating in the horse race betting industry, his relationship with RSI, the way they structured the rebates, and the way the rebating generally functioned for all parties involved. Racing Servs., 595 B.R. at 346. In other words, the court limited its reliance on Cuscuna’s testimony to general background about how rebates came to be and how they were structured. Cuscuna testified from his personal experience in helping Bala develop the variable rate model when he entered into RSI’s first ever rebating agreement. His testimony therefore helped support Bala’s testimony about how and why RSI gave out variable rebates instead of flat rebates.

Importantly, the court went no further and did not use his testimony to establish any specific terms of the agreement between RSI and PWE. No error can thus be found in the court’s limited reliance on Cuscuna’s testimony. f. RSI’s Payment of Taxes on All North Dakota Wagers Sixth, PWE argues that the Bankruptcy Court erred when it stated that “RSI paid the State of North Dakota taxes on all wagers that came through its North Dakota operation.” Id. at 345 (emphasis added). This is an erroneous statement, PWE asserts, because Bala and RSI were indicted for failure to pay taxes, the State filed a multi-million dollar claim in the bankruptcy proceedings for unpaid taxes, and those taxes were never paid. Even if RSI did not technically

pay taxes on every single wager in North Dakota, however, PWE has failed to explain how this statement by the court calls into question any key finding or otherwise merits reversal. It appears, after all, that RSI paid taxes on all the wagers made by the appellant players. Because PWE has failed to explain the significance of this statement, it at most constitutes harmless error. g. Professional Players’ Quest for an Effective Rate Seventh, PWE argues that the Bankruptcy Court committed a factual error when it stated that “professional players looked for an effective rate—an average rebate rate—that the player could achieve over all the different types of bets the player made.” Id. at 346. PWE argues that no document admitted into evidence referenced an “effective,” “variable,” or “average” rate and that no player other than Cuscuna testified that reaching an effective rate was the goal. To the contrary, the Bankruptcy Court’s finding is supported by sufficient evidence, and this Court is not left with the conviction that a mistake has been committed. Bala and Cuscuna testified about how variable rate rebates work and why they exist, and Wagner himself explained

why flat rate rebates do not work well. It is undisputed that at least one professional player— Cuscuna—was seeking to achieve a particular rate over time under the variable rate model. Regarding PWE specifically, Bala testified that Wagner told her that he was receiving 11% with his previous provider and that he was hoping for something similar with RSI. Wagner acknowledged that his prior rebate was close to 11% and that he may have told Bala that information but did not remember specifically doing so. PWE Exhibit 50, the spreadsheet RSI created during negotiations, shows how PWE’s rebate would vary based on the takeout rate of the pools and tracks he bet on.26 And the summary in Bala Exhibit 1 shows how Wagner’s effective rate over time did end up close to 11%. In short, the court’s finding is not clearly erroneous.

h. Rebate Almost Always Came to 11% Eighth, PWE argues that the Bankruptcy Court made a factual error when it said that PWE’s rebate “almost always came to around 11%.” Racing Servs., 595 B.R. at 348. While the court’s wording perhaps could have been more precise, the reality is not far off. While PWE’s weekly rebate varied from 7% (a clear outlier) to 12.7% during PWE’s time with RSI, the rebate fell within a half percent of 11% in a total of 76 out of 105 weeks. The summary in Bala Exhibit

26 Carlson Exhibit 2 shows the same for Carlson’s wagering. 1 also shows that PWE’s rebate ultimately averaged out to just short of 11%—more precisely, 10.924%. Thus, the actual data does not meaningfully vary from the court’s statement. On a similar note, PWE also asserts that if Bala’s version of the rebating agreement were correct, then PWE was shorted over a quarter of a million dollars. This is simply not the case. According to Bala, PWE was not promised that its rebate would equal 11% every week or

ultimately average out to exactly 11%. Instead, RSI developed a formula that enabled PWE to achieve an effective 11% rebate if PWE met the projected takeout rate, but ultimately let PWE control whether the rebate actually hit 11% based on which pools PWE bet into. As the takeout rate on PWE’s wagering varied from week to week, so too did the rebate rate. Thus, if Bala’s version is true, PWE received all it bargained for each week. i. PWE Not Conferring a Benefit on RSI Ninth and lastly, PWE argues that the Bankruptcy Court erroneously found that PWE could not show that it had conferred a benefit on RSI. More specifically, PWE asserts that it enriched RSI because PWE’s derivative claim efforts helped bring over $15 million back into the

bankruptcy estate and RSI made over $2.8 million in management fees on PWE’s wagering alone. However, even if either of these could constitute a benefit conferred upon RSI, in order to prevail, PWE would still need to meet its burden of proof on the other unjust enrichment elements. Unless this Court were to upset the Bankruptcy Court’s findings on all the other elements—which is not the case—the unjust enrichment claim would still fail on alternate grounds. Any error is thus harmless. j. Overall Review of the Denial of PWE’s Amended Claim To tie up any loose ends with PWE’s appeal, the Court will conclude with a brief look at the Bankruptcy Court’s determinations that Bala overcame the presumption of validity and that PWE failed to meet its ultimate burden of proof. First, the court did not err by finding that Bala presented some evidence to shift the burden of proof to PWE. While Bala’s evidence did consist primarily of her own testimony, she provided a significantly more detailed explanation of how the rebating agreement was reached. Her testimony also was corroborated by Cuscuna’s testimony about variable rate rebates and by the summary of all the weekly income statements.

The evidence also supports the Bankruptcy Court’s finding that PWE failed to meet its burden of persuasion on its claim. Having reviewed the entire appellate record, this Court is left unconvinced that PWE proved its version of the agreement by a preponderance. While PWE’s version is plausible, the contemporaneous evidence relied upon by PWE falls short. The parties allocated all the money that they believed was available at the time. PWE did essentially get the rest of the takeout after RSI’s 1% fee and expenses, but PWE has failed to show that RSI was limited to only that fee if taxes completely fell out of the equation retroactively. The evidence shows that the parties would have renegotiated if the taxes were lowered or eliminated and that the parties did not anticipate the retroactive return of the tax money during their original negotiations.

As such, it is reasonable to determine that the parties’ agreement simply did not provide for how to allocate this money. As far as unjust enrichment goes, the Court simply refers back to its prior analysis in this order. Ultimately, reversal of the Bankruptcy Court’s denial of PWE’s Amended Claim is not warranted on either contractual or equitable grounds. IV. CONCLUSION The Court has carefully reviewed the entire record, the parties’ filings, and the relevant law. The Bankruptcy Court’s rulings on Carlson’s Claim and PWE’s Amended Claim are supported by sufficient evidence, and Carlson and PWE have failed to demonstrate any legal or factual error that would merit reversal. The Bankruptcy Court’s Order is therefore AFFIRMED with regard to both claims. IT IS SO ORDERED. Dated this 18th day of June, 2020.

/s/ Peter D. Welte_________ Peter D. Welte, Chief Judge United States District Court