Opinion
Case No. 14-cr-00534-CRB-1
05-14-2018
UNITED STATES OF AMERICA, Plaintiff, v. JOSEPH J. GIRAUDO and KEVIN CULLINANE, Defendants.
ORDER SETTING SENTENCING GUIDELINES OFFENSE LEVEL FOR DEFENDANTS GIRAUDO AND CULLINANE
Defendants Joseph Giraudo and Kevin Cullinane have pled guilty to criminal violations of the Sherman Antitrust Act, 15 U.S.C. § 1. Unlike the other 21 defendants in this and related cases, Giraudo and Cullinane did not enter into plea agreements with the government. Not surprisingly, they contest the government's calculation of the appropriate offense level under the United States Sentencing Guidelines. The offense level determines the recommended length of custodial sentence under the guidelines.
The Court held a hearing on May 7, 2018, to set each defendant's offense level. The government calculated Giraudo's at 18 and Cullinane's at 17. Giraudo argued that his correct offense level was 9, while Cullinane contended that his was 12. The Court sided with the government's calculation with respect to both defendants. Having determined the appropriate offense level, the Court then proceeded to sentence Giraudo and Cullinane on May 8.
Given the complexity of this issue, the Court stated at the May 7 hearing that it would provide its reasoning as to the offense-level calculation in the form of a written order, rather than on the oral record. That reasoning follows.
I. BRIEF BACKGROUND
Having described the background of this case in detail in an earlier order, see Order Denying Mot. to Suppress (dkt. 248) at 1-5, the Court discusses it here only briefly. Giraudo and Cullinane are two of 23 defendants scheduled to be sentenced by this Court for their involvement in a conspiracy to rig bids at foreclosure auctions in the counties of San Francisco and San Mateo. According to the government, Giraudo and Cullinane orchestrated the scheme along with Ray Grinsell, Daniel Rosenblatt, and Mohammed Rezaian, a group that other bidders sometimes referred to as the "Big Five." The members of the Big Five allegedly conspired to drive down the prices of properties at foreclosure auctions by agreeing with other individuals and groups to refrain from bidding on certain properties. Giraudo and Cullinane, for their part, dispute the notion that they were involved in orchestrating the scheme, saying their roles were no different than those of the other 18 conspirators.
All of the defendants, including Giraudo and Cullinane, pled guilty to entering into a criminal agreement to restrain trade under the Sherman Antitrust Act, 15 U.S.C. § 1. Unlike their co-conspirators, however, Giraudo and Cullinane entered open pleas—that is, they did not plead pursuant to agreement with the government. See Applications to Enter Pleas (dkts. 263, 265). Accordingly, while admitting guilt, they dispute some of the facts on which the government bases its calculation under the guidelines, and challenge some of the government's legal arguments regarding guidelines interpretation.
II. LEGAL STANDARD
The Federal Sentencing Guidelines set out a uniform system of sentencing aimed at reducing disparities in sentences. United States v. Booker, 543 U.S. 220, 250, 253 (2005) (Breyer, J., delivering op. of Court in part); see also 28 U.S.C. § 991(b)(1)(B). To this end, the guidelines establish standards for calculating offenders' culpability in the form of "offense levels" that may be translated into a recommended range of time in custody. These guidelines are advisory, not mandatory. Booker, 543 U.S. at 259. Determining the sentence ultimately rests in the discretion of the district court.
Cullinane and Giraudo agree that § 2R1.1 of the guidelines, which apply to horizontal agreements in restraint of trade, provides the starting point for the analysis. See U.S.S.G. § 2R1.1 background (2017). The base offense level under § 2R1.1 is 12. § 2R1.1(a). The government agrees that Giraudo and Cullinane are entitled to a three-level reduction for accepting responsibility by pleading guilty.
Among other things, the guidelines provide for enhancements based on the magnitude of commerce affected, § 2R1.1(b)(2), and the defendant's role in the offense, § 2R1.1 note 1. They also provide for a one-point enhancement "[i]f the conduct involved participation in an agreement to submit non-competitive bids." § 2R1.1(b)(1).
A district court generally "uses a preponderance of the evidence standard when finding facts pertinent to sentencing," though a different standard applies when a sentencing factor has an "extremely disproportionate effect on the sentence relative to the offense of conviction." United States v. Berger, 587 F.3d 1038, 1047-48 (9th Cir. 2009) (emphasis omitted); see also United States v. Andreas, 216 F.3d 645, 678 (7th Cir. 2000).
III. DISCUSSION
A. Giraudo
1. Volume-of-commerce enhancements
The Sentencing Guidelines provide for enhancements in antitrust cases based on the "volume of commerce done by [a participant in a conspiracy] or his principal in goods or services that were affected by the violation." U.S.S.G. § 2R1.1(b)(2). "Affected" commerce "includes all sales made within the scope of the conspiracy." Andreas, 216 F.3d at 676. The parties disagree on both the proper methodology for calculating the volume of commerce and the factual basis of the calculation.
According to the Sentencing Guidelines, "Tying the offense level to the scale or scope of the offense is important in order to ensure that the sanction is in fact punitive and that there is an incentive to desist from a violation once it has begun." U.S.S.G. § 2R1.1 background. However, "[t]he offense levels are not based directly on the damage caused or profit made by the defendant[,] because damages are difficult and time consuming to establish. The volume of commerce is an acceptable and more readily measurable substitute." Id. The guidelines estimate that the average gain from horizontal agreements in restraint of trade is 10 percent of the total volume of commerce, and the offense-level adjustments are based on that assumption. See § 2R1.1 note 3.
The government calculates Giraudo's volume of commerce at $36,663,335.66, exceeding the threshold of $10,000,000 required for a four-level increase in the offense level. Giraudo, meanwhile, maintains that his volume of commerce was essentially negligible, warranting no increase at all.
a. Treatment of Big Five as single entity
First, Giraudo argues that the government improperly attributes the full purchase price of the properties he was involved in rigging to him, even for properties in which he only retained a partial ownership share. He contends that these sales should instead be allocated proportionately to each buyer.
Giraudo is quite right that the guidelines section applicable to violations of § 1 of the Sherman Antitrust Act, unlike those for other conspiracies, "counts every sale just once." United States v. Heffernan, 43 F.3d 1144, 1147 (7th Cir. 1994) (Posner, J.). In antitrust cases, co-conspirators are not liable for sales made by other co-conspirators solely by virtue of their association. However, the volume-of-commerce calculation does include "commerce done by [a participant] or his principal in goods or services that were affected by the violation." U.S.S.G. § 2R1.1(b) (emphasis added). The law treats members of partnerships as agents of the partnership, with the partnership functioning as the principal. 46 Noah J. Gordon, Am. Jur. 2d Joint Ventures § 36 (2018). So the government's treatment is appropriate.
Giraudo points out that the partnership agreements were not consistent from property to property, sometimes involving a different mix of owners. But this changes nothing: nowhere does § 2R1.1 state that a conspirator may only be liable for commerce done by a single principal. Such a rule would allow co-conspirators to avoid liability by forming an endless series of joint ventures with each other, frustrating the purpose of the antitrust laws.
In his papers, Giraudo made this point only obliquely. See Giraudo Response (dkt. 319) at 17. He pressed it at oral argument on May 7, however. While the argument is probably waived, the Court considers it on the merits because doing so does not prejudice the government, given the outcome of the analysis.
Giraudo also raises factual objections to the government's treatment of his involvement with alleged co-venturers on particular transactions. Those objections are addressed below, along with his other factual objections. See infra Part II.A.1.c.
But wait, Giraudo says: if the government is treating the Big Five as a joint venture, then there was no Sherman Act violation at all, because a single entity cannot conspire with itself to violate 15 U.S.C. § 1. Copperweld Corp. v. Independence Tube Corp., 467 U.S. 772, 777 (1984). True enough. . . . But there were 18 other co-conspirators who were not members of the Big Five. In its volume-of-commerce analysis, the government omits bids on properties on which members of the Big Five did not make an agreement with anyone outside the group. Gov't Sentencing Mem. (dkt. 307) at 8 n.9. In other words, the government did not treat members of the Big Five as co-conspirators with each other. The government's treatment is therefore entirely consistent with Copperweld.
b. Methodological disputes
Giraudo next objects to the government's methodology for calculating his volume of commerce. First, he insists that it is not necessary to calculate the volume of commerce at all, because a reliable measure of the actual gain is close at hand: the amount of the payoffs that co-conspirators accepted in exchange for refraining from submitting bids. Giraudo argues that, for any given property, the payoff will be equal to the difference between the market price (that is, what a property would have gone for had the auction been clean) and the rigged price (what the property actually went for). Indeed, Giraudo and Cullinane argue that, in some cases, the bids were entirely gratuitous, pushing the total payout for the property over the market price.
However, neither Giraudo nor Cullinane have identified any case or treatise supporting the proposition that the harm caused by bid-rigging may be measured solely by the amount of the payment given in exchange for a promise not to bid. Indeed, there is good reason to think that the price the properties would have sold for in a clean market is higher than the combination of the purchase price and payoff—perhaps much higher. For one thing, no rational bidder—and Giraudo had decades of experience in real estate transactions in general, and foreclosure auctions in particular—would choose to make a payoff to another bidder that is equal to the market price minus the purchase price. If the required payoff were that substantial, the bidder would instead simply purchase the property at the market price. There is evidence that this is exactly how members of the conspiracy viewed the auctions. See dkt. 307-7 at 3 ("Fong was not concerned about the payoff agreements as long as the total paid for a property including the payoff was equal to or less than the amount Fong was willing to pay for a property prior to the trustee sale.").
All of the evidence cited in this order comes in the form of summaries of witness interviews prepared by the government. Accordingly, all quotations are of the government's summaries, and do not necessarily reflect the witnesses' actual words.
Moreover, bid-rigging may be accomplished without any payoff at all, but rather with an agreement to "rotate" bids—I win this round, you win next round. (Indeed, that is the usual method.). Here, nearly every member of the conspiracy bid sometimes while agreeing to refrain from bidding at others. Accordingly, it is conceivable that the agreements consisted of reciprocal promises not to bid, in addition to monetary payments. See dkt. 307-8 at 5 ("Fung overheard Giraudo advise other people that Giraudo would allow them to buy a property if they agreed not to buy more property for a period of time."); dkt. 307-8 at 6 ("[B]idders at auctions [took] turns buying property in order to keep the price low.") (emphasis added); dkt. 307-11 at 3-4 ("Giraudo told Lipton that since he (Giraudo) let Lipton buy the Thomas Street property, Lipton should not bid against Giraudo for a property Giraudo wanted to buy."). The evidence that payoffs were often offset against other payoffs suggests that they served at least in part as a bookkeeping method to keep track of how many times bidders had agreed not to bid, rather than as compensation for so agreeing. See dkt. 307-9 at 8-9.
Other evidence indicates that the value of the non-compete agreement dwarfed the amount of the payoffs. For instance, there is evidence that the payoffs were much higher when the bidders involved were not repeat players. A co-conspirator who pled guilty in a related case, Laith Salma, stated that, in his first encounter with the Big Five—presumably before its members knew whether or not Salma would be a repeat player—Cullinane requested $150,000 in exchange for agreeing not to bid on a property. Dkt. 307-19 at 4-5. The scheme's longevity and the persistence with which Giraudo and other members of the Big Five pursued it, despite clear awareness of the legal risks, also belies the claim that it was not lucrative. See dkt. 307-9 at 5 (Grinsell said he was "scared to death" in the late 1990s when he heard a rumor that investigators were focusing on the foreclosure auctions.); dkt. 307-12 at 6-7 (describing changing tactics once Rezaian began to suspect law enforcement was looking into the bidding schemes).
Indeed, there are equally compelling reasons to believe that the volume-of-commerce analysis understates the profit made by Giraudo and his co-conspirators. Bid-rigging schemes at foreclosure auctions are likely to be more profitable than other price-fixing ventures because (1) the public nature of the bids makes it easier to catch and punish those who cheat on the cartel; (2) the cartel underpay does not reduce output (the same number of homes come to auction whatever the sale price); and (3) banks are relatively price-insensitive sellers (indeed, they were apparently oblivious to the conspiracy, which a number of co-conspirators described as obvious to anyone at the auctions, with participants sometimes paying off criers). See Heffernan, 43 F.3d at 1149. Moreover, individuals convicted of price-fixing often participate in cartels on behalf of large corporations, and thus appropriate for themselves only a small percentage of the total profit. See, e.g., Andreas, 216 F.3d at 649. Here, in contrast, the bidders acted on behalf of themselves, or as members of small partnerships. They therefore took home a much higher fraction of the gain than in the ordinary case.
Dkt. 307-10 at 8.
But this discussion only highlights a broader point: "[antitrust] damages are difficult and time consuming to establish." U.S.S.G. § 2R1.1 background. This is precisely why the guidelines calculate harm based on volume of commerce, rather than more direct measures. Id. And indeed, the volume-of-commerce analysis is the only method the Sentencing Guidelines provide for calculating the harm caused. There may be a case in which in which it is appropriate to fashion an alternative method, but if so, this is not it.
Giraudo's next contention is that the portion of the bid amount below the pre-set minimum selling price should be excluded from the volume-of-commerce analysis because the sellers would have made that amount in any event. In other words, that portion was not "affected" by Giraudo's conduct. This is a weird argument. Giraudo might have contended that the entire purchase price should be excluded because the sellers actually received that amount, and were therefore not harmed up to that amount. In other words, he might have argued that "affected" commerce, as used in § 2R1.1(b)(2), means the actual loss, measured by the delta between the market price and the purchase price. Of course, this is plainly not what the guidelines mean by volume of commerce, as evidenced by the explanation in the application notes that the harm resulting from horizontal agreements in restraint of trade represents a small fraction of the total "volume of affected commerce." U.S.S.G. § 2R1.1 note 3.
Giraudo's actual argument is even more perplexing, however. Under his logic, if the market value of a home is $150,000, the minimum bid is $100,000, and a rigged property is purchased for $100,000, the volume of commerce is $0. Meanwhile, if the purchase price on the same property is $150,000, the volume of commerce is $50,000. In other words, where the actual harm is $50,000, the volume of commerce is $0, and where the actual harm is $0, the volume of commerce is $50,000.
One need not be Judge Posner to know that this makes no sense. It is equivalent to arguing that a consumer is not harmed by a price-fixing scheme so long as he is not forced to buy a product at a price above what he is willing to pay. If that were the case, antitrust laws would be a dead letter: buyers and sellers would never be harmed by price-fixing or bid-rigging. Fortunately for the antitrust bar, this is not how the law sees it: a buyer is harmed by price-fixing whenever he is (1) forced to pay more than the market price or (2) dissuaded from buying at a price above what he is willing to pay—not only when he is somehow coerced into paying more than he would otherwise be willing to. So Giraudo's argument that banks were not harmed because they could have withdrawn the property if they did not receive any bids above the pre-set minimum is unavailing.
See U.S.S.G. 2R1.1 note 3 ("The loss from price-fixing exceeds the gain because, among other things, injury is inflicted upon consumers who are unable or for other reasons do not buy the product at the higher prices.").
Giraudo next likens this case to component-price-fixing cases in which courts calculate the volume of commerce based on the price-fixed component, rather than the end product sold to the consumer. He argues that the auctioned properties should really be separated into two—the portion up to the minimum bid (presumably equal to the amount of the encumbrance), and the portion that exceeds the minimum bid—and that he should only be taxed on the latter for properties he rigged. Some commodities can be separated into components because there are in fact two separate sales: the sale of the component to the manufacturer, and the sale of the entire commodity to the end user. If the component manufacturer engages in a price-fixing scheme on the component, then of course the volume of commerce is calculated with regard to only that component. But that is not the situation here: the homes were not sold piecemeal. Giraudo's argument is unavailing.
Next, Giraudo argues that sales which occurred more than five years prior to the filing of the indictment should not factor into the volume-of-commerce calculation because they are outside the statute of limitations. In support of this point, he cites a securities-fraud case concerning whether a statute of limitations applied to claims for disgorgement. Kokesh v. S.E.C., —U.S.—, 137 S. Ct. 1635, 1639 (2017). But the question here is not whether the statute of limitations at issue applies, but rather how. And that question has already been answered in the context of agreements in restraint of trade charged under 15 U.S.C. § 1. Every circuit to rule on this issue has held that "each sale in a price-fixing conspiracy is an overt act that restarts the statute of limitations." In re Pre-Filled Propane Tank Antitrust Litig., 860 F.3d 1059, 1065 (8th Cir. 2017), cert. denied sub nom. Ferrellgas Partners, L.P. v. Morgan-Larson, LLC, 138 S. Ct. 647 (2018) (citing cases). So Giraudo's argument fails.
c. Factual disputes
Giraudo next contends that the government has not put forth sufficient evidence that he actually participated in the transactions it uses to calculate his volume of commerce. The government bases its volume-of-commerce calculation on some 206 transactions in which it claims Giraudo was involved. Having reviewed the evidence, and adopting the government's methodology, the Court finds by a preponderance of the evidence that Giraudo's volume of commerce is more than $10,000,000.
In challenging the factual basis of the government's volume-of-commerce calculation, Giraudo has taken a bit of a now-you-see-it, now-you-don't approach. The government represents that it provided the evidence supporting Giraudo's involvement in the listed transactions to his counsel over three years ago, see Gov't Response (dkt. 358) at 2 n.2, and gave his attorneys a list of the transactions cross-referenced to exhibits three months ago, id. at 8. Giraudo did not rebut these representations at oral argument.
Nevertheless, in his sentencing memorandum, Giraudo made only minor objections to the government's list of transactions, noting in footnotes that he had found several "discrepancies" and "factual errors." See Giraudo Sentencing Mem. (dkt. 313) at 20 n.30, n.34. Then, after receiving the government's sentencing memorandum—which included a chart of the allegedly rigged properties and cross-referenced (without actually attaching) supporting evidence—Giraudo responded that "the government has failed to offer a single piece of evidence to the Court to support the 206 property entries listed" as part of his volume of commerce. Giraudo Response (dkt. 319) at 14.
In response to this argument, the government duly served Giraudo and the Court with a thumb drive containing the referenced evidence. Gov't Reply (dkt. 358). After receiving the thumb drive, on the morning of the hearing set for determining the appropriate guidelines range, Giraudo identified 13 additional properties which he contended he had no role in purchasing. Reply (dkt. 363) at 2. Nevertheless, he did not ask for either a continuance or an evidentiary hearing to allow him further opportunities to contest the government's evidence. Instead, he urged the Court to refuse to consider the government's submission as untimely filed and find that his volume of commerce was $0.
Insofar as Giraudo moves the Court to exclude the evidence on the ground that the Court lacked sufficient time to review it, that motion is denied. Having reviewed the materials, the Court is satisfied that they are reliable and that the government has summarized them accurately. In the absence of specific objections by Giraudo that, if credited, would reduce the volume of commerce below $10,000,000, this is all that is required. See United States v. Gerald, 71 F. App'x 231, 232 (4th Cir. 2003).
The arguments Giraudo included in his reply to the government's response to his sentencing memorandum regarding specific transactions are waived. Giraudo had the evidence on which the government based its calculation for three years, and had the specific list of transactions the government considered for three months. The government also filed the list as an attachment to its own sentencing memorandum, which Giraudo could have addressed in his response. In any event, even if the Court were to exclude all of the transactions that Giraudo lists in his three briefs, the volume of commerce would still be well over $10,000,000. --------
Giraudo's contention that the Court should exclude the evidence on the ground that the government failed to give him notice consistent with due process also fails. Giraudo notes that "due process requires . . . that the defendant be given an opportunity to refute . . . . any information presented at sentencing." United States v. Giltner, 889 F.2d 1004, 1008 (11th Cir. 1989). But he does not cite any cases in which a court held that this information must be disclosed to the defendant via the court's filing system, rather than through discovery. He cites a case in which a court excluded evidence of sales used to calculate volume of commerce as untimely where the government submitted the evidence after the court had already determined the volume of commerce. United States v. SKW Metals & Alloys, Inc., 6 F. App'x 65, 66 (2d Cir. 2001). That case is obviously inapposite: here, the government submitted its evidence prior to the hearing at which the Court was to determine the guidelines range.
Even if the Court were to find that Giraudo lacked adequate notice of the evidentiary basis for the government's calculation of the volume of commerce, it would hold that the proper remedy would have been a continuance, not exclusion of the evidence. However, Giraudo's attorney emphatically stated at the May 7 hearing that he was not requesting a continuance. In doing so, he waived any argument that his client was entitled to a court-ordered remedy for the government's alleged delay in providing the evidence. Giraudo wants to have his cake, eat it too, and share it with his friends; but his lawyers, excellent as they may be, are not magicians.
2. Bid-rigging enhancement
Sentencing Guidelines § 2R1.1 provides for a one-point enhancement for defendants who participate "in an agreement to submit non-competitive bids." § 2R1.1(b)(1). The government argues that this enhancement applies to Giraudo's conduct. Giraudo counters that it only applies where defendants agree to rotate bids without making payoffs to each other. He cites Judge Posner's opinion in Heffernan for this proposition. See 43 F.3d at 1147.
The panel in Heffernan held that the bid-rigging enhancement in § 2R1.1 is meant to capture the scenario in which a defendant agrees not to make a bid, "hence has not made a sale, hence has no volume of commerce." Id.; see also U.S.S.G. § 2R1.1 note 6. The panel held that the enhancement did not apply in the circumstances of that case, in which the defendants had agreed to submit matching bids. Heffernan, 43 F.3d at 1147-50. Giraudo's case is readily distinguishable: he frequently agreed not to bid. Heffernan made no distinction between cases in which the agreement not to bid involves a payoff, and cases in which it does not—and such a distinction would not make sense. In either case, the party on the other side of the transaction is harmed because the winning bid is artificially high or low; but the volume-of-commerce calculation does not account for this fact with regard to the party who does not bid. And there are other reasons why agreements to submit noncompetitive bids tend to be more harmful than agreements to submit matching bids or offers. See supra Part II.A.1.b.; see also Heffernan, 43 F.3d at 1149. Accordingly, to the extent Giraudo contends that the bid-rigging in this case was not particularly harmful compared to bid-rigging that does not involve payoffs, this argument is unavailing.
In the alternative, Giraudo argues that the payoffs he received for not bidding are reflected in the requirement that he pay restitution, and that he shouldn't be punished further. But the guideline increase only bears on the recommended length of the custodial sentence. There is nothing in the guidelines to indicate that the applicability of the enhancement in § 2R1.1(b)(1) depends on the magnitude of the fine levied on the defendant.
3. Role enhancement
The government argues that Giraudo is deserving of a four-point enhancement for his role as "an organizer or leader of a criminal activity that involved five or more participants or was otherwise extensive." U.S.S.G. § 3B1.1; see also § 2R1.1 note 1. The guidelines advise courts to consider the following factors in determining whether this enhancement is warranted: (1) the exercise of decision-making authority, (2) the nature of participation in the commission of the offense, (3) the recruitment of accomplices, (4) the claimed right to a larger share of the fruits of the crime, (5) the degree of participation in planning or organizing the offense, (6) the nature and scope of the illegal activity, and (7) the degree of control and authority exercised over others. § 3B1.1 note 4.
a. Applicability of enhancement
Giraudo first appears to argue that this enhancement is not applicable to white-collar criminals: "The four-level enhancement is intended for sophisticated drug cartels or organized crime syndicates, not octogenarian real estate investors." Sentencing Mem. at 13. But the notes to the guidelines for bid-rigging make express reference to the role-enhancement provision. U.S.S.G. § 2R1.1 note 1. So this argument fails.
b. Organization of the cartel
Giraudo next contends that the enhancement does not apply because there was no "organization" to speak of: "This was a disorganized group of people who would show up at auctions and make ad hoc agreements to bid or not bid when an opportunity arose." Sentencing Mem. at 13-14. Cullinane extends this argument in his brief, contending that "[n]o one was empowered to tell another when or how much to bid; each participant made those decisions for his/her self." Cullinane Sentencing Mem. (dkt. 297) at 5. Cullinane continues: "In reality, [the participants] were rivals who agreed on the structure for their rivalry but acted independently within it." Id. Certain bidders may have had more influence or resources than others, but "each participant crafted his or her own strategy and made bids on the basis of self-interest." Id. at 5-7.
Price-fixing cartels need not always have leaders in order to function effectively. Indeed, "some markets are by their structure so conducive to collusion-like behavior that it is to be expected even in the absence of a legal 'agreement' among the parties." 12 Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law: An Analysis of Antitrust Principles and their Application ¶ 2002a (3d ed. 2012). But it is almost inconceivable that a cartel of the type at issue here could agree to rig bids without a strong management structure. Schemes such as this one require a high degree of organization and coordination. Id. at ¶ 2002f4. While agreeing to charge the same price for commodity products may require only a single agreement, with periodic updates to account for market fluctuations, see, e.g., Andreas, 216 F.3d 645, successive bid-rigging requires many (here, hundreds) of successive accords. And each agreement was likely more difficult to reach, given that members were agreeing not just to buy at a particular price, but to entirely forego the benefits of bidding in most transactions. In sum, it is extremely unlikely that cartel members spontaneously agreed when to bid and when not to bid, without any cajoling or assurances of future benefits from cartel leaders.
Controlling cheaters on a large cartel also requires strong management. See Areeda & Hovenkamp, Antitrust Law, supra, at ¶¶ 2002d, 2002d2, 2002f1 ("[T]he managers of the cartel must be vigilant about detecting cheating and disciplining the cheater."). Cheating on the cartel price becomes more profitable as the group becomes larger, and strong leadership is required because there are more members who need to be kept in line. See id. ¶ 2002d ("[C]artels become less stable as the number of members increases."). It is not in the interest of any individual bidder to try to out-bid a renegade without an understanding that he will be compensated for doing so. See id.; see also United States v. Romer, 148 F.3d 359 (4th Cir. 1998), cert. denied, 525 U.S. 1141 (1999). Strong management solves this collective-action problem.
Guidelines determinations require more than cartel theory, of course: they also need actual evidence. The theory helps, however, to put the evidence into context. Here, the evidence overwhelmingly establishes that the Big Five managed the conspiracy, getting others involved and disciplining those who did not get with the program. According to Michael Navone, "It was known that Rezaian, Grinsell, Giraudo, Rosenbledt, and Cullinane worked together and were known as 'the group' or 'the big five.'" Dkt. 307-13 at 5; see also 307-8 at 6 (testimony of Florence Fung, describing the group as the "Board of Directors"). According to Salma, the Big Five set the rules of the bidding among co-conspirators and determined how much of a share each bidder would get in a given property, often dividing the shares unequally to give themselves a bigger cut of the pie. Dkt. 307-22 at 13. "Giraudo . . . explained to Salma the rule that if you were interested in a property, you had to tell Giraudo or someone else in the Big Five, and the Big Five made the determination as to who got the property." Dkt. 307-22 at 5. According to Grinsell, Giraudo kept a list tracking who had been paid to refrain from bidding on various properties. Dkt. 307-9 at 7. Later, Rezaian took on this role. Dkt. 307-16 at 1-2.
The Big Five also took responsibility for disciplining would-be cheaters on the cartel and threatening those who declined to participate. According to Salma, Giraudo and other members of the Big Five would bid up properties to punish those who did not follow the bid-rigging rules set by the Big Five, or would threaten to do so. Dkt. 307-22 at 5-6. According to Navone, "It was common for a person not associated with the group to have difficulty purchasing property . . . unless [he] agreed to pay the group." Dkt. 307-13 at 5. In sum, the evidence shows that this was a well-organized cartel in which the members of the Big Five played a managerial role.
c. Giraudo's role
Giraudo next argues that the leadership enhancement does not apply because he did not enjoy a leadership role within the Big Five. Rather, he and his four partners "operated as equals." But the evidence shows that Giraudo made the final decision over which co-conspirators could join the partnership; that Giraudo settled disputes within the conspiracy; that Giraudo gave direction to other members of the Big Five; and that other members of the Big Five regarded Giraudo as the leader.
Giraudo's representation to the Court that he only began bidding at the auctions during the financial crisis, apparently joining an already existing conspiracy, is belied by the statements of numerous co-conspirators that he had been bid-rigging much earlier, with a rotating cast. See, e.g., dkt. 307-9 at 4-5 (Grinsell); dkt. 307-6 at 2-3 (Abraham Farag). Rezaian, Rosenbledt and Cullinane only joined Giraudo and Grinsell once the foreclosure crisis was underway in 2009, when more people began attending the auctions. Dkt. 307-10 at 4 (According to Grinsell, the number of attendees more than doubled in 2009, from six to twelve or fifteen.). Grinsell maintained that Giraudo bullied him into joining the conspiracy, and admitted others to the "Big Five" partnership over Grinsell's objections. Dkt. 307-9 at 8-9. The members of the Big Five may have worked collaboratively in certain ways, but when there were disagreements, the other members consistently deferred to Giraudo. According to Rezaian, "Everyone at the auctions looked to Giraudo to make deals." Dkt. 307-16 at 4.
According to Salma, other co-conspirators referred to Giraudo as the "boss," "don," or "king" of the auctions: "From attending the auctions, one learned Giraudo was the boss and his position of being in charge was clear. Those attending the auctions learned that you did not 'fuck' with Giraudo." Dkt. 307-22 at 6. Cullinane would tell other bidders, "Don't make Joe angry." Dkt. 307-22 at 6-7. Giraudo would bully and shame other bidders, calling one the "Bad Indian." Dkt. 307-22 at 5-6. One bidder labeled the Big Five "Joe's Group." Dkt. 307-23 at 5.
Giraudo argues that Rezaian was the real ring-leader, pointing to Rezaian's role as an enforcer and intimidator, and his active direction of co-conspirators. Giraudo Response at 12-13. The evidence clearly indicates that Rezaian played an important part, orchestrating bidders and often threatening them. Grinsell noted that Giraudo decided to partner with Rezaian partly in order to "eliminate a competitor" (meaning Rezaian), a fact which suggests that Rezaian had significant leverage. Dkt. 307-10 at 4. But Grinsell also stated that "Rezaian went along with anything Giraudo said," and that "Rezaian would not have opposed Giraudo as it would have undermined Rezaian wanting to become Giraudo's friend and right hand man." Dkt. 307-10 at 3.
Rezaian may well have qualified for the leadership enhancement, had the government requested it. But "[t]here can, of course, be more than one person who qualifies as a leader or organizer of a criminal association or conspiracy." U.S.S.G. § 3B1.1 note 4. That other members also played crucial roles does not establish that Giraudo was not nonetheless a leader, and he has pointed to no evidence indicating that he took orders from anyone.
Giraudo contends that his co-conspirators' statements regarding his involvement are unreliable because they were not subject to cross-examination. But Giraudo does not point to any specific inconsistencies, and "hearsay is admissible at sentencing, so long as it is accompanied by some minimal indicia of reliability." United States v. Littlesun, 444 F.3d 1196, 1200 (9th Cir. 2006). At the May 7 hearing, Giraudo's counsel contended that the government has hung its entire argument on the testimony of a single co-conspirator, Laith Salma. But it should be evident from the foregoing summary of the record that a bevy of other co-conspirators also spoke to Giraudo's leadership role within the Big Five, and the conspiracy as a whole. He is deserving of the enhancement.
B. Cullinane
Cullinane and the government disagree about the proper calculation of the volume of commerce affected by him. Cullinane also argues that an enhancement based on his role in the offense is improper. The Court takes these arguments in turn.
1. Volume of commerce
Like Giraudo, Cullinane argues that the Court should allocate his volume of commerce based on his ownership stake in the rigged properties, and that the volume-of-commerce calculation provided for in the guidelines overstates the actual profit he made. The Court rejects these arguments for the reasons already stated with respect to Giraudo. See supra Part II.A.1.
Cullinane also argues that the volume-of-commerce enhancement is only meant to address "collusion among providers of services or products to extract a higher price from consumers for those services and products." Cullinane Sentencing Mem. at 4. The enhancement is not designed for cases like this one, in which the agreements "served to reduce the profits made by banks and lenders on their foreclosed mortgages." Id.
It is not quite clear what rationale Cullinane thinks supports this conclusion. He may be saying that the guidelines only apply to upstream (or selling) cartels, not downstream (or buying) ones. But "a buying cartel's low buying prices are illegal and bring antitrust injury and standing to the victimized suppliers. Clearly mistaken is the occasional court that considers low buying prices pro-competitive or that thinks sellers receiving illegally low prices do not suffer antitrust injury." Knevelbaard Dairies v. Kraft Foods, Inc., 232 F.3d 979, 988-89 (9th Cir. 2000) (quoting 2 Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law ¶ 375b at 297 (rev. ed. 1995)). "To hold otherwise would be contrary to long-established antitrust law," id., not to mention binding Ninth Circuit precedent, see id.
Cullinane's attorney offered a different logic at the May 7 hearing, maintaining that the volume-of-commerce enhancement does not contemplate the conduct at issue here because that conduct only harmed lenders, not homeowners—and that it was the abusive lending practices of the banks themselves which precipitated the widespread sales of homes at foreclosure auctions like the ones rigged by the defendants in this case. Cf. dkt. 307-22 ("Salma recalled Giraudo always saying 'fuck the banks.'"). But while civil antitrust law does distinguish between the directness of the harm suffered by plaintiffs in determining who has standing to sue, see Assoc. Gen. Contractors of Calif., Inc. v. Calif. State Council of Carpenters, 459 U.S. 519, 545 (1983), the victim's conduct is otherwise of no account. "[T]he Sherman Act does not authorize horizontal price conspiracies as a form of marketplace vigilantism." United States v. Apple, Inc., 791 F.3d 290, 332 (2d Cir. 2015).
2. Role enhancement
Cullinane argues that he is not deserving of a three-level role enhancement as "a manager or supervisor." U.S.S.G. § 3G1.1(b). To the extent he is contending that the Big Five did not manage the conspiracy, that argument fails for the reasons discussed above. See supra Part II.A.3.b.
Cullinane also argues that he did not personally play a sufficiently significant role within the group to warrant the enhancement. He argues that the government has "fail[ed] to identify any specific conduct by [him] that satisfies the requirements of U.S.S.G. § 3B1.1(b) or to provide any evidence for the attribution to him of responsibility for the specific conduct of others." Cullinane Sentencing Mem. at 7. Cullinane argues that his role within the Big Five was "clearly secondary," noting that he principally researched the status and value of properties and "rarely participated in negotiations and agreements." Id.
Cullinane was the last to join the group, and the government does not dispute that he had less sway within it than the others. See dkt. 307-10 at 4-5. Nevertheless, he played an important role within the Big Five, managing the real estate it acquired, and shared equally in the profits on most properties. Id. at 5-6. This is sufficient for the Court to find by a preponderance of the evidence that he is deserving of the three-level enhancement. In addition, the application notes to Guideline § 3B1.1 state than "[a]n upward departure may be warranted . . . in the case of a defendant who did not organize, lead, manage, or supervise another participant, but who nevertheless exercised management responsibility over the property, assets, or activities of a criminal organization." U.S.S.G. § 3B1.1 note 2. According to Grinsell, Cullinane managed and held title to most of the properties. Dkt. 307-10 at 5-6. So in the alternative, the Court finds that Cullinane is deserving of the enhancement based on his role in managing properties the Big Five acquired through rigging bids.
IV. CONCLUSION
For the foregoing reasons, the Court finds pursuant to the Sentencing Reform Act of 1984 that Giraudo's offense level is 18 and Cullinane's is 17.
IT IS SO ORDERED.
Dated: May 14, 2018
/s/_________
CHARLES R. BREYER
United States District Judge