Opinion
1:18-cr-00258-BLW
08-30-2023
UNITED STATES OF AMERICA, Plaintiff, v. PAVEL BABICHENKO, PIOTR BABICHENKO, TIMOFEY BABICHENKO, DAVID BIBIKOV, MIKHAIL IYERUSALIMETS, and ARTUR PUPKO, Defendants.
MEMORANDUM DECISION & ORDER RE RESTITUTION
B. LYNN WINMILL, U.S. DISTRICT COURT JUDGE
INTRODUCTION
Following a jury trial, five of the ten defendants in this matter-Pavel, Piotr, and Timofey Babichenko, David Bibikov, and Mikhail Iyerusalimets-were convicted of conspiring to commit wire fraud and traffic in counterfeit goods. Four of the five were also convicted of discrete acts of wire fraud, mail fraud, and trafficking in counterfeit goods. A sixth defendant, Artur Pupko, pleaded guilty to three counts of trafficking in counterfeit goods. At sentencing, the Court ordered these defendants to pay restitution but deferred deciding the amounts until after conducting an evidentiary hearing.
At the restitution hearing, the government asked the Court to order $15,887,176.52in restitution, broken down as follows: $4,799,880 to Apple Inc.; $9,225,607.61 to Samsung Electronics Co., Ltd. and Samsung America, Inc. (collectively, “Samsung”); and $1,861,688.91 to UL LLC. The Court will largely deny this request, as the government failed to demonstrate that Apple and Samsung lost profits or that UL was deprived of testing and certification fees by virtue of defendants' conduct. The Court will also deny the request to reimburse Samsung and UL for expenses they incurred participating in the government's investigation and prosecution of this case, though this portion of the request will be denied without prejudice. Before awarding such expenses, Samsung and UL will need to submit billing records for their attorneys, and UL will need to provide the hourly rates paid to the salaried employees who participated in the case.
The government actually requested $15,887,176.50, see Gvt. Br., Dkt. 1834, at 1, but the individual components of the restitution request add up to $15,887.176.52.
BACKGROUND
In August 2018, a grand jury indicted ten defendants, charging three overarching conspiracies: (1) conspiracy to commit wire fraud; (2) conspiracy to traffic in counterfeit goods; and (3) conspiracy to launder money. Additionally, defendants were variously charged with discrete acts of mail fraud, wire fraud, money laundering, and trafficking in counterfeit goods. Broadly, the government alleged that defendants sold counterfeit cell phones and accessories on online platforms such as Amazon and then laundered the proceeds.
In September 2019, a little over a year after the indictment was filed, Defendant Artur Pupko pleaded guilty to three counts of trafficking in counterfeit goods. The remaining nine defendants, all of whom are members of an extended family,proceeded to trial, which took place during the summer of 2021. After the government rested, the Court granted Defendant Gennady Babitchenko's Rule 29 motion and dismissed all charges against him. See Dkt. 1120. The Court also dismissed two additional money-laundering charges against Piotr Babichenko and David Bibikov. See Aug. 12, 2021 Trial Tr. (Day 28), Dkt. 1113, at 5315. Otherwise, the Court denied or reserved ruling on the Rule 29 motions. Before defendants began presenting evidence, however, the government dropped all remaining money-laundering counts. See Dkts. 1114, 1117.
Defendants Gennady, Pavel, Piotr, Timofey, and Anna are siblings. Three defendants are spouses of these Babichenko siblings: Natalya Babichenko is Pavel's spouse, Kristina Babichenko is Timofey's spouse, and Mikhail Iyerusalimets is Anna's spouse. David Bibikov is a cousin.
The jury began deliberations on the slimmed-down case in late August 2021 and returned verdicts shortly thereafter. The jury fully acquitted Defendant Natalya Babichenko, returned a handful of not-guilty verdicts on some counts against three other defendants, and deadlocked on the remaining counts. The Court declared a mistrial on the deadlocked counts.
The case was retried in the summer of 2022, this time with seven defendants rather than nine. After another lengthy trial, the jury fully acquitted two defendants (Kristina Babichenko and Anna Iyerusalimets) and convicted the other five on most counts. To recap, then, five of the original ten defendants were convicted, four were acquitted, and one pleaded guilty.
In March 2023, the Court sentenced the defendants. Pavel Babichenko was sentenced to 72 months' incarceration and fined $21,000. Piotr Babichenko was sentenced to 48 months' incarceration and fined $10,000. Timofey Babichenko were sentenced to 48 months' incarceration and fined $10,500. Mikhail Iyerusalimets was sentenced to two months' incarceration. David Bibikov was sentenced to one month's incarceration. Artur Pupko was sentenced to five years' probation and fined $10,000. The Court also entered forfeiture orders, which included the following money judgments:
a. A judgment of $33,708,700.30 against Pavel Babichenko;
b. A judgment of $3,316,882.40 against Piotr Babichenko;
c. A judgment of $9,230,196.82 against Timofey Babichenko;
d. A judgment of $953,411.95 against Mikhail Iyerusalimets; and
e. A judgment of $4,710,207.11 against David Bibikov.See Dkt. 1838, at 46; Dkt. 1793, at 18. The final, sentencing-related task is to determine the amount of the restitution orders.
GOVERNING LEGAL STANDARD
If a defendant is convicted of certain “offenses against property,” including trafficking in counterfeit goods, the Mandatory Victims Restitution Act (MVRA) U.S.C. §§ 3663A-3664, applies and the Court is required to order the defendant to pay restitution to the victims. See 18 U.S.C. § 3663A(c)(1)(A)(ii); see also 18 U.S.C. §§ 2320(c), 2323. The basic goal under the MVRA is to make victims whole. See United States v. Anderson, 741 F.3d 938, 951 (9th Cir. 2013). As such, a restitution award is limited to the victim's “actual losses” that were the direct and proximate result of the defendant's offense. See id.
The government has the burden of establishing, by a preponderance of the evidence, that the victim's damages were caused by the defendant's conduct. United States v. Rice, 38 F.3d 1536, 1540 (9th Cir. 1994); 18 U.S.C. § 3664(a), (e). Although the Court has wide discretion and “flexibility” in fashioning restitution orders, “any award is limited to the victim's actual losses” as demonstrated by reliable evidence. United States v. Anderson, 741 F.3d 938, 951 (9th Cir. 2013).
To determine actual loss for restitution purposes, the Court compares what actually happened with what would have happened if the defendant had acted lawfully. See generally United States v. Hunter, 618 F.3d 1062, 1064 (9th Cir. 2010). In cases involving intellectual property infringement, the victim's actual loss is typically the victim's lost profit from sales because customers purchased the defendant's goods while believing they were buying from the victim. See, e.g., Anderson, 741 F.3d at 951 (citing United States v. Fair, 699 F.3d 508, 514 (D.C. Cir. 2012)). Thus, the central aim of a “lost-profits on diverted-sales theory” is to figure out how many sales the counterfeiters diverted from the victim. In tallying up the victim's lost sales, however, the Court cannot simply assume that customers who purchased counterfeit devices from defendants would have purchased full-priced versions from Apple or Samsung. Rather, “the government must offer sufficient evidence to establish both the profit margin per sale and the number of sales lost.” Id. at 952 (citing Fair, 699 F.3d at 514).
DISCUSSION
1. Requests for Reimbursement of Lost Profits and Testing & Certification Fees
The Court will begin with the government's request that Apple and Samsung be reimbursed for lost profits and then turn to UL.
A. Apple's & Samsung's Requests for Lost Profits
Apple says defendants' sales of counterfeit goods caused it to lose nearly $4.8 million in profits. Samsung says it lost nearly $9 million. Both companies offer straightforward loss theories. They say each counterfeit item sold by a defendant caused them to lose a corresponding sale of the same item. For example, in tallying up its losses, Apple starts with the fact that defendants sold 1,336 counterfeit iPhone 3s. See Rollins Dec., Dkt. 1834-1, ¶ 3. To calculate its loss for those sales, Apple says it lost the gross profit it would have made from selling 1,336 genuine Apple iPhone 3s. That same methodology was applied across the board - for all types of Apple products defendants sold. Samsung followed the same methodology. See Diaz Dec., Dkt. 1843-3, ¶ 6; June 14, 2023 Hg. Tr., Dkt. 1852, at 56:14-19. In other words, both companies used the “lost profits on diverted sales” approach described above. See Anderson, 741 F.3d at 951.
The problem is that the government failed to demonstrate that any given customer who purchased a counterfeit device from a defendant would have purchased a full-priced, corresponding item from Apple or Samsung. Not only is that assumption unproven, it's quite unlikely, given the evidence at trial. To be sure, the government points to various emails showing that many of defendants' customers were shopping for new, genuine phones. See Gvt. Memo, Dkt. 1834, at 3. The government also points to a sampling of other evidence offered at trial to support its request. See Gvt. Reply, Dkt. 1843, at 4. Then, without elaboration or explanation, the government contends that this sampling “more than satisfies the requirements for restitution of lost profits through diverted sales.” Id. The Court concurs with the defense on this point - these unexplained evidentiary citations fail to support the government's argument. See Bibikov Br., Dkt. 1859, at 5 n.3.
Moreover, the larger point is that defendants' customers typically were not shopping in the primary market, where Apple and Samsung sell phones. Rather, they were shopping in the secondary market, where customers could purchase older model phones at reduced prices. That defendants' customers were shopping in the secondary market raises two issues. First, even assuming that each customer who purchased from defendants would have purchased a new, genuine item from Apple or Samsung, it does not appear that that they could have done so because Apple and Samsung typically were not selling the item the customer had purchased. For example, returning to those 1,336 would-be Apple customers discussed above-the ones who purchased counterfeit iPhone 3s from defendants-it's not been proven that Apple was even selling iPhone 3s during the times when these customers bought those phones. This point was clarified during the evidentiary hearing, when an Apple representative offered the following testimony:
Q: ... [W]ere all of these phones, looks like close to 30,000 phones, were all of those phones at the time of sale, was Apple still selling that same model?
A: Sitting here right now, I don't have a specific answer for you. I believe there were certain instances when a new phone model, for example, the iPhone 3 was not being sold at the same time period. I do believe that there were refurbished products that were being sold in that relevant time period.
Q: I'm referring only to new product.
A: I believe there were certain instances.
Q: Where they were not?
A: Where they were not.
Q: And others where they were?
A: Yes, sir.
Q: You don't have them broken down in some way?
A: I have not broken that down.June 14, 2023 Hearing Tr., Dkt. 1852, at 20:1-23.
Likewise, Samsung did nothing to account for the fact that defendants were selling older model phones that were behind the release date by around two or more years. See Restitution Hg. Tr., Dkt. 1852, at 54:1 to 55:12 & 72:11 to 73:14 (the accountant who calculated Samsung's losses testified that he did not know if Samsung was selling certain Samsung devices at the same time defendants were selling counterfeit versions of those devices).
This failure of proof, standing alone, is enough to defeat Apple's and Samsung's requests for restitution. That brings us to the second issue raised by the fact that defendants were selling in the secondary phone market: Even assuming defendants' customers could have purchased a new device from Apple or Samsung, the more critical question is, “Would they have?” Or, perhaps a more factually accurate question is this: Would these customers have bought a different, later-model genuine phone from Apple or Samsung (for example, an iPhone 5 rather than an iPhone 3) if defendants had not been offering their counterfeit devices for sale? Or would these customers have decided, instead, to get a less expensive, refurbished device? Or would they have gone to a different, cheaper alternative altogether?
The Court has nothing to grab onto here, in terms of what these customers would have done. It bears repeating that courts cannot just assume that a customer who purchased a counterfeit device at a discounted price would have instead purchased an authentic version at full price. See Anderson, 741 F.3d at 953. In United States v. Hudson, 483 F.3d 707, 710 (10th Cir. 2007), the Tenth Circuit reversed a district court's restitution order, stating that it was “very skeptical of the implicit suggestion” that a customer who purchased copies of low-priced counterfeit software proves that that customer would have agreed to purchase the same number of copies from the legitimate seller for many times more. The D.C. Circuit approvingly cited Hudson in United States v. Fair, 699 F.3d 508 (D.C. Cir. 2013), and the Ninth Circuit, in turn, relied on Fair in deciding United States v. Anderson, 741 F.3d 938, 951 (9th Cir. 2013).
In Anderson, the defendant was charged with copyright infringement after selling Adobe software advertised as “full,” “complete,” or “OEM” (original equipment manufacturer). See 741 F.3d at 942. The district court accepted the government's straightforward loss theory - essentially agreeing that restitution would be calculated simply by multiplying the number of copies defendant sold by Adobe's retail price. The Ninth Circuit reversed and remanded, explaining:
We ... recognize that applying the MVRA can be complex and somewhat burdensome. The MVRA requires restitution for certain victims but also requires some precision when calculating restitution. Speculation and rough justice are not permitted. Nonetheless, exact precision is not required and district courts do have a degree of flexibility in accounting for a victim's complete losses; thus a “reasonable estimate” will suffice. Ultimately, however it is the government's burden to establish that the victim suffered an actual loss in a quantifiable amount. Where the alleged loss is not quantifiable to any degree of certainty, the government's burden has not been satisfied, and no restitution should be ordered.Id. at 954 (internal citations omitted, emphasis added). The Anderson Court then echoed Hudson and Fair's concerns with assuming that a defendant's customers would purchase full-priced items from the victim, observing that the district court had failed to consider “whether purchasers of Anderson's discounted software would have, in fact, purchased full price authentic software if Anderson's software had not been available.” Id. at 953-54.
Here, the government has not met its burden of proving that defendants' customers would have been willing to spend more money on new, genuine products from Apple or Samsung if defendants' products had not been available. The government could have addressed this failure of proof - perhaps by surveying a representative sampling of defendants' customers to find out what they would have done. The D.C. Circuit discussed a similar failure of proof in United States v. Fair, 699 F.3d 508 (D.C. Cir. 2012). In that case, as in Anderson, the defendant was convicted of selling illegal copies of Adobe software, and the government had acknowledged that a market existed for “people who are likely to buy older but genuine products.” Id. at 516. The Fair Court observed that Adobe conceivably could have surveyed defendant's customers to determine how many likely would have purchased full-cost, updated versions of those Adobe products. Id. The same is true here. The government didn't do that (or anything else) to demonstrate that defendants' customers would have turned to Apple or Samsung for new, genuine devices.
The government attempts to distinguish Anderson by arguing that the district court there had a “barren” record, whereas in this case, the Court has ample evidentiary support to find defendants' customers would have purchased genuine devices from Apple or Samsung if defendants hadn't been selling their counterfeits on Amazon. See Gvt. Br., Dkt. 1857, at 2. But to support this assertion, the government mainly relies on the rather unremarkable fact that-in a counterfeiting case-defendants advertised their wares as being “genuine.” The government says this is “direct proof” that Apple and Samsung lost profits from diverted sales because the Amazon sales reflect “the customers' own stated desire to purchase genuine goods and the Defendants' fraudulent assurances to these customers that they were buying genuine Apple and Samsung products.” Id. at 4.
This is not a principled distinction from Anderson. The overriding point in both cases is that the consumers who purchased the counterfeits were paying significantly reduced prices. Plus, although Anderson didn't advertise his infringing software as being genuine, he did advertise it as being “full,” “complete,” or “OEM.” Clearly, OEM (an acronym for “original equipment manufacturer”) strongly suggested to the consumer that the Adobe software was genuine. And in United States v. Fair, 699 F.3d 508, 511 (D.C. Cir. 2012), which Anderson relied upon, the defendant advertised his products as “genuine.” So the government is straining when it says, “In Anderson, unlike in this case, there was no evidence of consumers' stated desire to purchase genuine software, nor did Anderson engage in any deception that the much-reduced pirated software was genuine.” See Gvt. Br., Dkt. 1857, at 4 (emphasis added). In that passage, the government makes it sound as though Anderson was obviously selling infringing copies of Adobe-and that's all Anderson's customers were after-whereas the defendants in this case are different because they were deceptive and their customers were looking for genuine devices. But in both cases, the defendants engaged in deception and, in both cases, some customers complained about having been duped and some demanded refunds when they figured out that they had not purchased genuine products. See Anderson, 741 F.3d at 942; see also, e.g., June 27, 2022 Trial Tr. (Day 19), Dkt. 1560, at 3184:23 to 3185:18 (Defendant Artur Pupko testifies about customer complaints); Gvt. Restitution Ex. 4, Dkt. 1843-1; Trial Exs. 4025, 4027, 4044, 4069.
Likewise, the Court is not persuaded by the government's argument that defendants were selling in the same market as Apple and Samsung. To shore up this argument, the government asserts that the price difference between the products defendants were selling vis-a-vis those sold by Apple or Samsung does not show there were two separate markets. See Gvt. Reply, Dkt. 1843, at 2; Dkt. 1857, at 4. The evidence and arguments at trial, however, seriously undermine that assertion. Throughout the trial, the parties acknowledged the existence of the secondary phone market, and there was ample evidence that the defendants sold their products in that marketplace - at a discount. As just a few representative examples, witnesses offered the following testimony during the trial:
• Defendant Piotr Babichenko testified that he sold older model phones on Amazon, priced at 30 to 40% less than the retail price of those phones. July 14, 2022 Trial Tr. (Day 26), Dkt. 1582, at 4780:1-5.
• Nikolay Bukhantsov testified that while working for defendants, he sold phones that were not the latest models from Apple and Samsung but were instead “a couple years behind,” and, further, that these phones were not sold at “retail price.” July 12, 2022 Trial Tr. (Day 24), Dkt. 1580, at 4333:22 to 4334:9.
• Defendant Artur Pupko testified that during the five years he sold phones for Pavel Babichenko (2013 to 2018), he sold older phones - “you know five or four years or so - even more . than three years” although there might have, at times, been
some slight overlap between the phones defendants were selling vis-a-vis those being sold in brick-and-mortar stores. June 27, 2022 Trial Tr. (Day 19), Dkt. 1560, at 3259:23 to 3260:17 & 3370:5 to 3371:9.
• Steve Kroll (a customer of a defendant) testified that he often purchased older-model, inexpensive phones on Amazon or eBay for the purpose of testing apps. June 16, 2022 Trial Tr. (Day 14), Dkt. 1535, at 2155:13 to 2156:12; 2174:16-19, & 2176:9-14.
• Apple expert Leah Caras confirmed that the phones she examined for this case were “older model phones.” June 24, 2022 Trial Tr. (Day 18), Dkt. 1545, at 3044:20-23.
Consistent with this evidence, during its opening statement the government acknowledged the existence of the secondary market, describing it as follows:
You will hear that the defendants were selling in a secondary market for electronic devices that operated as a black market for counterfeits. And unsuspecting customers thought they were buying new and genuine products when, in fact, they were not. And the defendants sold millions of dollars in this market, including $54 million from Amazon alone, and they made a lot of money doing it.May 23, 2022 Trial Tr. (Day 4), Dkt. 1491, at 31:25 to 32:6.
Against this backdrop, the government points to a single customer-Ed Teders-who paid near-retail for an iPhone 6 he purchased from a defendant. See Dkt. 1857, at 4-5; (citing Tr. Ex. 4069 & Restitution Ex. 1). But this is a single example in a case involving thousands upon thousands of sales. And the great bulk of the evidence was that defendants' customers shopped in the secondary market and paid reduced prices for older-model phones that typically could not be purchased from Apple or Samsung at the time.
The government also cites cases such as United States v. Hucks, No. 11-326, 2013 WL 654397 (E.D. Pa. Feb. 20, 2013) in an effort to show that defendants weren't selling in a separate marketplace. See Gvt. Reply, Dkt. 1843, at 2-3. But Hucks isn't particularly helpful, as it involved an entirely different factual scenario. Hucks sold counterfeit Cialis and Viagra pills at flea markets and bars and on the street. At sentencing, the government asked the court to order restitution to the makers of these pills (Eli Lilly and Pfizer). The court declined, explaining that because Hucks was not licensed to purchase drugs, he couldn't have made legitimate purchases from the manufacturers. And Hucks' clients, in turn, didn't have prescriptions, so the drug manufacturers didn't miss out on any sales. In that context, the court noted that “[t]he counterfeit market was entirely separate from Eli Lilly and Pfizer's legitimate market.” Id. at *4.
The government suggests that this is what two separate marketplaces look like and, from there, argues that defendants' customers in this case weren't really shopping in a separate marketplace. But just because the two markets here were not as starkly separate as those in Hucks doesn't change the fact that that there were, indeed, two separate markets. Because the government did not meaningfully engage with the realities of the secondary market in seeking restitution for Apple and Samsung, it failed to distinguish Anderson. As the Court sees it, Anderson forecloses a restitution award to Apple and Samsung.
Finally, the Court is not persuaded by the government's post-hearing effort to sidestep the causation issue by pointing out that Apple and Samsung were extraordinarily conservative in tallying up the number of counterfeit devices defendants sold. See Gvt. Br., Dkt. 1857, at 5-7. The government also says Apple and Samsung didn't include items seized from warehouses in their loss calculations.
Pointing to the items in the warehouses doesn't help. As the government acknowledged in the initial round of briefing, the tally of diverted sales should only include goods that were actually placed into the market. See Gvt. Br., Dkt. 1834, at 3, 5; see generally Fair, 699 F.3d at 514 (“If the record does not demonstrate that the counterfeit goods ever reached the market, . . . courts have held that no actual loss can be shown and restitution therefore is inappropriate.”). More broadly, this facet of the government's argument is essentially an appeal for rough justice. The government is basically saying that even if the assumptions underlying the loss calculations are inaccurate, the Court should nonetheless feel comfortable ordering restitution because Apple and Samsung surely incurred losses and those losses were far greater than the restitution requests. If a rough-justice approach were allowed, the Court might go along. But the Ninth Circuit has stated that although courts do have flexibility, “the MVRA requires more precision. A ‘back-of-the-envelope approach simply will not do.” Anderson, 741 F.3d at 953.
For all these reasons, the Court will decline to order restitution to Apple and Samsung for their asserted lost profits. Because of this ruling, the Court does not need to address defendants' remaining arguments. For the sake of completeness, however, the Court will briefly note that those arguments were not persuasive. Among other things, defendants argued that Apple and Samsung didn't properly account for “avoidable” costs. They also argued that Apple and Samsung didn't “lose out completely” because they would have made money from their “services arms” (such as the App Store and music and payment services). The “avoidable costs” argument is not persuasive because company representatives testified that fixed costs wouldn't have changed. And it's irrelevant if Apple and Samsung were able to eventually make some money off defendants' customers by way of apps and payment services; presumably, they would have made that money regardless. The critical issue here is whether they lost money on the initial sale of a device.
B. UL's Request for Lost Testing & Certification Fees
Although government says defendants' conduct caused UL to suffer reputational injury as well as lost certification and testing fees, it has acknowledged that “calculation of reputational harm is too difficult to quantify for the current ‘reasonable estimate' calculation.” Gvt. Reply, Dkt. 1843, at 11. The Court agrees and will therefore focus only on the claim for lost fees.
The next issue is whether UL suffered losses because of defendants' unlawful conduct. UL does not directly sell goods bearing its mark. Instead, UL performs safety testing and certification for tens of thousands or products -including Apple and Samsung products. See UL Letter, Dkt. 1843-3, at 2. UL says it missed out on $1,707,500 in testing and certification fees because counterfeit manufacturers churned out products the defendants sold, placed UL's mark on those products (or batteries used in the products), and never retained or paid UL. Specifically, UL says defendants “provided an outlet for more than twenty unscrupulous factories to dump counterfeit goods on unsuspecting consumers in the United States.” UL Letter, Dkt. 1834-3, at 7.
Using the existence of these 20-plus factories as a working assumption, UL set about determining how many different products each factory produced. To get that data, UL turned to Trial Exhibit 1300, which is a list of packages that were shipped to various defendants from overseas senders but intercepted and seized by Customs and Border Protection (CBP). See UL Letter, Ex. A thereto, Dkt. 1834-3. If these CBP seizure packages contained items bearing the UL mark, UL assumed that (1) the person listed as the “sender” on any given package was, in fact, a manufacturer-not a distributor or a middleman of some sort, and (2) this manufacturer would have paid UL between $20,000 and $25,000 to review the product inside the package, plus another $7,500 for factory surveillance. See id.
Here is a representative sampling from UL's loss-calculation chart, which employs the approach just described:
Id. at 14, 16, 17.
Sender
“Manufacturer” Model
UL Certification
Estimated Lost Fees
Shenzhen Opqnciq Co Hong Kong
Apple iPhones (20)
Y (Battery)
$32,500
Logic Pro Technology Ltd Honk Kong
Apple iPhones (53)
Y (Battery)
$32,500
Azeuredff Cargo Co Ltd Hong Kong
Apple iPhones (20)
Y (Battery)
$32,500
Alliance Telecom Hong Kon
Apple iPhones (100)
Y (Battery)
$32,500
UL says the following information can be gleaned (or should be assumed) from the four relevant seizures shown in this chart: (1) each entity listed in the “sender” column is a manufacturer; (2) each sender thus has a manufacturing facility where it produced counterfeit Apple iPhones; (3) the batteries in the seized iPhones were marked as being UL certified; and (4) each manufacturer would have paid UL a minimum of $32,500, which represents the cost to certify the batteries and conduct factory surveillance for one year.
There are several problems with these assumptions and with UL's loss theory. First, UL's loss theory is at odds with the diverted-sales theory underlying the government's request for restitution for Apple and Samsung. See generally Linscott Dec., Dkt. 1839-1, at 9. As explained above, Apple and Samsung assumed that every defrauded consumer who purchased a counterfeit item from the defendants would have purchased a corresponding, genuine item from them. But if that were the case, the genuine products would have been tested and certified by UL. As the government has explained, defendants' customers “wanted to buy genuine Apple and Samsung products that had been tested and certified by UL.” Gvt. Br., Dkt. 1834, at 4 (emphasis added). As such, UL wouldn't be entitled to collect additional testing and certification fees. Obviously, Apple and Samsung would have sold more product using that assumption - that's the entire point of the lost-profits-on-diverted-sales theory. And UL has acknowledged that certification fees are flat fees - sales volume doesn't matter. See UL Letter, Dkt. 1834-3, at 9. (“The fees that the defendants would have paid during this process would be independent of the quantity of products sold to consumers.”). Moreover, both Apple and Samsung have said their fixed fees (such as legal, marketing and HR expenses) would not have increased the cost to produce the relatively small number of products at issue in this request. See Gvt. Reply, Dkt. 1843, at 8; June 14, 2023 Hearing Tr., Dkt. 1852, at 44:11-18 (testifying that “over hundreds of millions” of Apple iPhones were sold between 2012 and 2018); Id. at 61:1-12 & 69:7-11 (“Samsung sells tens of millions of products every year.”); Id. at 32:21 to 33:18 (explaining that because “we're only talking about less than 30,000 units” in an established market, fixed costs such legal, marketing, and HR would not increase). Along those same lines, it stands to reason that once Apple or Samsung had paid the relevant testing fees, UL would not be in a position to collect additional fees if defendants had not diverted sales from Apple and Samsung. For this reason, the Court finds that the government failed to prove, by a preponderance of the evidence, that UL suffered any lost testing and certification fees because of defendants' unlawful conduct.
Alternatively, engaging with UL's loss theory reveals at least two significant flaws. First, at times UL appeared to assume that the defendants themselves were manufacturers. See UL Letter, Dkt. 1834-3, at 9 (“UL Solutions has been directly and proximately harmed by the defendants' counterfeiting of UL's Marks.”) (emphasis added). But the evidence at trial didn't show that defendants were manufacturers. Nor is the Court persuaded by the government's arguments that the defendants were “more than simply downstream consumers of these manufacturers' goods.” Gvt. Reply, Dkt. 1843, at 11. In its reply brief, the government supported this assertion by stating that Piotr, Pavel, and Timofey Babichenko “instructed manufacturers on how to create more passable counterfeits.” Id. But the trial evidence doesn't support that argument. Rather, the key takeaway from the trial evidence was that the defendants were purchasers of counterfeit goods - not manufacturers.
The government cited four exhibits to support its argument that defendants “intimately coordinated” the manufacturing of their counterfeit goods - Exhibits 3005-03, 3005-39, 3005-40, and 3006-04.” Reply, Dkt. 1843, at 11. Two of the cited exhibits (3005-39 and 3005-40) were not admitted at either trial. (The Court also checked pages 39 and 40 of Exhibit 3005, which was admitted at the first trial, but those two pages do not show that defendants “intimately coordinated” the manufacture of counterfeit goods.) And the Court is not persuaded that the other two cited trial exhibits (3005-03 and 3006-04) show that the defendants were intimately involved in the manufacturing process.
A second problem with UL's approach is that it assumes every “sender” listed on the CBP seizure chart is a manufacturer, and, therefore, a potential entity that would hire UL and pay certification and testing fees. See Ex. A to UL Letter, Dkt. 1843-3, at 13 of 38, n.1. But there is no proof that this is so. As defendants have argued, the evidence suggested that these senders were suppliers. Further, a closer look at Trial Exhibit 1300 reveals the dangers with assuming that the “sender” listed on any given package was a manufacturer-or even a supplier, for that matter. In some instances, the “sender” of the package is merely a play on the recipient's name. For example, one of the packages seized at the border was addressed to Mobi Recycle LLC - which is one of defendant's businesses. The “sender” for that package is listed as “Mobi Recycle LLC, Hong Kong.” See Tr. Ex. 1300, at 3, ln. 15; see also UL Letter, Dkt. 1834-3, at 16-17. Similarly, several packages seized by CBP were addressed to Vadim Dmitruk, an unindicted coconspirator. The “senders” for some of these packages are variations of Mr. Dmitruk's name, such as “Vadim Electronic Technology Co,” and “Dmitruk Electronic Technology.” See Tr. Ex. 1300, at 9:6-8. If UL's logic were applied, Mr. Dmitruk had two or three Hong Kong-based manufacturing companies and Mobi Recycle had one. There was no such evidence at trial.
For all these reasons, the government failed to demonstrate that UL suffered lost testing and certification fees by virtue of defendants' conduct. Accordingly, the Court will deny this aspect of the restitution request.
2. Requests for Reimbursement of Expenses Incurred During the Investigation and Prosecution of the Case
The government next argues that Samsung and UL should be reimbursed for expenses associated with the government's investigation and prosecution of this case. Samsung says it incurred $281,332.37 in such expenses, and UL says it incurred $154,188.91. For the reasons explained below, the Court will hold off on ordering restitution of these types of expenses-as well as determining how to apportion such expenses amongst the defendants-until after it has received more detailed information regarding the expenses.
In pursuing these expenses, the government relies upon 18 U.S.C. § 3663A(b)(4), which states that defendants convicted of certain crimes must be ordered to “reimburse the victim for lost income and necessary childcare, transportation, and other expenses incurred during participation in the investigation or prosecution of the offense or attendance at proceedings related to the offense.” 18 U.S.C. § 3663A(b)(4) (emphasis added). Focusing on the italicized, “other expenses” clause, the Ninth Circuit at one time had adopted a broad view of that clause - holding that victims could recover investigation costs incurred by private parties so long as those costs were a direct and foreseeable result of defendants' conduct. See United States v. Gordon, 393 F.3d 1044, 1057 (9th Cir. 2004). But in Lagos v. United States, 138 S.Ct. 1684, 1687 (2018), the Supreme Court expressly rejected that Ninth Circuit precedent (and other, similar circuit holdings), choosing instead to adopt a “more limited interpretation” of the statute. Lagos held that “the words ‘investigation' and ‘proceedings' in 18 U.S.C. § 3663A(b)(4) are limited to government investigations and criminal proceedings.'” Id. (emphasis added). Thus, corporate victim General Electric could not recover the millions of dollars it had spent investigating Lagos's crime and participating in his bankruptcy proceedings. Id.
In light of Lagos, the Court must ensure that Samsung's and UL's claimed expenses were incurred as part of their participation in a government investigation or as part of their attendance at a proceeding in this case. The Court will address each company's claimed expenses in turn below. But at the threshold, the Court notes that it is not persuaded by defendants' assertion that these companies cannot claim expenses under 18 U.S.C. § 3663A(b)(4) because they did not prove defendants' conduct caused them to lose profits, or, in UL's case, testing and certification fees. Defendants argue that because of this failure, the companies do not qualify as “victims,” and, therefore, the Court cannot order reimbursement of “other expenses” under 18 U.S.C. § 3663A(b)(4). Defendants have not cited any authority for this proposition, aside from the statute itself. The statute, however, does not condition an award of “other expenses” on a finding that the claimant was victimized in some other way as well. Rather, it defines a victim as one who was “directly and proximately harmed as a result of the commission of an offense for which restitution may be ordered ....” 18 U.S.C. § 3663A(a)(2). And just because Samsung and UL were unable to demonstrate that they lost profits or testing fees does not mean that defendants' conduct did not directly and proximately cause other harms - such as costs of participating in a government investigation or attending proceedings. Accordingly, the Court not only may, but must, order that Samsung and UL be reimbursed for such expenses, assuming, of course, that the companies put forth sufficient evidence.
A. UL's Claimed “Other Expenses”
UL says it incurred two categories of “other expenses:” attorney time and monies paid to salaried employees while those employees participated in the investigation or attended court proceedings.
Starting with the salaried employees, UL's in-house counsel and three other UL employees produced after-the-fact estimates of time spent “performing work relating to the case.” Sharma Dec., Dkt. 1834-3, at 36, ¶ 2. Defendants quibble with the fact that these employees estimated their time after the fact. But while contemporaneous logs would have been preferable, three of the four employees (Robert Pollock, Kevin Olive, and Randy Johnson) have nevertheless provided detailed descriptions of the work performed, and the Court finds that those descriptions align with the hours it would take to do the work. (The fourth employee is an attorney, so the Court will address her time below, when it discusses UL's request for attorney's fees.)
UL employee Robert Pollock estimates that he spent 250 hours preparing for and testifying at a Daubert hearing and at both trials, overseeing testing of some of defendants' devices prior to the second trial, and participating in additional hearings. He also reportedly “supported the seizure” and “field[ed] questions relating to the counterfeit products.” Id. ¶ 4.
Next up is UL employee Kevin Olive, who “attended the seizure in this case at the request of the government and assisted in verifying that the UL certification marks on the seized product were counterfeit.” Id. ¶ 3. He also assisted Mr. Pollock with testing, which involved traveling to Boise, inspecting and setting up equipment, and recording data relating to that testing. Id. Mr. Olive estimates that he spent 50 hours on these tasks.
The third UL employee is Randy Johnson. Mr. Johnson reports that he assisted Mr. Pollock in testing some of defendants' products. This included travel to Boise, inspection and setup, and recording data relating to that testing. He estimates that he spent 40 hours on these tasks. Id. ¶ 5.
Based upon these descriptions, as well as the Court's familiarity with this complex litigation, including UL's participation in the trial, the Court finds that the work just described was “participating” in a government investigation or “attending” a proceeding. See generally Lagos, 138 S.Ct. at 1688.
Defendants argue that UL shouldn't be reimbursed for these salaried employees' hours because they were able to keep up with their regular work at UL, so UL failed to prove any real losses. After all, UL had to pay these employees their regular salaries regardless, and they were apparently able to absorb the extra work, so the company didn't really lose any money. Or so the argument goes. See Post-Hearing Br., Dkt. 1860, at 3-4. The Court doesn't agree. Because of defendants' conduct, these employees spent significant chunks of time focused on the government's investigation or attending proceedings - rather than doing their regular work. In that sense - they “missed” work, which is precisely the sort of loss 18 U.S.C. § 3663A(b)(4) aims to reimburse. See Lagos, 138 S.Ct. at 1688.
But defendants are correct when they say UL failed to prove the amount of its losses for these hours. UL isn't just asking to be reimbursed for the monies it actually paid these employees for these hours. Rather, for two of these employees (Pollock and Olive), UL's in-house counsel says she “assigned” billing rates. And for Mr. Johnson, UL doesn't clearly say it paid Mr. Johnson an hourly wage as a UL employee. Rather, UL asks the Court to reimburse it for Mr. Johnson's “ordinary billing rate for field work.” See Sharma Dec. ¶ 5, Dkt. 1834-3. This approach is problematic because reimbursement should be at the rates UL actually paid these employees. That is the extent of UL's losses. Neither UL nor the government has explained why UL should be paid some different (presumably higher) rate. Accordingly, the Court will hold off on ordering restitution to UL. Instead, UL may submit a supplemental declaration which contains a more detailed calculation of the amounts these employees were actually paid.
As for attorney time, UL seeks compensation for time spent by in-house attorney, Sanjana Sharma, as well as time spent by employees of the Greenberg Traurig law firm. As a threshold matter, attorneys' fees remain recoverable as an “other expense” after Lagos - so long as the time is reasonable and was spent either participating in or attending a government investigation or proceeding. See United States v. Afriyie, 27 F.4th 161, 163 (2d Cir. 2022) (concluding that preLagos Second Circuit authority holding that attorneys' fees can sometimes be “other expenses” referred to in 18 U.S.C. § 3663A(b)(4) remains good law after Lagos). The problem, though, is that the Court doesn't have any billing records or detailed descriptions before it. So it's not possible to ascertain whether the attorney time spent was reasonable or whether all of the time was spent in connection with the government investigation or the proceedings in this case. If UL wishes to pursue these fees, it will need to submit billing records for outside counsel. For inhouse counsel, the Court would likewise need additional detail, as well as the hourly rate UL pays Ms. Sharma, in order to determine whether the entirety of her hours are reimbursable, and, if so, at what rate.
B. Samsung's Claimed “Other Expenses”
The Court is also not in a position to determine whether all time spent by Samsung's outside counsel is reimbursable. Samsung's initial request for $281,332.37 in attorney's fees and costs was based on nothing more than a chart setting forth time spent by 13 different timekeepers for a 10-year period. See Apr. 10, 2023 Letter, Ex. A thereto, Dkt. 1834-2, at 3. There was no description whatsoever of the work performed. On reply, Samsung followed up with a slightly more detailed chart, which contains helpful summary descriptions. See Rinkoff Dec., Ex. A thereto, Dkt. 1843-4, at 3. But even with this chart, the Court is not in a position to order restitution to Samsung. One obvious problem is that Samsung seeks fees stretching back to 2014. The indictment was filed in August 2018. The government obviously began investigating before that date, but the Court would need to have more information - both as to when the investigation began and how Samsung became involved in that investigation - before awarding fees.
On top of that, the Court would need to review billing records, not just a summary chart (although that is certainly helpful), to determine whether the time spent was both reasonable and incurred in connection with Samsung's attendance at, or participation in, a government investigation or proceeding. Accordingly, the Court will deny this request without prejudice. The Court will allow Samsung and UL to submit additional materials to support the requests for attorneys' fees, and any such submissions should include billing records and attorney timesheets.
The government says any review of billing records should be in camera. See Gvt. Reply, Dkt. 1843, at 14 n.2 (citing 18 U.S.C. § 3663A(d)(4) and Eyraud, 809 F.3d at 470). But courts routinely review attorney's bills in the context of determining whether the fees are reasonable. As one court has noted, “[t]o the extent a district court must review a law firm's timesheets during restitution proceedings, this is a requirement mandated by the MVRA itself-no different from the court's obligation to review a victim's parking and child care receipts.” Afriyie, 27 F.4th at 170. Further, though district courts may authorize in camera submissions, see Eyraud, 809 F.3d at 470, they don't necessarily have to do that; courts often receive attorneys' bills by way of regular filings. See, e.g., United States v. Bahel, 662 F.3d 610, 648 (2d Cir. 2011) (billing records supplied shortly before restitution hearing); United States v. Edwards, 19 F.Supp.3d 366, 372-73 (D. Mass. 2014) (reviewing an attorney billing list submitted by the victim, which included individualized entries); United States v. Gupta, 925 F.Supp.2d 581, 584 (S.D.N.Y. 2013) (noting that victim had submitted 542 pages of attorney billing records). Still, though, to protect Samsung's and UL's privacy, the Court will allow them to file attorney's billing records under seal.
3. Defendant Artur Pupko's Motion to Dismiss
Finally, the Court will address whether Defendant Artur Pupko should be ordered to reimburse victims in this matter. Mr. Pupko argues that the Court cannot order restitution against him because the government has not asked for it. But the governing statute, 18 U.S.C. § 2323(c), directs courts to order restitution without reference to whether the government has so requested. And the Ninth Circuit has held that a district court may order restitution so long as “someone”-not necessarily the government-“proves the amount by a preponderance of the evidence.” United States v. Gamma Tech Indus., Inc., 265 F.3d 917, 924 (9th Cir. 2001); see also United States v. Eyraud, 809 F.3d 462 (9th Cir. 2015).
Nevertheless, because Mr. Pupko cooperated early, and because the great bulk of UL and Samsung's expenses were likely incurred in connection with the trials, the vast majority of any such expenses will not be attributable to Mr. Pupko. But to the extent Samsung and UL incurred qualifying expenses prior to the date Mr. Pupko pleaded guilty-and they likely did-the Court likely will order Mr. Pupko to pay a very small portion of any such expenses. Before making this determination, however, the Court needs to see supplemental materials from Samsung and UL. For that reason, the Court will not specify the amount of a restitution award for any defendant at this time. That will be done after the Court reviews the supplemental materials.
ORDER
IT IS ORDERED that:
1. The Government's Request for Restitution as it relates to Apple, Samsung, and UL, is DENIED in part, and DENIED WITOUT PREJUDICE, in part, as follows: a. The Court DENIES the government's request to order restitution to Apple in the amount of $4,799,880 for alleged lost profits. b. The Court DENIES the government's request to order restitution to Samsung in the amount of $8,944,275.24 for alleged lost profits. c. The Court DENIES the government's request to order restitution to UL in the amount of $1,707,500 for alleged lost testing and certification fees. d. The Court DENIES WITHOUT PREJUDICE the government's request that Samsung and UL be reimbursed for claimed “other expenses” under the MVRA, in the amounts of $281,332.37 and $154,188.91, respectively. If these entities wish to pursue such expenses, the government shall submit supplemental materials (including attorneys' billing records for both companies, and, for UL, hourly rates paid to the relevant salaried employees). These materials may be filed under seal, and must be filed within 21 days of this 34 Order. The government may accompany these materials with a brief, not to exceed 15 pages, at its option. Defendants shall have 14 days within which to respond to these materials. Defendants' briefs, likewise, shall not exceed 15 pages. The government shall then have 14 days in which to file an optional reply brief, which shall not exceed 10 pages. 2. Artur Pupko's Motion to Dismiss (Dkt. 1841) is DENIED. The Court will determine the amount of restitution after it has had a chance to review supplemental materials, if any, that are submitted.