UNITED STATES of America, Plaintiff-Appellee, v. Dainius VYSNIAUSKAS, Defendant-Appellant, 593 F. App'x 518

BOGGS, Circuit Judge. Dainius Vysniauskas (true name Valen-tinas Babakinas) is a Lithuanian national who pleaded guilty to bank fraud. He now appeals his top-of-guidelines-range sentence of 71 months. Specifically, he argues that the district court incorrectly calculated the total loss from his fraud, applied sentencing enhancements not supported by the facts, and imposed a procedurally and substantively unreasonable sentence. We affirm. I A Vysniauskas entered the United States illegally in 1999. In 2002, he was caught attempting to shoplift at a grocery store under the false name Denius Norkus.

Immigration and Customs Enforcement subsequently ordered his removal in October 2002, but Vysniauskas eluded immigration authorities until he was deported in 2006. He returned illegally to the United States at an unknown time.' After his return to the United States, Vysniauskas became involved with Tatsia-na Leichanka, a co-conspirator in the bank-fraud scheme and the mother of his now 4-year-old child. Starting in 2008, the two engaged in an extensive scheme of bank and retail fraud.

Using false names backed by various forms of fake and stolen identification, the defendants set up dozens of accounts at several banks over the course of two years. The co-conspirators used these false accounts to defraud the banks and others in several different ways. First, they wrote checks from one of their accounts that did not have sufficient funds to another one of their accounts. Vys-niauskas and Leichanka exploited the gap between the time a depository (payee) bank credits funds and the time a drawee (payor) bank returns a not-sufficient-funds check.

During this gap, typically two or three days, they would make large cash withdrawals or other purchases. Once the check was dishonored, the account would go into a negative balance and they would abandon it. The government calculated that Vysniauskas and Leichanka wrote $44,250 in such returned checks, with actual damages in the range of $80,000. Second, it appears that Vysniauskas and Lei-chanka would also overdraw their accounts in a more straightforward manner, taking advantage of the overdraft protection offered by the banks.

Often in tandem with the deposit of a bad check, defendants would withdraw funds that exceeded the balance of available funds and, as before, abandon the account without repaying. The government does not give a precise breakdown of overdraft losses, as these were a relatively small component of the overall fraud. Third, Vysniauskas and Leichanka fraudulently disputed over $8,000 in charges and ATM withdrawals that they in fact made themselves. Fourth, Vysniauskas used not-sufficient-funds checks to make over $5,000 in purchases from Costco.

Vysniauskas disputes that the fourth category of loss is relevant, as the bad checks were written and cashed before the more extensive bad-check-depositing scheme began. Nevertheless, as with the later fraud, the checks were drawn on an account made in a false name (Eduardus Zenkevicius) that was later used for other fraudulent accounts. Finally, Vysniauskas also made unauthorized purchases totaling $186.85 with the credit card of Anthony Demaree.

Vysniauskas’s scheme was exposed in 2010, when Leichanka attempted to open a Chase bank account in Sterling Heights, Michigan, using the same false name that she had unsuccessfully tried to use the day before at a different branch. Leichanka fled to the parking lot, where she and Vysniauskas were arrested by the police. Upon arrest, both Leichanka and Vys-niauskas gave false names to the police, who discovered several false identification documents, bank cards, and a large amount of cash in their car.

At the station, Leichanka ultimately confessed to the fraud; Vysniauskas continued to deny involvement and gave yet another false name. Unknown to the police, Vysniaus-kas and Leichanka had left their then 16-month-old son alone in their apartment. Not wanting to involve the police — due to the incriminating evidence in the apartment — Vysniauskas made a call to a friend, whom he asked to break into the apartment to get the child and also to “take and keep” anything that would “get him in more trouble.” The friend was unable to gain entry, and eventually called to police to help.

The police broke in at 2:35 a.m., finding the child sleeping on the floor after being left unattended for approximately 35 hours. While inside, the police observed boxes of checks and a crib full of mail addressed to different people.

As a result, they obtained a search warrant for the apartment, finding vast amounts of stolen mail, mail sent to Vys-niauskas’s various aliases, false identification documents, hundreds of -checkbooks, and 103 credit cards and other “financial transaction devices.” Vysniauskas pleaded guilty to identity theft in Macomb County, Michigan, Circuit Court, receiving a sentence of 78 days in custody on July 7, 2010. He escaped shortly thereafter, but was re-arrested trying to visit his son, again using false identification. He was sentenced to a year in jail for the escape.

B In the meantime, the FBI had initiated an investigation into Vysniauskas’s bank fraud. Agent Neil Gavin compiled bank records from the affected financial institutions into a spreadsheet, confirming over $200,000 in deposits and withdrawals from accounts associated with the various aliases of Vysniauskas and Leichanka. On November 30, 2010, Vysniauskas was charged with bank fraud, in violation of 18 U.S.C. § 1344; and attempted bank fraud and conspiracy to commit bank fraud, in violation of 18 U.S.C. § 1349.

He pleaded guilty without the benefit of a plea agreement, and, after an evidentiary hearing on disputed sentencing issues, the district court sentenced him to 71 months, the top of the calculated guidelines range. In reaching this sentence, the district court applied, and Vysniauskas objected to, several sentencing enhancements. First, the district court found that the amount of loss caused or intended by the fraud was $138,466.27, warranting a sentence enhancement of ten levels for theft losses in the range of $120,000 to $200,000. USSG § 2Bl.l(b)(l)(F).

This loss total was comprised of six components: 1) $51,500 in presumed loss attributable to the 103 fraudulent access devices (applying the $500 per device minimum provided by the Guidelines); 2) $500 in presumed loss from the unauthorized use of Anthony De-maree’s credit card (actual loss was only '$186.35); 3) $28,042.21 in actual loss, as reported by the banks; 4) $44,250 in intended loss, from returned checks; 5) $8,517 in intended loss from the disputed-transaction claims; and 6) $5,657.06 in intended loss from not-sufficient-funds checks written to Costco.

The district court’s loss total was significantly less than the probation office’s calculation, since the probation office included the total amount of all withdrawals from the fraudulent accounts. The district court also excluded actual losses that were not proven by a loss submission from one of the banks. Second, the district court applied a two-level enhancement for an offense involving ten or more victims. USSG §. 2Bl.l(b)(2)(A)(I).

To reach ten victims, the court added the three banks with proven losses, two individuals whose names were used to set up accounts (Anthony Demaree and Maria Babanina), and five individuals whose mail, sent by federal agencies, was stolen. Although the Government submitted proofs that many more victims were involved, the court stopped after finding the ten necessary to support the enhancement. Third, the district court applied a two-level enhancement for an offense involving “sophisticated means.” USSG § 2Bl.l(b)(9)(C) (2010).

While finding the individual steps in Vysniauskas’s scheme not particularly sophisticated, when viewing the scheme in its totality, the court found that the sheer number of the fraudulent transactions, dozens of bank accounts, false identification documents, and his illegal alien status rendered the means of the offense “sophisticated.” Fourth, the district court found Vys-niauskas’s call to his friend supported a two-level enhancement for obstruction of justice. USSG § 3C1.1.

The court reasoned that providing false names and a false address was not enough to “significantly impede” the investigation, and so Leichanka did not receive an enhancement for her false statements. However, Vys-niauskas’s call was an active attempt to conceal and destroy material evidence, warranting an enhancement for him. Vysniauskas timely appealed these sentencing enhancements, and now argues on appeal that the sentence was also procedurally and substantively unreasonable.

II A district court determines the amount of loss under the sentencing guidelines by a preponderance of the evidence, and we review that finding for clear error. United States v. McCarty, 628 F.3d 284, 290 (6th Cir.2010). In its calculation, the district court “need only make a reasonable estimate of the loss.” Ibid, (quoting USSG § 2B1.1 cmt. n. 3(C)). Nevertheless, we review the district court’s “methodology for calculating loss” de novo. Ibid.

Under the Guidelines, loss is calculated as “the greater of the actual loss or the intended loss.” USSG § 2B1.1 cmt. n. 3(A). “Intended loss” includes intended harm “that would have been impossible or unlikely to occur.” USSG § 2B1.1 cmt. n. 3(A)(ii). A Before proceeding to the individual loss-calculation arguments, we discuss Vys-niauskas’s fraudulent scheme in greater detail to help clarify our analysis. The district court described the conduct as “check kiting,” and cases with similar facts have used the same term. See United States v. Geevers, 226 F.3d 186 (3d Cir.2000).

The mere labeling of the scheme as “check kiting,” however, is not particularly useful for determining the proper loss-calculation methodology-as one commentator has noted, check kiting “is a loose concept meaning different things for different purposes, used more often to describe egregious behavior than to define it.” Alvin C. Harrell, Some Surprising New (and Old) Perspectives on Check-Kiting, 57 Consumer Fin. L.Q. Rep. 214, 214 (2003). In a traditional check-kiting scheme, the kiter draws a check on one bank account — knowing the check is not backed by sufficient funds — and deposits it in another account.

Taking advantage of the gap, or “float,” between the time checks are honored and cleared, the kiter then draws a second check on the second account and deposits it in the first account, to cover the first check. As long as the kiter continues writing new checks within the temporary clearance period, the balances of his accounts are permanently inflated, resulting in an interest-free loan in the amount of the kited check. See, e.g., Williams v. United States, 458 U.S. 279, 281 n. 1, 102 S.Ct. 3088, 73 L.Ed.2d 767 (1982); United States v. Stone, 954 F.2d 1187, 1188 n. 1 (6th Cir.1992).

Although the scheme can work with two accounts, to avoid detection kiters will often use multiple accounts through which to circulate bad checks. This “circular” version of check kiting is distinct from the scheme engaged in by Vysniauskas. Vysniauskas had no intent of obtaining short-term loans from the banks, but instead, he intended to systematically defraud them by extracting the value of bad checks in cash. Such a scheme requires numerous bank accounts and false names, since each bank account must typically bé abandoned after cashing only one bad check, when the balance becomes irretrievably negative.

Vysniauskas passed relatively small bad checks (about $1,500 on average), so everyday ATM withdrawals and purchases were sufficient to extract most of the face value of the deposited check. In similar schemes involving much larger checks, however, the kiter may reasonably expect to extract only a small percentage of the value of the bad check. See, e.g., United States v. Higgins, 270 F.3d 1070, 1075 (7th Cir.2001) ($69,990 out of a $420,000 check); Geevers, 226 F.3d at 189 (out of $2,000,000 in bad checks, $400,000 attempted and $160,000 successfully withdrawn).

Although there was significant circular activity in Vys-niauskas’s accounts, it appears the purpose was to mask his overdrafts and large withdrawals, not to obtain a short-term loan. Given the variety of potential kiting schemes, we must be careful to employ a loss methodology appropriate to the case at hand: loss calculations in prior check-kiting cases might not be relevant precedent. With this in mind, we now address the specific issues with the district court’s loss calculation.

B The first question is whether, in a non-circular check-kiting scheme, the entire face value of dishonored checks may be presumed to be the intended loss. Vys-niauskas argues that the actual withdrawals taken from the bank accounts are a “clear manifestation of [his] intent,” since during the “temporary credit” period after the deposit of the kited check, the amounts withdrawn were within his control. Appellant’s Br. at 26. If he had intended to cause more loss, Vysniauskas contends, he would have. The district court properly rejected this argument.

An intended loss need not be likely or even possible, USSG § 2B1.1 cmt. n. 3(A)(ii), and as Geevers explained, “expectation is not synonymous with intent when a criminal does not know what he may expect to obtain, but intends to take what he can.” Geevers, 226 F.3d at 193. The same principle is applicable here. Vysniauskas may not have expected to withdraw the full face value of the bad checks, but if a bank had been slow in dishonoring the check, there was nothing to stop Vysniauskas from continuing to make withdrawals.

Vysniauskas’s argument is especially weak here, where the actual losses caused by Vysniauskas were more than 50% of the face value of the bad checks, and in some cases Vysniauskas successfully withdrew more than the face value of the check. A defendant is free to introduce evidence to overcome the presumption that intended loss is less than the face value of bad checks, see, e.g., Higgins, 270 F.3d at 1075, but Vysniaus-kas has made no such showing here. This methodological holding should not be extended to “circular” check-kiting schemes.

The district court was correct in not following the methodology used in United States v. Swanson, 360 F.3d 1155 (10th Cir.2004), and United States v. Deutsch, 987 F.2d 878 (2d Cir.1993), both, of which used the amount of overdrafts instead of the face value of bad checks to calculate loss. This is not because those cases had unsound reasoning, but because those cases dealt with fundamentally different types of check kiting — there is no “circuit split,” as suggested by Vys-niauskas. Appellant’s Br. at 34 & n. 4.

Swanson involved a classic “circular” check-kiting scheme, in which a Kansas businessman used overdrafts and non-sufficient-funds checks circulated among his eleven bank accounts to fund his day-trading activities. Swanson, 360 F.3d at 1158-59. The bank ultimately detected the unusual activity and immediately charged all outstanding checks, resulting in an overdraft of over $500,000. Id. at 1160. Over the course of his scheme, Swanson had circulated over $70,000,000 in kited checks. Ibid.

No party suggested that $70,000,000 was the proper figure for intended loss, and $500,000 — the amount of the overdraft — was used for sentencing. Id. at 1167. Deutsch involved a more unusual variant of check kiting, which might be termed “telescoping” rather than “circular.” Deutsch obtained a credit card from AT & T with a limit of $8,000, using false information. Deutsch, 987 F.2d at 881. AT & T had a policy of giving immediate credit upon the receipt of payments. Ibid. Taking advantage of this, whenever his credit limit was exhausted, Deutsch would pay off his balance with a check drawn on a closed account.

Ibid. When the check was dishonored — as expected — Deutsch would simply write a new, larger check to cover it and any further expenses that had been incurred. Ibid. This was repeated nine times over three months, allowing Deutsch to spend $34,000 before being caught. Ibid. The district court added up the face value of all the checks to get at an intended loss amount of $114,000. Id. at 886. The Second Circuit disagreed, reasoning that “[e]ach check, in effect, replaced all the previous checks and was not cumulative.” Ibid. As a result, only the face value of the last check was relevant to the intended loss.

Taken together, Deutsch, Swanson, and Geevers stand for a simple proposition: in a fraudulent scheme, potential losses— however unlikely — and impossible losses may be included in the loss calculation, but double counting is not acceptable. See also United States v. Watkins, 994 F.2d 1192, 1196 n. 5 (6th Cir.1993) (noting that “the record contains indications that the sentencing judge might have counted some funds twice by adding together several transactions involving the same amount of bogus funds.”). Factually impossible and legally impossible frauds involve actual intent to cause loss.

See, e.g., United States v. Younes, 194 Fed.Appx. 302, 315-16 (6th Cir.2006) (all fraudulent attempts to obtain student aid included in loss total, although some students were eligible in fact); United States v. Schlei, 122 F.3d 944, 996 (11th Cir.1997) (government sting). Bad checks passed in a circular fashion, on the other hand, may not always be individually intended to cause loss. A methodology involving double counting must be avoided, not because such losses are impossible, but because they are not intended.

C Vysniauskas argues that the district court improperly added actual losses to intended losses despite the fact that the two kinds of loss overlap. In particular, he argues that the intended-loss calculation of $44,250 based on the face value of the returned checks overlaps with the actual-loss estimate of $28,042.21 as reported by the banks. If a fraud involves both successful and unsuccessful attempts, total loss may include both actual and intended losses. United States v. Brown, 151 F.3d 476, 489-90 (6th Cir.1998). But as intended loss “necessarily includes all actual losses,” United States v.

Tate, 136 Fed.Appx. 821, 826 (6th Cir.2005), courts must be careful not to double count. The government does not dispute the legal proposition that actual losses should not be double counted in the intended-loss estimate, but argues that “[t]he evidence submitted to the court properly separated actual from intended loss such that the district court could determine both in its finding of total loss.” Appellee’s Br. at 27. Vysniauskas’s argument is perhaps best illustrated with an example.

Consider the following scenario: Vysniauskas writes a $500 check on his account with Bank A and deposits it in his Bank B account. Assume that both accounts initially have zero balances. Bank B then gives Vysniauskas temporary credit for the face value of the check, and Vysniauskas withdraws the $500 from Bank B. Now, when Bank B presents the check to Bank A, either Bank A honors it, in which case Bank A is out the $500, or Bank A returns it to Bank B, in which case Bank B is out the $500. Either way, the actual loss is $500, and is equal to the intended loss of $500, as reflected in the face value of the check.

Vysniauskas’s argument is that, where the check was returned, adding its value to the amount of the actual loss sustained by Bank B effectively double-counts that loss. But Vysniauskas fails to account for the fact that the victims of his actual and intended losses in the above scenario are different — in other words, the losses are not, in fact, identical. Consider the above hypothetical from the perspective of Bank A only. If Bank A is presented with the check and honors it, it incurs an actual loss of $500.

If Bank A dishonors the check and returns it, it incurs no loss, but there is nonetheless an intended-but-unrealized loss with respect to Bank A of $500. So, with respect to Bank A, the correct way to compute the amount of loss would be to add the actual losses (from checks honored or amounts otherwise withdrawn despite insufficient funds) to the intended-but-unrealized losses (from returned checks). That appears to be exactly what the district court did. The court appears to have calculated the amount of loss by adding up the actual losses and the intended-but-unrealized losses for each victim bank.

As Agent Gavin testified, only checks that were actually returned were used to support the intended-loss figure. This is supported by his spreadsheet, and is methodologically sound: honored checks are like those in a “circular” check-kiting scheme, and should not be included in intended loss. In his testimony, Agent Gavin also agreed that “[i]f the check was not honored by a bank ... then it would not be included in that affidavit of actual loss” for that bank.

This means that the actual losses tallied in the bank affidavits are not included in the intended-loss calculation from the face values of the returned checks. Although Vysniauskas’s double-counting argument may be appealing at first blush, it falls short for several reasons. First, although Vysniauskas’s gain in the above example might only be $500 — no matter which bank it ultimately came from — he exposed each bank to the risk of loss.

Even if only one bank would ultimately incur the loss, “[i]ntended loss ... includes intended pecuniary harm that would have been impossible or unlikely to occur.” See USSG § 2B1.1 cmt. n. 3(A)(ii). Corroborating this point is the fact that, as explained in

Part II-E of this opinion, one may be found guilty of bank fraud under 18 U.S.C. § 1344 for writing not-suffieient-funds checks to retailers, since the bank is exposed to an actual risk of loss, regardless of whether any loss is ultimately incurred. United States v. Reaume, 338 F.3d 577, 581 (6th Cir.2003) (“[W]hen the Bank receives an NSF check, it makes a decision to either honor the check anyway or to dishonor the check. If the check is dishonored, the Bank does not lose any money, but if the check is honored, and the account holder fails to pay back that debt to the Bank, the Bank suffers a loss.”).

In the example described above, holding Vys-niauskas responsible for only one bank’s actual or intended loss would be inconsistent with the undisputed fact that both banks are “victims” under law. In addition, the district court’s approach of calculating the amount of loss separately for each victim finds significant support in the Guidelines.

Application Note 3(C)(iv) explains that a reasonable loss estimate may be based on “[t]he approximate number of victims multiplied by the average loss to each victim,” where, again, “loss” is defined as “the greater of actual loss or intended loss,” USSG § 2B1.1 cmt. n. 3(A). A fair reading of Application Note 3(C)(iv), then, is that loss estimates may be based on “[t]he approximate number of victims multiplied by the average [of the greater of actual loss or intended loss] to each victim.” The Guidelines commentary concerning losses in “Ponzi and Other Fraudulent Investment Schemes” is also instructive.

That commentary explains that “the gain to an individual investor in the scheme shall not be used to offset the loss to another individual investor in the scheme.” Id. at cmt. n. 3(F)(iv). The point is: victims are not fungible. Finally, it is no answer to say that, in the scenario in which Bank A does not honor the check, Vysniauskas should somehow receive a “credit” for the intended-but-unrealized loss to Bank A based on the actual loss to Bank B.

Under the Guidelines, criminals may receive “Credits Against Loss” for the amount of “money returned ... by the defendant or other persons acting jointly with the defendant, to the victim before the offense was detected.” Id. at cmt. n. 3(E)(i) (emphasis added). Vysniauskas should not receive credit for the fact that Bank A decided not to honor his checks when presented with them. After all, Bank A’s losses in that scenario were reduced not by Vysniaus-kas’s actions but by the bank’s own actions.

The Seventh Circuit appears to have considered this same question — i.e., whether the amount of loss may include both the intended loss to one victim and the actual loss to another, -with respect to the same sum. See United States v. Lauer, 148 F.3d 766, 767 (7th Cir.1998). Lauer involved a Ponzi scheme with numerous victims. The court held as follows: If the defendant intended to steal $5 million and succeeded in stealing $1 million, it would be absurd to reckon the loss at $6 million, for this would result in his being punished more severely than if his theft had been a complete success.

But here we have multiple victims, and we cannot think of any objection to adding the actual loss of one to the intended loss of another, any more than it would be objectionable to punish the defendant separately for murdering X and attempting to murder Y, as distinct from punishing him separately for murdering and attempting to murder X. Ibid, (emphasis added). To be sure, one may distinguish the Lauer court’s murder example on the ground that, in that case, both deaths are intended, whereas here, only one loss is ultimately intended. Nevertheless, that distinction does not affect the outcome.

It is helpful to consider Vysniauskas’s actions in the example given earlier sequentially. First he deposits a non-sufficient-funds check into Bank B and redeems the cash, causing a loss to Bank B. That loss may eventually be passed on, but at this point, Bank B has incurred the loss, and Vysniauskas has acted with the requisite intent with respect to that loss. A few days later, Bank B presents the check to Bank A. Depending on how it responds, Bank A will suffer an actual loss or an intended-but-unrealized loss, and Vysniauskas will have acted with the requisite intent with respect to that loss as well.

To summarize, Vysniauskas imposed the loss on Bank B, and then attempted to impose that same loss on Bank A. Like the Lauer court, we see no reason not to hold Vysniauskas accountable for the amount of loss that he both actually imposed or intended to impose on each victim. That approach is consistent with the text of the Guidelines and their intention to accord victims individualized treatment. Accordingly, we cannot conclude that the district court’s methodology for calculating loss was improper.

D Vysniauskas also objects to the district court’s addition of $500 in damages for each of the 108 access devices (i.e., credit cards, bank cards, and PIN numbers) found in his apartment. As a general matter, such a presumption is “reasonable,” and authorized by the Sentencing Guidelines, which provides that the loss associated with each such device “shall not be less than $500.” USSG § 2B1.1 cmt. n. 3(F)(i).

Admitting this, Vysniauskas makes three specific objections: 1) unusable cards should be excluded, 2) cards in his and Leichanka’s actual names should be excluded, and 3) cards actually used to set up fraudulent accounts should not be double counted with the actual and check-face-value loss totals. Under the Sentencing Guidelines, “unauthorized access device” has the meaning that the term is given in 18 U.S.C. § 1029, which deals with fraud in connection with access devices.

Section (e)(1) of the statute defines “access device” as “any card ... or other means of account access that can be used ... to obtain money, goods, services, or any other thing of value, or that can be used to initiate a transfer of funds.” 18 U.S.C. § 1029(e)(1).

Section (e)(3) adds that “unauthorized” refers to devices “lost, stolen, expired, revoked, canceled, or obtained with intent to defraud.” 18 U.S.C. § 1029(e)(3). Actual use is not required. United States v. Shifu Lin, 508 Fed.Appx. 398, 403-04 (6th Cir.2012), cert. denied, — U.S. —, 133 S.Ct. 1300, 185 L.Ed.2d 227 (U.S.2013); United States v. Gilmore, 431 Fed.Appx. 428, 430 (6th Cir.2011). Although neither the statute nor the Guidelines specifically require “usability,” in interpreting the statute the Ninth Circuit has read in such a requirement. United States v. Onyesoh, 674 F.3d 1157, 1159 (9th Cir.2012).

To the extent we recognize a “usability” requirement, it must be permissive enough to encompass the “expired, revoked, [and] canceled” devices expressly included in the statute. Thus a device need not be usable in its primary “access” sense: unauthorized charges are not the only way credit cards can be misused. In a check-fraud case, bank and credit cards — whether expired, fake, or un-activated — can be used to create and corroborate false identities. Where the usability of access devices is “self-evident” from the nature of the fraud, further proof is unnecessary. See Onyesoh, 674 F.3d at 1160.

Nor is there good reason to exclude the cards in Vysniauskas’s and Leichanka’s true names. The statutory definition includes not just lost or stolen devices in other people’s names, but any device “obtained with intent to defraud.” This phrase naturally encompasses devices obtained under real names as well as false names, especially when backed by other false information or an overriding fraudulent purpose. Here, as the district court found, there is sufficient evidence that the eight devices in their true names were associated with accounts involved in the overall fraudulent scheme.

Even if all of the purchases and transactions made through these accounts were legitimate, they helped to provide a cover of ordinary activity that lowered the banks’ suspicions of the fraudulent withdrawals. Vysniauskas’s final contention is likewise unpersuasive. The Guidelines entitle the government to apply a presumption of $500 loss per card unless it seeks to establish a higher loss amount. USSG § 2B1.1 cmt. 3(F)(i).

Vysniauskas argues that the government cannot apply the $500-per-card figure because it failed to show that it did not already establish a loss amount for each of the 103 cards in calculating actual and intended losses to banks. However, the government does not need to show the absence of double counting in order to benefit from this presumption because the burden of rebutting the government’s presumption falls upon Vysniauskas. See, e.g., St. Mary’s Honor Ctr. v.

Hicks, 509 U.S. 502, 507, 113 S.Ct. 2742, 125 L.Ed.2d 407 (1993) (“To establish a ‘presumption’ is to say that a finding of the predicate fact ... produces a ‘required conclusion in the absence of explanation’.... ”); see also, Usery v. Turner Elkhorn Mining Co., 428 U.S. 1, 26, 96 S.Ct. 2882, 49 L.Ed.2d 752 (1976) (“Each presumption is explicitly re-buttable, and the effect of each is simply to shift the burden of going forward with evidence from the claimant to the [defendant].”).

Vysniauskas does not produce any evidence that the government doubled counted the loss amount "with respect to any of the 103 cards and, therefore, the government is entitled to apply the $500-per-card presumption. E Vysniauskas also objects to the inclusion in the loss total of $5,657.09 in non-sufficient-funds checks written to Costco, since 1) the government did not prove the loss by a preponderance of the evidence, 2) passing bad checks to retailers is not bank fraud, and 3) the activity was too remote from the offense of conviction.

The district court did not clearly err in finding the not-sufficient-funds loss to be proven by a preponderance of the evidence. The loss was established by a report from the Oakland County Sheriffs Department; the checks were signed by Eduardas Zenkevicius, one of Vysniaus-kas’s known aliases; the associated bank account had been subpoenaed by the FBI during its fraud investigation; and numerous Costco membership cards were found in the search of Vysniauskas’s apartment.

It is true that Agent Gavin misspoke, in stating that the Comerica bank account on which the checks were drawn was included in the government’s spreadsheet of fraudulent accounts. However, the district court did not rely on this statement, specifically noting that “in addition to the fraudulent checks summarized by the Government’s Exhibit 8 [the spreadsheet], the Government provided evidence ... that Vysniaus-kas ... issued non-sufficient funds checks to Costco.” Dist. Ct. Op. at 14 (emphasis added). Vysniauskas’s second argument is incorrect.

As mentioned, writing not-sufficient-funds checks to merchants can support bank fraud under 18 U.S.C. § 1344, since the bank is exposed to an actual risk of loss. Reaume, 338 F.3d at 581. Vysniauskas argues, however, that his 2008 Costco fraud is not sufficiently related to his kiting to be considered “relevant conduct” under the Guidelines. For the purposes of sentencing, judges may only take into account “relevant conduct,” that is, offenses “part of the same course of conduct or common scheme or plan as the offense of conviction.” United States v.

Hill, 79 F.3d 1477, 1481 (6th Cir.1996) (quoting USSG § lB1.3(a)(2)). The Guidelines require the offenses to be either “substantially connected to each other by at least one common factor, such as common victims, common accomplices, common purpose, or similar modus operandi,” USSG § 1B1.3 cmt. n. 9(A), or “part of a single episode, spree, or ongoing series of offenses,” USSG § 1B1.3 cmt. n. 9(B). Conduct not charged in the indictment may be considered, although if the offense is sufficiently remote in time, the district court will err in considering it. United States v.

Kappes, 936 F.2d 227, 229, 230-31 (6th Cir.1991). The district court did not clearly err in considering the not-sufficient-funds checks to Costco as relevant to his check-kiting fraud. Although the primary victims were different, both schemes shared the same underlying characteristic: passing bad checks under a false name, with intent to cash out as much of the value as possible. In addition, a search of Vys-niauskas’s apartment turned up numerous Costco and other membership cards, presumably still a part of the ongoing fraud.

The relevance of the Costco scheme is especially clear when viewed in light of the diversity of Vysniauskas’s fraud: in addition to the kiting, he fraudulently disputed transactions, made charges on a stolen credit card, and simply overdrew accounts. The district court did not clearly err in finding evidence of a “single evolving [] conspiracy” rather than “several discrete and separate” frauds. United States v. Barbour, 393 F.3d 82, 92 (1st Cir.2004). Vysniauskas argues that the Costco checks should not be included because they predated the conduct charged in the indictment by 17 months.

However, the spreadsheet compiled by the FBI showed kiting activity going back a full year before the date indicated in the indictment. The gap in the records, therefore, is only five months (May to October 2008). This is not enough to negate the conclusion that Vsyniauskas engaged in a prolonged, albeit varied, course of bank fraud. Until 2010, even the check-kiting activity was intermittent, with multiple multi-month gaps between bad checks — the longest from January to August 2009. This “ongoing series of offenses” is sufficient to constitute the “same course of conduct” under the Guidelines.

Ill Vysniauskas also raises various other challenges to sentencing enhancements applied by the district court. As with loss, the district court’s factual determinations here are reviewed for clear error. United States v. Maken, 510 F.3d 654, 656-57 (6th Cir.2007); see also United States v. Kraig, 99 F.3d 1361 (6th Cir.1996) (“sophisticated means” is question of fact). Whether a person is a victim under the Guidelines, on the facts found by the district court, is a legal question we review de novo. United States v. Stubblefield, 682 F.3d 502, 510 (6th Cir.2012).

Likewise, the “mixed law- and-fact conclusion that the facts found by the district court constituted an obstruction of justice” is reviewed de novo. United States v. Lineberry, 7 Fed.Appx. 520 (6th Cir.2001). A Vysniauskas contends that the district court improperly applied an enhancement for ten or more victims under § 2Bl.l(b)(2)(A), specifically arguing that there was insufficient evidence that one of his alleged identity-theft victims, Maria Babanina, was a real person; and that there was insufficient evidence that his mail-theft victims were real, or that their mail was undelivered.

The notes to the Sentencing Guidelines define “victim” as “any person who sustained any part of the actual loss.” USSG § 2B1.1 cmt. n. 1. However, in addition to this general definition, the Guidelines also include “any person who was the intended recipient, or addressee, of [ ] undelivered United States mail” and “any individual whose means of identification was used unlawfully or without authority.” USSG § 2B1.1 cmt. n. 4(C), (E). Thus for such victims the government need not prove actual loss. The district court did not clearly err in finding that Maria Babanina was an identity-theft victim.

Agent Gavin recovered an identity document, which he determined to be legitimate, with the name, information, and picture of a person named Maria Babanina. He also recovered a fraudulent identification document with the same name and information, but Leiehan-ka’s picture. The natural conclusion is that Babanina’s ID was lost or stolen, and Leichanka used the information to forge her own. Vysniauskas has provided no other explanation for the existence of this seemingly legitimate identification document. Nor did the district court clearly err in finding five victims of mail fraud.

The stolen mail was official correspondence from the Social Security Administration and the United States Treasury; it is reasonable to assume these agencies were corresponding with real people. Further, the mail was addressed to various apartments in the same complex as Anthony Demaree, a victim of Vysniauskas who actually suffered fraudulent charges on his credit card. The evidence also supports a finding that the mail was undelivered: the apartment complex’s mailboxes were outdoors and unsecured, and “mail taken from [an] addressee’s mail box” is considered undelivered. USSG § 2B1.1 cmt. n. 4(C)(iii).

B Vysniauskas also argues the district court erred in applying a “sophisticated means” enhancement under § 2B1.1(b)(9), characterizing his fraud as time-consuming, but no.t “especially complex or especially intricate.” USSG § 2B1.1 cmt. n. 8(B). If mere use of false identities is enough to constitute sophistication, he argues, nearly every instance of fraud will necessitate this enhancement. In applying this enhancement, courts do not look to the “individual steps” of the scheme, but instead to “the totality of the defendant’s conduct.” United States v.

Masters, 216 Fed.Appx. 524, 526 (6th Cir.2007); see also United States v. Tandon, 111 F.3d 482, 491 (6th Cir.1997). Here, Vysniauskas set up dozens of bank accounts at several financial institutions, stole real identification information and forged identification documents, and tried to cover up fraudulent withdrawals with nearly $200,000 of “legitimate” transactions. Our ease law supports that such repeated use of false identities in an extensive and repetitive scheme supports such an enhancement, even if “the information used was not itself unduly complex or difficult to obtain.” United States v.

Kopietz, 126 Fed.Appx. 708, 711 (6th Cir.2005) (filing 15 fictitious and fraudulent tax returns); see also United States v. Crosgrove, 637 F.3d 646, 667 (6th Cir.2011) (use of pseudonym and issuance of fraudulent insurance certificates); United States v. Middleton, 246 F.3d 825, 848 (6th Cir.2001) (tax fraud concealed by opening of several bank accounts under different company, names, repetitive small withdrawals, and use of cash).

The district court did not err in ruling that this extensive fraud was accomplished through “sophisticated means.” C Vysniauskas challenges the two-level obstruetion-of-justiee enhancement imposed as a result of his attempt to have hid friend conceal evidence. USSG § 3C1.1. The request to have his friend “take and keep” the “envelopes, money, watches,” and other “stuff in his apartment which will get him into more trouble” is within the core definition of obstruction of justice under the Guidelines.

USSG § 3C1.1 cmt. n. 4(D) (“directing or procuring another person to destroy or conceal evidence that is material”). Vysniauskas objects that his obstruction did not occur during the federal “investigation, prosecution, or sentencing of the instant offense,” as only a state investigation had been initiated at the time. USSG § 3C1.1. The Guidelines, however, do not draw any such distinction between federal and state investigations. See United States v.

Ayers, 416 F.3d 131, 134-35 (2d Cir.2005) (enhancement warranted where defendant ordered friend to destroy gun during state-police investigation); see also United States v. Roberts, 243 F.3d 235, 238 (6th Cir.2001) (enhancement warranted where defendant escaped from state authorities before start of federal investigation).

The only requirement is that the state and federal investigations be for the same “instant offense.” Here, the state investigation arose out of exactly the same offense that the federal charges were ultimately based on: attempting to set up bank accounts for fraudulent purposes, using false identification. Vysniauskas’s argument that he had no reason believe there was a “further investigation” underway after his arrest also fails.

When the Guidelines were amended in 2006, the Commission replaced the language “during the course of the investigation, prosecution, or. sentencing” with “with respect to the investigation, prosecution, or sentencing” to make clear that pre-investigative conduct can form the basis of an adjustment under § 3C1.1, expressly rejecting United States v. Baggett, 342 F.3d 536, 542 (6th Cir.2003), in favor of the majority view. USSG Appendix C, amend. 698.

Even under the prior version of the Guidelines, however, Vysniauskas’s actions are clearly within the scope of the enhancement: arrest is typically sufficient to trigger the start of police investigation or prosecution. See USSG § 3C1.1 cmt. n. 4(D) (expressly including destroying or concealing evidence “contemporaneously with arrest”); United States v. Waldon, 206 F.3d 597, 608 (6th Cir.2000) (enhancement for phone call to conceal evidence six hours after arrest); see also United States v. Micklin, 89 Fed.Appx. 977, 982 (6th Cir.2004), vacated due to United States v.

Booker, 543 U.S. 220, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005) (enhancement for disposal of evidence several hours after execution of search warrant). While in custody, both Leichanka and Vysniauskas were being actively questioned, and Vysniaus-kas’s suggestion that state police were planning no further investigation after finding fraudulent documents in the initial car search is pure speculation. A phone call to conceal evidence while in post-arrest custody is clearly obstruction “with respect to the investigation, prosecution, or sentencing.” USSG § 3C1.1.

IV Finally, Vysniauskas argues that his sentence is both procedurally and substantively unreasonable. Sentences are reviewed for procedural and substantive unreasonableness under an abuse-of-discretion standard. Gall v. United States, 552 U.S. 38, 51, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007). A district court commits procedural error by “failing to calculate (or improperly calculating) the Guidelines range, treating the Guidelines as mandatory, failing to consider the § 3553(

a) factors, selecting a sentence based on clearly erroneous facts, or failing to adequately explain the chosen sentence.” Ibid. A sentence is substantively unreasonable if the sentence is “greater than necessary,” as when the district court “selects the sentence arbitrarily, bases the sentence on impermissible factors, fails to consider pertinent § 3553(

a) factors or gives an unreasonable amount of weight to any pertinent factor.” United States v. Tristan-Madrigal, 601 F.3d 629, 632-33 (6th Cir.2010). Where a party has failed to object to a procedural defect, as is the case here, we review claims of procedural unreasonableness for plain error. United States v. Vonner, 516 F.3d 382, 385-86 (6th Cir.2008) (en banc). To show plain error, a defendant must show 1) error 2) that was obvious or clear, 3) that affected defendant’s substantial rights and 4) that affected the fairness, integrity, or public reputation of the judicial proceedings. Vonner, 516 F.3d at 386.

First, Vysniauskas argues that the sentence was procedurally unreasonable because the district court relied on facts that it specifically found were not supported by a preponderance of the evidence. The presentence report stated that ten of the aliases used by Vysniauskas and Lei-chanka were identified as those of real individuals. In a subsequent opinion and order, the district court found that only two of the aliases were proven by a preponderance of the evidence to be of real people, for the purposes of the ten-or-more-victims enhancement.

In apparent contradiction to this finding, at the sentencing hearing the district court referred back to the presentence report, stating that “ten of [Vysniauskas’s aliases] were identified as legitimate and real individuals” and explaining that it was “troubling” that he was “using the identification of real individuals who were going to have real consequences in terms of their credit, their fives.” Vysniauskas did not object at the hearing, thus we review for plain error. Despite appearances, these statements are not completely inconsistent.

As the district court recognized, to be an identity theft “victim” under § 2Bl.l(b)(2)(A), the false identification must be “used.” USSG § 2B1.1 cmt. n. 4(E). As a result, the court excluded these individuals from the victim count only because there was no proof of use of their information, not because of lack of proof of their existence. At sentencing, the district court could consider the potential risk of harm faced by these individuals, even though lack of actual use prevented their characterization as “victims” for enhancement purposes.

Nor was the district court’s concern about the “real consequences” of using the identifieation of real individuals problematic simply because there was no evidence of actual credit problems. The district court made its statement only in terms of possibility (“were going to” as opposed to “did”), and a court is free to consider the risk of foreseeable harms in sentencing. See United States v. Karro, 257 F.3d 112, 121 (2d Cir.2001) (holding that “the risk of non-monetary harms associated with identity theft” is a permissible basis for upward departure).

Vysniauskas also argues that the district court unreasonably failed to give him either a concurrent sentence or credit for time served on his state offense, and also failed to provide a reason for the denial. A district court has discretion to impose a concurrent sentence under USSG § 5G1.3 and 18 U.S.C. § 3584(a); USSG § 5G1.3 also permits the district court to adjust the sentence downward for time served. Since, in exercising this discretion, the court is to consider same § 3553(

a) factors that must be addressed in determining the length of the sentence, 18 U.S.C. § 3584(b), the district court’s determinations of sentence length and concurrency are “intertwined.” United States v. Johnson, 640 F.3d 195, 208 (6th Cir.2011). Thus there is no need for “the district court to state a ‘specific reason’ for a consecutive sentence,” especially because the Guidelines “presume that multiple terms of imprisonment imposed at different times are consecutive.” Id. at 209. Here, the district court heard argument, recognized its discretion, and addressed the appropriate factors. There was no error.

Vysniauskas’s substantive unreasonableness argument is based solely on his never having served a jail sentence before the 78-day term imposed after his April 2010 arrest. To start, a within-Guidelines sentence is presumptively reasonable on appellate review. United States v. Christman, 607 F.3d 1110, 1118 (6th Cir.2010). The main problem with the argument, however, is that Vysniauskas had only avoided jail in the past by failing to appear for arraignment, using multiple false identities, and illegally immigrating back into the United States.

As the district court recognized, this is reason to impose a harsher, not a more lenient sentence. V Because Vyniauskas’s arguments are without merit, we AFFIRM his sentence. . Federal law requires banks to promptly provide access to funds from deposited checks. Expedited Funds Availability Act, 12 U.S.C. §§ 4001-4010. . The court also applied a two-level enhancement for unlawful use of a means of identification; Vysniauskas does not object to this. .

For example, in the National City account under the name Ulugbek Khaydarov, Vys-niauskas successfully withdrew $2000 after depositing a bad $1600 check, using the overdraft protection offered by the bank. Gov't App’x at 5 [Gov’t Exh. 8]. . Watkins, which relied on the Sixth Circuit’s prior rule that "intended loss must have been possible to be deemed relevant,” United States v. Khan, 969 F.2d 218, 221 (6th Cir.1992), was effectively abrogated in 2001 by the amendment to the Sentencing Guidelines now at USSG §'2B1.1 cmt. n. 3(A)(ii), which specifically allowed for impossible losses.

USSG App’x C, Amendment 617 (effective November 1, 2001). . Note that calculation of amount of loss for sentencing purposes is distinct from restitution, which may only be awarded for losses actually caused by the defendant, and not for intended losses.

Younes, 194 Fed.Appx. at 317 (“While for sentencing purposes 'loss’ is defined as the greater of either the ‘actual’ or the ‘intended’ amount lost due to the fraud, for restitution purposes the [Mandatory Victims Restitution Act, 18 U.S.C. § 3663] implicitly requires that the restitution award be based on the amount of loss actually caused by the defendant’s offense.”). . Tellingly, Vysniauskas did not object at the time, failing to challenge the ten-or-more-victims enhancement at all. (Amended Sentencing Memorandum, R.54 at 5.)