UNITED STATES of America, Appellee, v. David STAPLES, also known as Andrew N. Blatt, Appellant, 410 F.3d 484

MORRIS SHEPPARD ARNOLD, Circuit Judge. David Staples appeals from his convictions and sentence for bank fraud, see 18 U.S.C. § 1344(1), (2), misuse of a social security number, see 42 U.S.C. § 408(a)(7)(B), and fraud with identification documents, see 18 U.S.C. § 1028(a)(7). With respect to his convictions, he argues that the district court should have suppressed evidence from two pretrial identifications, excluded two in-court identifications, and granted him a new trial because he received ineffective assistance of counsel. As for his sentence, Mr.

Staples argues that the district court miscalculated the intended loss under the sentencing guidelines and, in contravention of Blakely v. Washington, 542 U.S. 296, 124 S.Ct. 2531, 159 L.Ed.2d 403 (2004), made the predicate factual findings for two sentencing enhancements rather than allowing the jury to make them. We affirm the convictions, but remand the case for resentencing. I. Mr. Staples argues that we should reverse his convictions and grant him a new trial because the district court deprived him of his due process rights when it denied his motion to suppress evidence of two pretrial identifications.

A man posing as Dr. Andrew Blatt bought a house with a counterfeit check. The check bounced, and the impostor became a wanted man. A few months after the closing, a television station in St. Louis broadcast a picture of Mr. Staples, as his picture was on the driver’s license used by the impostor and photocopied by the employees of the title company at the closing. A state probation officer and a St. Louis police officer independently saw the broadcast, identified the man in the photograph as Mr. Staples, and called the agents working on the case to tell them so.

After receiving this information, the case agents showed the same photo spreads separately to Michelle Clemons and Mary Layne, the two employees of the title company who saw the impostor on the day of the closing. Prior to showing the arrays, the agents did not ask either witness to work with an artist on a composite drawing or to give a full description of the face of the impostor. Nor did the agents show either witness a blank lineup (ie., a lineup without Mr. Staples’s picture in it) before showing the challenged photo spreads. The agents told each witness that the picture of the person who posed as Dr.

Blatt might be in the spread and then asked if the impostor’s picture was in the array. Ms. Clemons and Ms. Layne each identified the picture of Mr. Staples as the man who posed as Dr. Blatt. At trial, both witnesses again identified Mr. Staples as the fake Dr. Blatt. The jury convicted Mr. Staples on all three counts of the indictment. We review the district court’s

interpretation of the protections afforded by the fifth amendment’s due process clause de novo, while we review its underlying factual determinations for clear error. See United States v. Rose, 362 F.3d 1059, 1066 (8th Cir.2004); United States v. Smith, 383 F.3d 700, 703-04 (8th Cir.2004). To succeed in challenging the validity of the photo-array identifications on due process grounds, Mr. Staples must show that the identification procedure was impermissibly suggestive and that the suggestive procedure created a “very substantial likelihood of irreparable misidentification.” Simmons v.

United States, 390 U.S. 377, 384, 88 S.Ct. 967, 19 L.Ed.2d 1247 (1968); Robinson v. Clarke, 939 F.2d 573, 575 (8th Cir.1991) (per curiam). Mr. Staples contends that the identical lineups were unduly suggestive because the witnesses were not asked to create composite sketches, to give full descriptions of the impostor’s face, or to view blank lineups prior to seeing the challenged photo spreads.

Additionally, he argues, the lineups were suggestive because the witnesses may have looked at the photocopy of the driver’s license after the closing and before the lineups, and because they likely saw the photograph of Mr. Staples on television. Mr. Staples maintains that the latter two circumstances matter because someone stole his driver’s license, cut out the picture, and used it on the fraudulent Dr. Blatt driver’s license. So, Mr. Staples insists, if Ms. Clemons and Ms.

Layne had looked at the photocopy of the license after the closing or had seen the news story which used the same picture, they would have seen his face and started to believe that he was the man at the closing, though he was not. He concludes that the suggestive lineups tainted his trial. We are unpersuaded by Mr. Staples’s arguments. We reject the argument that a witness’s failure to work on a composite drawing renders a photo array suggestive, because Mr. Staples does not explain why this is so and does not cite any case for this proposition. See Watson v. O’Neill, 365 F.3d 609, 615 (8th Cir.2004).

For the same reason, we reject the argument that the agents corrupted the lineups by failing to obtain descriptions of the impostor’s face. Next, as we have held previously, a photo array is not unconstitutionally suggestive just because it was not preceded by a blank lineup. United States v. Amrine, 724 F.2d 84, 88, 88 & n. 5 (8th Cir.1983). Finally, we refuse to conclude that the lineups were suggestive because the witnesses might have reviewed the photocopy of the license or seen the television news story: Mr.

Staples points us to no evidence showing that either witness looked at the photocopy after the closing or saw the story. Because the lineups were not unduly suggestive, we do not need to address the question of whether they caused the pretrial and in-court identifications to be unreliable. See Robinson, 939 F.2d at 575. We note that Mr. Staples’s brief states that “Ms. Layne ... did not deny looking at the photograph” in the file after the closing and before viewing the photo array. It also says that Ms. Layne did not deny seeing the news story about Mr. Staples.

The difficulty with these statements is that during her trial testimony Ms. Layne was never asked whether she looked at the photograph after the closing and before the array or whether she saw the news story. Thus, while the statements may be true, they are also misleading. The argument is entirely disingenuous. II. Mr. Staples also argues that his convictions should be reversed because his trial counsel (who withdrew before sentencing) offered ineffective assistance under the sixth amendment by failing to call two witnesses, Ronald Harlan and Mr.

Staples’s uncle, who would have testified that he was not the man who passed himself off as Dr. Blatt. Mr. Harlan was the seller of the house, and he has pleaded guilty to bank fraud in a related case. According to the defendant, Mr. Harlan met the impostor several times and would have testified that Mr. Staples was not the impostor. Mr. Staples’s uncle has also pleaded guilty in a related case. Mr. Staples insists that his uncle would have testified that Mr. Staples’s driver’s license was stolen from his (the uncle’

s) car (which bolsters Mr. Staples’s story that a third party presented his driver’s license picture at the closing) and that Mr. Staples could not have been involved in the fraud. Before proceeding to the merits of Mr. Staples’s claim, we must decide if it is amenable to review on direct appeal. Because ineffective-assistance-of-counsel claims often encompass facts outside of the district court record, they are typically best raised collaterally via motions filed under 28 U.S.C. § 2255. United States v. Ford, 918 F.2d 1343, 1350 (8th Cir.1990).

When the claim does not turn on information outside of the district court record, however, disposition on direct appeal is appropriate. See id. We have previously decided on direct appeal ineffective-assistance claims premised on an attorney’s failure to call witnesses, e.g., United States v. Hunter, 95 F.3d 14, 17-18 (8th Cir.1996), and we do so again here. Mr. Staples’s claim requires us to assess the hypothetical costs and benefits of calling his uncle and Mr. Harlan, and this task does not require off-the-record information (as do, for example, claims about counsel’s advice given in confidence).

To prevail on his ineffective-assistance claim, Mr. Staples must show that his trial counsel’s representation “fell below an objective standard of reasonableness,” Strickland v. Washington, 466 U.S. 668, 687-88, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), and that he was prejudiced such that -“there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different,” id. at 694, 104 S.Ct. 2052.

When assessing attorney performance, “[cjourts should avoid ‘the distorting effects of hindsight’ and try to evaluate counsel’s conduct by looking at the circumstances as they must have appeared to counsel at the time.” Sera, 267 F.3d at 874 (quoting Strickland, 466 U.S. at 689, 104 S.Ct. 2052). Because of the difficulty of avoiding such bias, courts “must indulge a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance.” Strickland, 466 U.S at 689, 104 S.Ct. 2052. Mr. Staples’s ineffective-assistance-of-counsel claim is untenable.

The decision not to call a witness is a “ ‘virtually unchallengeable’ ” decision of trial strategy, see United States v. Davidson, 122 F.3d 531, 538 (8th Cir.1997), cert. denied, 522 U.S. 1034, 118 S.Ct. 639, 139 L.Ed.2d 617 (1997) & 523 U.S. 1033, 118 S.Ct. 1329, 140 L.Ed.2d 490 (1998) (quoting Bowman v. Gammon, 85 F.3d 1339, 1345 (8th Cir.1996), cert. denied, 520 U.S. 1128, 117 S.Ct. 1273, 137 L.Ed.2d 350 (1997)), and Mr. Staples’s attorney’s decisions do not fall outside this wide perimeter of protection. Testimony from Mr. Staples’s uncle and Mr.

Harlan might have provided some benefits, but given the matters described below, it was not objectively unreasonable for Mr. Staples’s attorney to decide that the potential costs of calling the witnesses outweighed the potential benefits. The prosecutor’s cross-examination of Mr. Harlan likely would have diminished greatly the value of his testimony. Mr.

Harlan was under indictment' in a related fraud case, and the conduct underlying that indictment almost certainly would have been suitable material for impeachment pursuant to Federal Rule of Evidence 608(b), as fraudulent conduct implicates a witness’s character for truthfulness, e.g., United States v. Smith, 80 F.3d 1188, 1193 (7th Cir.1996). In addition, any claim by Mr. Harlan that Mr. Staples was not the ersatz Dr. Blatt would have been contradicted by the eyewitness testimony of Ms. Clemons and Ms. Layne. (This analysis assumes, perhaps unrealistically, that Mr.

Harlan would not have invoked his fifth amendment right against self-incrimination when called to testify even though he was facing related criminal charges.) Impeachment specifics aside, there is considerable risk inherent in calling any witness because if the witness does not hold up well on cross-examination, the jurors might draw unfavorable inferences against the party who called him or her. See Lema v. United States, 987 F.2d 48, 54 (1st Cir.1993). As Mr. Harlan would have faced a vigorous cross-examination, there was a not insignificant possibility that he would not have held up well. Mr.

Staples’s uncle would have been an even less convincing witness, because he shared all of Mr. Harlan’s weaknesses as a witness (he too was under indictment in a related fraud case, would have contradicted the two eyewitnesses, and faced a tough cross-examination) and possessed one more: Because he is related to Mr. Staples, the prosecutor could have argued that his testimony was shaped by familial loyalty. Mr. Staples did not receive ineffective assistance from his lawyer. In light of the deferential presumption, the prospects for impeachment, and the risk inherent in calling any witness, we conclude that Mr.

Staples’s attorney did not perform in an objectively unreasonable manner by deciding not to call the two witnesses. Because Mr. Staples’s attorney’s performance was not unreasonable, we need not address the question of whether his performance prejudiced Mr. Staples’s defense. III. Mr. Staples- maintains that the district court erred in determining, his sentencing range because it miscalculated the loss that he intended to cause. We briefly review the relevant facts. At a real estate closing, Mr. Staples gave the title company a fake check for $249,493, the purchase price of the house.

About $76,000 of the purchase price represented the seller’s equity in the property. The title company issued Mr. Harlan, the seller (and an active participant in the fraud), a check for this amount, which he cashed before the title company discovered that Mr. Staples’s • check was counterfeit. About $170,000 of the purchase price represented money still owed to the bank holding the mortgage. The title company paid the bank the amount of the mortgage, but the bank returned the funds when it learned that Mr. Staples had given the title company a fake check. Subsequently, the bank foreclosed on and sold the house.

It received only about $85,000 for the house at auction, thus incurring a loss of around $85,000. (A few thousand dollars of the purchase price of the house went for fees related to the purchase, but we ignore these as they do not affect the outcome.) To calculate the sentencing range for a fraud case, we need to determine the amount of loss caused by the fraud. U.S.S.G. § 2B1.1. (All citations to the sentencing guidelines are to the 2002 edition, which the district court used in sentencing Mr. Staples.) The loss amount for sentencing purposes is the greater of the actual or intended loss.

U.S.S.G. § 2B1.1 comment. (n.2(A)). The parties agree that the actual loss in this case was about $161,000, which is the sum of the approximately $76,000 lost by the title company and the approximately $85,000 lost by the bank. The parties do not agree on the amount of the intended loss, however, because they disagree about whether to deduct the apparent value of the house (i.e., what a reasonable person at the time would have thought the house to be worth) from the intended loss. Intended loss is “the loss the defendant intended to cause to the victim.” United States v. Wells, 127 F.3d 739, 746 (8th Cir.1997).

The government. must prove the intended loss by a preponderance of the evidence. United States v. Jackson, 155 F.3d 942, 948 (8th Cir.1998). Absent other evidence of the defendant’s intent, the size of the maximum loss that a fraud could have caused is circumstantial evidence of the intended loss which satisfies the preponderance of the evidence standard. See Wells, 127 F.3d at 746-47.

Importantly, though, as our goal is to capture the defendant’s intent, we compute the maximum possible loss from the perspective of a reasonable person in the defendant’s position at the time of the fraud; we do not assume either omniscience or irrationality. Cf. United States v. Wheeldon, 313 F.3d 1070, 1072-1073 & n. 2 (8th Cir.2002). In other words, the maximum possible loss is the maximum loss that a reasonable person in the defendant’s position could have intended.

Stripped to its core, the intended loss disagreement is a dispute about whether collateral should be considered when determining intended loss: if an immobile object is the subject of a fraudulent purchase, then it is in essence collateral securing the fraudulent purchase because it can be levied on (assuming it has not been sold by the fraudulent purchaser to a bona fide purchaser). We review the district court’s resolution of this issue de novo because it involves the

interpretation and application of the guidelines. United States v. Mashek, 406 F.3d 1012, 1016 (8th Cir.2005). Mr. Staples insists that he could not have intended to defraud anyone of the apparent value of the house because a reasonable person would have known that it could be sold by the bank (or perhaps the title company) to mitigate any damages. The government responds that the fact that losses were recouped by a later sale of the house is irrelevant under the guidelines. At Mr.

Staples’s sentencing, the district court determined that he intended to cause a loss of $249,493; in other words, it held him responsible for the apparent value of the house. We hold that courts should consider collateral when determining the intended loss amounts attendant to transactions like the one involved in this case. If a reasonable person intends for the collateral to revert to the defrauded party (or understands that it will), then he or she does not intend to obtain the value of that collateral.

We do not mean that the value of the collateral necessarily must be deducted from the intended loss; the defendant’s intent is the touchstone. For example, if a car were collateral in a fraudulent loan procurement case, and the defendant were to hide the car, then the court should not deduct the value of the collateral from the intended loss because under those eircumstances the defendant intended the loss to encompass the value of the collateral. See United States v. Williams, 292 F.3d 681, 686-87 (10th Cir.2002).

We are not alone in concluding that collateral is relevant to determining intent, as the Ninth Circuit reached the same conclusion in United States v. McCormac, 309 F.3d 623, 629 (9th Cir.2002). The government’s argument to the contrary is unpersuasive. According to the government, that collateral is irrelevant because, in its words, the guidelines instruct courts to ignore “later re-payments or recoupments of money by victims” when calculating loss. Intended loss is meant to be a measure of a defendant’s culpability. U.S.S.G. § 2B1.1 comment, (background); United States v.

McBride, 362 F.3d 360, 375 (6th Cir.2004). With this in mind, courts do not subtract from the intended loss repayments made by a defendant to his victim after the detection of the offense, as such payments, given their timing, likely do not indicate anything about the defendant’s culpability. United States v. Sivanson, 360 F.3d 1155, 1168-69 (10th Cir.2004); see U.S.S.G. § 2B1.1 comment. (n.2(E)(i)). For the opposite reason, though, courts do subtract payments that were made before the offense was discovered. See U.S.S.G. § 2B1.1 comment. (n.2(E)(i)).

To us, collateral is more like pre-detection repayment than post-detection repayment because it bears on the defendant’s culpability: the existence of collateral could indicate that the defendant intended to cause a smaller loss than would have occurred absent the collateral. Thus, sentencing courts should subtract the value of the collateral from the intended loss amount when the facts warrant it. Applying the principles discussed above to the facts of this case, we conclude that the intended loss was smaller than the actual loss. The only evidence of Mr.

Staples’s intent is the maximum possible loss that could have been caused by the fraud. This being so, we apply our inferential rule and conclude that a reasonable person in Mr. Staples’s position would have intended a loss of less than the actual loss of $161,000. First, a reasonable person would not have thought that he could keep the house because a house cannot be hidden like a car or other mobile form of security.

Second, a reasonable person would have thought that the house was worth at least as much as the outstanding value of the mortgage (if not more, as houses often appreciate), and hence could be sold to recoup much of the loss. Here, the house was worth far less than the outstanding value of the mortgage, but we see no indication that a reasonable person in the defendant’s position would have known this fact. In short, a reasonable person would have thought that at least the value of the mortgage would be recouped and so would have intended a loss of less than $161,000.

Since the actual loss of about $161,000 was greater than the intended loss, the court should have used the actual loss amount to calculate Mr. Staples’s sentencing range. Instead, the court calculated Mr. Staples’s sentencing range using an intended loss amount of $249,493. The court thus set Mr. Staples’s offense level two levels higher than it should have; his final offense level is 18, not 20. See U.S.S.G. § 2B1.1 (b)(1). As Mr. Staples falls in criminal history category III, his sentencing range is 33-41 months’ imprisonment, and not 41-51 months’ imprisonment as determined by the district court.

See U.S.S.G. § 5A. We turn now to the appropriate remedy for this error. If the district court has incorrectly interpreted or applied the guidelines, then pursuant to 18 U.S.C. § 3742(f)(1) we are to remand the case for resentencing, unless the error is a harmless one. Mashek, 406 F.3d at 1015, 1017. An error is harmless if it is clear from the record that the district court would have given the defendant the same sentence regardless of which guidelines range applied. See United States v. Bassett, 406 F.3d 526, 527 (8th Cir.2005) (per curiam). There is no indication here that the court would have given Mr.

Staples a 44-month sentence regardless of what guidelines range applied: the court did not sentence Mr. Staples in the overlap between the correct and incorrect guidelines ranges (41 months’ imprisonment) and did not otherwise suggest that the guidelines range was irrelevant to its sentence. On the record before us, then, we cannot conclude that the miscalculation did not affect the district court’s selection of the sentence imposed, and we thus remand the case for resentencing. See Williams v. United States, 503 U.S. 193, 203, 112 S.Ct. 1112, 117 L.Ed.2d 341 (1992); Mashek, 406 F.3d at 1020.

Before concluding, we briefly address Mr. Staples’s argument that his sentence runs afoul of Blakely. We interpret this as an argument pursuant to United States v. Booker, — U.S.-, 125 S.Ct. 738,160 L.Ed.2d 621 (2005), and conclude that it is moot. We have already decided to remand Mr. Staples’s case for resentencing, and the district court will necessarily resentence him in accord with Booker. See id. at 769. TV. For the above stated reasons, we affirm Mr. Staples’s convictions and remand the case for resentencing.