United States v. Howard , 784 F.3d 745 ( 2015 )


Menu:
  •                                                                                FILED
    United States Court of Appeals
    Tenth Circuit
    PUBLISH
    April 28, 2015
    UNITED STATES COURT OF APPEALS
    Elisabeth A. Shumaker
    Clerk of Court
    TENTH CIRCUIT
    UNITED STATES OF AMERICA,
    Plaintiff - Appellee,
    v.                                                         No. 14-1075
    ROGER KEITH HOWARD,
    Defendant - Appellant.
    APPEAL FROM THE UNITED STATES DISTRICT COURT
    FOR THE DISTRICT OF COLORADO
    (D.C. No. 1:12-CR-00039-RBJ-1)
    Elizabeth L. Harris, Law Office of Elizabeth L. Harris, LLC, Denver, Colorado, for
    Defendant -Appellant.
    Paul Farley, Assistant United States Attorney (John F. Walsh, United States Attorney,
    with him on the brief), Denver, Colorado, for Plaintiff - Appellee.
    Before KELLY, LUCERO, and HARTZ, Circuit Judges.
    HARTZ, Circuit Judge.
    Defendant Roger Howard pleaded guilty to three counts of wire fraud, see
    
    18 U.S.C. § 1343
    , and one count of money laundering, see 
    id.
     § 1957, arising from his
    participation in three mortgage-fraud schemes. His participation included identifying
    property buyers, arranging for their applications for mortgage loans to overstate assets
    and incomes, and obtaining inflated property appraisals and kickbacks to himself and
    some buyers. All buyers defaulted on their mortgage notes. Some notes had been sold
    by the original lenders to downstream lenders, who may themselves have resold the
    notes.
    The United States District Court for the District of Colorado sentenced Defendant
    to 108 months’ imprisonment and ordered him to pay $8,862,191.18 in restitution. He
    argues that the district court made two errors in imposing the sentence: (1) it improperly
    increased his offense level by miscomputing the loss to the mortgage lenders, see USSG
    § 2B1.1; and (2) it awarded restitution to alleged victims without evidence of their actual
    losses. Exercising jurisdiction under 
    28 U.S.C. § 1291
     and 
    18 U.S.C. § 3742
    , we affirm
    the determination of loss under USSG § 2B1.1, which was calculated in accordance with
    our precedents, but we largely agree with Defendant’s restitution argument and therefore
    reverse the restitution order and remand for further proceedings.
    I.       LOSS UNDER USSG § 2B1.1(B)(1)
    Under the sentencing guideline for fraud, the offense level is based on the amount
    of loss. See USSG § 2B1.1(b)(1). The district court calculated the loss caused by
    2
    Defendant to be $8,961,191.18. For losses exceeding $7 million but less than $20
    million, the offense level is 20. See id. § 2B1.1(b)(1)(K).
    In mortgage-fraud cases like this, “[a]ctual loss” under USSG § 2B1.1 cmt.
    n.3(A)(i) “is the unpaid portion of the loan as offset by the value of the collateral.”
    United States v. Crowe, 
    735 F.3d 1229
    , 1241 (10th Cir. 2013) (internal quotation marks
    omitted), cert. denied, 
    134 S. Ct. 1565
     (2014). “[S]o long as it is foreseeable that loans
    will be sold or repackaged, both the original lenders and downstream lenders are
    foreseeable victims of the fraud, and the general formula applies.” United States v.
    Smith, 
    705 F.3d 1268
    , 1276 (10th Cir. 2013). “That is so because any gains or losses
    sustained by the original lender will be offset by a corresponding loss or gain by the
    downstream lender, leaving the total loss to equal mortgage balance minus foreclosure
    price.” 
    Id.
     As a result, “the number of lenders involved and the amount of profit made
    [or loss suffered] by the original lender or any intermediate lenders is mathematically
    irrelevant to the calculation of the total loss caused by the fraud.” 
    Id.
     (emphasis and
    internal quotation marks omitted).
    Defendant does not dispute that the district court’s method of calculating loss was
    the method dictated by our precedents. Instead, he challenges the loss amount based on
    three arguments not raised below: (1) the government’s evidence was insufficient to
    prove $709,588 in losses on eight loans included in the loss amount; (2) the court should
    have reduced the loss amount by $973,935 to account for interest payments made on the
    loans; and (3) the court should not have included $313,261 in losses to a downstream
    3
    noteholder that purchased three loans after the buyers had defaulted. Based on these
    arguments, he concludes that the correct total loss amount is $6,964,407 (instead of
    $8,961,191). See Aplt. Br. at 27. Because his offense level and guidelines range remain
    the same unless the net actual loss is $7 million or less, see USSG § 2B1.1(b)(1)(K),
    Defendant cannot prevail if we reject any argument challenging more than $35,593.
    When the defendant objects to the loss calculation below, we review the district
    court’s factual findings for clear error and calculation methodology de novo. See Crowe,
    735 F.3d at 1235‒36. But because Defendant failed to object below on the grounds
    argued here, we review only for plain error. See id. at 1242. Relief is available under the
    plain-error standard only if Defendant establishes four elements: “(1) the district court
    committed error; (2) the error was plain—that is, it was obvious under current well-
    settled law; (3) the error affected the Defendant’s substantial rights; and (4) the error
    seriously affected the fairness, integrity, or public reputation of judicial proceedings.”
    United States v. Gantt, 
    679 F.3d 1240
    , 1246 (10th Cir. 2012) (brackets and internal
    quotation marks omitted). Defendant must establish all four elements. See 
    id.
     “[T]he
    failure of any one will foreclose relief and the others need not be addressed.” 
    Id.
    Defendant argues that the district court committed plain error by counting losses
    on certain loans totaling $709,588. Among those losses are the amounts of five second-
    mortgage loans totaling $422,588. The court calculated the losses from printouts from
    UTLS Default Services. The printouts showed the amounts of the loans and named the
    noteholders. Defendant asserts that the amounts cannot be believed because each
    4
    printout is indisputably wrong in naming the holder of the first-mortgage note on the
    property. This challenge to the loss calculation raises solely a question of fact—was
    there a second mortgage in the amount stated on the printout? But “factual disputes
    regarding sentencing not brought to the attention of the district court do not rise to the
    level of plain error.” United States v. Lewis, 
    594 F.3d 1270
    , 1288 (10th Cir. 2010)
    (brackets and internal quotation marks omitted).
    Defendant relies on United States v. Goode, 
    484 F.3d 676
    , 681 (10th Cir. 2007),
    for the proposition that sufficiency of the evidence can be reviewed for plain error.
    Goode, however, involved a challenge to the sufficiency of the evidence of guilt at a
    criminal trial, where different considerations are in play than with sentencing.1 More
    importantly, the issue raised by Defendant is one of admissibility of evidence—was the
    printout sufficiently reliable to be used to establish the amount of the second mortgage
    notes?—not sufficiency. Because Defendant failed to object to the evidence below, there
    was no need for the government to explain why the printout was likely to be accurate.
    Defendant has given us no reason to believe that the government could not present
    reliable evidence on remand of the amount of the second-mortgage loans. See Lewis, 594
    1
    For example, if the appellate court determines that there was insufficient evidence of
    guilt, the Double Jeopardy Clause forbids the government from gathering more evidence
    and subjecting the defendant to a second trial, see Burks v. United States, 
    437 U.S. 1
    ,
    14‒17 (1978); but the Double Jeopardy Clause does not apply to noncapital sentencing,
    see Monge v. California, 
    524 U.S. 721
    , 728‒29 (1998). As a result, only in the
    sentencing context does an appellate court reviewing for plain error consider whether
    anything ultimately would be accomplished by a remand for further proceedings. See
    Lewis, 
    594 F.3d at 1288
    .
    5
    F.3d at 1288. We are not disposed to ignore our binding precedents regarding the scope
    of plain-error review of sentencing determinations.
    II.    RESTITUTION
    The award of restitution in this case is governed by the Mandatory Victims
    Restitution Act of 1996 (MVRA), which “requires certain offenders to restore property
    lost by their victims as a result of the crime.” Robers v. United States, 
    134 S. Ct. 1854
    ,
    1856 (2014). Defendant’s principal challenge is to the method of calculating the loss to
    downstream lenders—that is, lenders who did not originate the mortgage loan but
    purchased it from the original lender or an earlier downstream lender. Because we agree
    with this challenge and remand for further proceedings, we need not address his other
    challenges, which had not been raised below and can be considered on remand.
    The MVRA requires that a defendant convicted of an offense against property,
    including any offense committed by fraud or deceit, be ordered to pay restitution to
    victims of the offense. See 18 U.S.C. § 3663A(a)(1), (c)(1)(A)(ii). Payment is to be
    made to “an identifiable victim or victims [who] suffered . . . pecuniary loss.” Id.
    § 3663A(c)(1)(A)(ii), (B). A victim is “a person directly and proximately harmed as a
    result of the commission of an offense for which restitution may be ordered.” Id.
    § 3663A(a)(2). “Restitution must not unjustly enrich crime victims or provide them a
    windfall.” United States v. Ferdman, 
    779 F.3d 1129
    , 1132 (10th Cir. 2015). District
    courts thus “may not order restitution in an amount that exceeds the actual loss caused by
    the defendant’s conduct, which would amount to an illegal sentence.” 
    Id.
     (internal
    6
    quotation marks omitted); see also United States v. Griffith, 
    584 F.3d 1004
    , 1019 (10th
    Cir. 2009) (“a district court that orders restitution in an amount greater than the total loss
    caused by the offense thereby exceeds its statutory jurisdiction and imposes an illegal
    sentence”) (internal quotation marks omitted)).
    In disputes over the amount of a victim’s loss, the government bears the burden of
    persuasion by a preponderance of the evidence. See 
    18 U.S.C. § 3664
    (e). The district
    court “may consider hearsay evidence that bears minimal indicia of reliability so long as
    the defendant is given the opportunity to refute the evidence.” United States v.
    Rodriguez, 
    751 F.3d 1244
    , 1261 (11th Cir. 2014) (internal quotation marks omitted), cert.
    denied, 
    135 S. Ct. 310
     (2014); cf. United States v. Sunrhodes, 
    831 F.2d 1537
    , 1544 (10th
    Cir. 1987) (admission of hearsay testimony with substantial indicia of reliability in
    restitution proceeding did not violate hearsay rule or Confrontation Clause). We review
    the restitution order for an abuse of discretion, which requires us to review factual
    findings for clear error and application of the MVRA de novo. See Ferdman, 779 F.3d at
    1131; United States v. Battles, 
    745 F.3d 436
    , 460 (10th Cir. 2014), cert. denied,
    
    135 S. Ct. 355
     (2014); cf. Cooter & Gell v. Hartmarx Corp., 
    496 U.S. 384
    , 405 (1990)
    (reviewing sanction under Fed. R. Civ. P. 11: “A district court would necessarily abuse
    its discretion if it based its ruling on an erroneous view of the law or on a clearly
    erroneous assessment of the evidence.”).
    The “victim” identified for each loan was the holder of the note when the property
    went into foreclosure. The district court calculated the restitution amount for each
    7
    identified victim using the same method it employed in calculating loss under
    USSG § 2B1.1—by adding the unpaid principal balances on each loan held by the victim
    and subtracting the amounts recovered from sales of the properties securing the loans.
    (The total restitution amount was $99,000 less than the loss under § 2B1.1 because the
    noteholder on one loan had not been identified.)
    Defendant contends that the court “applied an incorrect methodology for
    computing restitution.” Aplt. Br. at 10. He argues that the MVRA limits restitution to
    “actual, out-of-pocket losses,” id., and that the measure of actual loss to a downstream
    lender is “the difference between what the successor lender paid for the loan . . . and
    proceeds obtained from payments and sale of collateral,” id. at 10‒11. The method used
    to calculate restitution in this case—subtracting the amounts recovered through
    foreclosure sales from the unpaid principal balances on the loans—does not reflect actual
    loss to downstream noteholders, he says, because they could have paid less than the
    unpaid balance to acquire the notes. See id. at 11, 30‒31.
    Defendant’s argument is correct. Although the total-loss calculation under
    USSG § 2B1.1 does not depend on which lender in the chain of title of a mortgage note
    suffered what loss, that information is necessary to avoid windfalls in awarding
    restitution. A hypothetical example illustrates why. Say, the original mortgage note was
    for $500,000; the original noteholder sold the note to a downstream lender for $200,000;
    the borrower made no payments on the note; and foreclosure on the property netted
    $100,000. Under the district court’s methodology, Defendant would have to pay the
    8
    downstream lender restitution of $400,000 ($500,000 less $100,000 from the foreclosure
    sale), although its loss was only $100,000. That would create an unlawful windfall for
    the downstream lender.
    Other circuits agree with this analysis. The Ninth Circuit has noted that
    “[b]ecause the value of [the] loan is not necessarily its unpaid principal balance, but may
    vary with the value of the collateral, the credit rating of the borrower, market conditions,
    or other factors, the loan purchaser may have purchased the loan for less than its unpaid
    principal balance.” United States v. Yeung, 
    672 F.3d 594
    , 602 (9th Cir. 2012), overruled
    in part on other grounds by Robers, 
    134 S. Ct. 1854
    . As a result, it said, “To calculate a
    victim’s restitution award using the outstanding principal balance of the loan, if the
    victim only paid a fraction of that amount to obtain the loan on the secondary market,
    would cause the victim to receive an amount exceeding its actual losses.” Id.; see United
    States v. Chaika, 
    695 F.3d 741
    , 748 (8th Cir. 2012) (“The ultimate foreclosure sale price
    is irrelevant to an initial lender who sold the loan, while the purchasing secondary lender
    may not be a victim, and if it is, actual loss will turn on its purchase price in the
    secondary market, whether it remained on the loan all the way to foreclosure, and perhaps
    other factors.”); United States v. v. Beacham, 
    774 F.3d 267
    , 278‒79 (5th Cir. 2014)
    (following Chaika and Yeung).
    The government contends that this point was not raised below, but we disagree.
    Defendant’s sentencing memorandum argued that the restitution calculations in the
    probation office’s presentence report were flawed because they did not examine the
    9
    purchase prices paid by downstream holders of the mortgage notes, and he reminded the
    district court of the issue at the sentencing hearing. Once that objection to the
    government’s methodology was clear, Defendant was not also required to object to the
    evidence offered in reliance on the challenged methodology. He gave ample notice that
    he objected to a restitution calculation that did not identify the specific losses of
    individual noteholders in the chain of title of a mortgage note. When the government did
    not put on such evidence, it took the risk that we would agree with Defendant’s legal
    argument.
    We recognize that the government may be able to explain (on remand) why the
    restitution calculation is “a reasonable estimate of the loss,” United States v. James, 
    564 F.3d 1237
    , 1248 (10th Cir. 2009) (internal quotation marks omitted); but it did not
    provide such an explanation in the district court and, equally important, the court itself
    made no finding on the point. See Ferdman, 779 F.3d at 1133 (district courts may not
    “dispense with the necessity of proof as mandated by the MVRA and simply ‘rubber
    stamp’ a victim’s claim of loss based upon a measure of value unsupported by the
    evidence,” and restitution awards may not be based on “[s]peculation and rough justice”
    (internal quotation marks omitted)). The presentence report stated that the victims it
    identified had not responded to correspondence from the probation office. The FBI agent
    who testified at the sentencing hearing about the losses caused by Defendant said that he
    did not know whether original lenders had suffered losses on any loans sold, and that he
    had no information about the amounts paid by downstream noteholders to purchase the
    10
    loans. He never testified that the information was not available and, for all the record
    reveals, he never asked for it. Neither the failure of a victim to respond to a request for
    evidence of actual loss nor the government’s unexplained failure to obtain the necessary
    proof suffices to justify a restitution award. There is no public interest served by
    requiring that restitution be paid to an alleged victim who declines to cooperate in
    providing the evidence necessary to establish its loss.
    In short, although the impact of sales of mortgage notes to downstream lenders is
    generally irrelevant to the total-loss calculation under USSG § 2B1.1, it is highly relevant
    in calculating restitution under the MVRA. See United States v. James, 
    592 F.3d 1109
    ,
    1116 n.6 (10th Cir. 2010) (“the calculation of loss for sentencing purposes does not
    necessarily establish loss for the purpose of awarding restitution under the MVRA”). We
    remand with instructions that the district court vacate its restitution order and redetermine
    the amount of actual loss to identified downstream-noteholder victims. Should the court
    find the existing record to be insufficient to permit a proper calculation of a victim’s
    actual loss, the court may “(1) ask the Government to submit additional evidence,
    (2) hold an evidentiary hearing, or (3) decline to order restitution.” Ferdman, 779 F.3d at
    1133.
    III.    CONCLUSION
    We AFFIRM the district court’s calculation of loss under USSG § 2B1.1 and
    REMAND with instructions that the court VACATE its restitution order for
    redetermination of the amount of actual loss to apparent victims.
    11