UNPUBLISHED
UNITED STATES COURT OF APPEALS
FOR THE FOURTH CIRCUIT
No. 17-7613
UNITED STATES OF AMERICA,
Plaintiff – Appellee,
v.
ROBERT M. FELDMAN,
Defendant – Appellant.
Appeal from the United States District Court for the District of Maryland, at Baltimore.
James K. Bredar, Chief District Judge. (1:13-cr-00457-JKB-1; 1:17-cv-02412-JKB)
Argued: October 29, 2019 Decided: December 3, 2019
Before MOTZ, AGEE and KEENAN, Circuit Judges.
Reversed, vacated, and remanded by unpublished per curiam opinion.
ARGUED: Brian Patrick Quinn, O’MELVENY & MYERS LLP, Washington, D.C., for
Appellant. Martin Joseph Clarke, OFFICE OF THE UNITED STATES ATTORNEY,
Baltimore, Maryland, for Appellee. ON BRIEF: Jonathan D. Hacker, Patrick Jones,
O’MELVENY & MYERS LLP, Washington, D.C., for Appellant. Robert K. Hur, United
States Attorney, Isabelle M. Jensen, Student Law Clerk, OFFICE OF THE UNITED
STATES ATTORNEY, Baltimore, Maryland, for Appellee.
__________________
Unpublished opinions are not binding precedent in this circuit.
PER CURIAM:
Federal prisoner Robert M. Feldman appeals from the district court’s judgment
denying his 28 U.S.C. § 2255 motion to vacate his sentence. He argued that his trial counsel
provided ineffective assistance by failing to object to an incorrect Sentencing Guidelines
calculation. Applying the two-prong analysis set out in Strickland v. Washington, 466 U.S.
668 (1984), the district court determined that counsel’s performance was deficient, but that
Feldman failed to demonstrate prejudice. For the reasons set forth below, we hold that the
district court erred in concluding Feldman failed to show prejudice. Because Feldman is
entitled to § 2255 relief, we reverse the district court’s judgment, vacate Feldman’s
sentence and remand for resentencing.
I.
In September 2013, Feldman pleaded guilty to one count of conspiracy to commit
wire fraud, in violation of 18 U.S.C. § 1349, for his role in a complex and large-scale Ponzi-
like scheme “to defraud investors and lenders in medical accounts receivable and to obtain
money and property from such investors and lenders[.]” J.A. 16. Feldman’s written plea
agreement included several provisions relevant to his Guidelines calculation, including that
“[p]ursuant to U.S.S.G. § 2B1.1(b)(1)(N), the base offense level is increased by twenty-six
(26) levels because the loss was more than $100,000,000 but less than $200,000,000[.]”
J.A. 34. Elsewhere, it noted Feldman’s agreement to the entry of a restitution order “for
the full amount of the victims’ losses of approximately $148,251,859.” J.A. 35.
Although Feldman pleaded guilty in 2013, he was not sentenced until August 2016.
In the intervening period, § 2B1.1(b)(1)(N) was amended to increase the amount of
2
financial loss corresponding to different offense levels. 1 In relevant part, the 2016
Guideline amendments increased the base offense level to 26 when the loss involved more
than $150,000,000 and less than $250,000,000 rather than for losses of more than
$100,000,000 and less than $200,000,000, as was the applicable range at the time of
Feldman’s plea. The Government noted this Guidelines change in a footnote of its
sentencing memorandum, but claimed a base offense level of 26 remained proper. Feldman
did not challenge that assertion in the sentencing process.
Consequently, at the sentencing hearing, the district court applied
§ 2B1.1(b)(1)(N)’s 26-level enhancement “to reflect [a] loss in excess of $150 million.”
J.A. 97. Other adjustments to the offense level resulted in the court using an overall offense
level of 23 and a criminal history category of I, which yielded a Guidelines range of 46 to
57 months’ imprisonment.
Following the parties’ 18 U.S.C. § 3553(a) arguments as to a proper sentence, the
district court sentenced Feldman to a term of imprisonment at the bottom of his Guidelines
range: 46 months. In so doing, the district court discussed the relevance of the § 3553(a)
factors, including the Sentencing Guidelines. Of particular note, it stated that it sentenced
Feldman “today with that [Guidelines] calculation very much in mind.” J.A. 134. The court
explained that although it believed the Guidelines’ fraud loss table “is often given too much
significance in its influence on the sentence that ultimately is imposed, . . . in this case,
1
Ordinarily, a defendant is sentenced using the version of the Guidelines in effect
at the time of sentencing. 18 U.S.C. § 3553(a)(4)(A)(ii). That principle is applicable to
Feldman’s sentencing.
3
when the size of the fraud was as enormous as it is, there is no getting around the fact that
just the big loss is itself the driving factor that has to be considered in sentencing[.]” J.A.
135. Feldman did not file a direct appeal.
In August 2017, Feldman filed a pro se § 2255 motion alleging that he was entitled
to resentencing because trial counsel violated his right to effective assistance of counsel.
In sum, he argued that counsel provided deficient performance by failing to object to the
district court’s use of the 26-level enhancement because his plea agreement stipulated a
loss amount of $148,251,859 and, under the 2016 version of § 2B1.1(b)(1)(N) applicable
at sentencing, that loss amount corresponded to a 24-level enhancement. Feldman asserted
this error led the district court to incorrectly calculate his Guidelines range, which, in turn,
created a reasonable probability that he was sentenced to a longer term of imprisonment
than if his Guidelines range had been properly calculated.
The district court determined that Feldman had demonstrated deficient performance
based on counsel’s failure to object to the court’s use of a 26-level enhancement. But the
court concluded Feldman failed to show prejudice arising from that error because the
advisory Guidelines range was just one of the § 3553(a) factors used in selecting his
sentence, it had thoroughly explained why the § 3553(a) factors supported the sentence,
and “the scale and magnitude of the fraud” well supported a sentence of 46 months’
imprisonment. J.A. 58. The court stated that although Feldman raised a “remote
possibility” that he would have received a lower sentence, that did not satisfy his burden
of showing a “reasonable” possibility of a lower sentence. J.A. 58.
4
We granted Feldman a certificate of appealability under 28 U.S.C. § 2253(c)(1)(B)
on the following issue:
Whether the district court erred in concluding that Feldman was not
prejudiced by counsel’s failure to object to the Sentencing Guidelines
enhancement for the loss amount associated with the fraud conspiracy.
6/1/2018 Order, ECF No. 13. We have jurisdiction under 28 U.S.C. § 2253(a).
II.
A.
We review de novo the district court’s decision to deny Feldman’s § 2255 motion.
United States v. Poindexter, 492 F.3d 263, 267 (4th Cir. 2007).
B.
Strickland’s two-prong test governs Feldman’s ineffective assistance of counsel
claim. It requires him to show “(1) that his lawyer afforded him defective representation;
and (2) that there exists ‘a reasonable probability that, but for counsel’s unprofessional
errors, the result of the proceeding would have been different.’” United States v. Stitt, 552
F.3d 345, 350 (4th Cir. 2008) (quoting Strickland, 466 U.S. at 694).
The first prong is not at issue in this appeal. The district court concluded that
Feldman satisfied his burden of showing defective representation at sentencing. The
Government does not contest that determination. Thus, only the second prong of the
Strickland test is at issue: whether Feldman satisfied his burden of showing prejudice
resulting from trial counsel’s failure to object to the Sentencing Guidelines calculation.
The Supreme Court has described prejudice in the context of counsel’s deficient
performance at sentencing to mean the existence of a reasonable probability that the
5
defendant would have been sentenced differently if the error had not occurred. See Sears
v. Upton, 561 U.S. 945, 956 (2010). “A reasonable probability is a probability sufficient to
undermine confidence in the outcome.” Strickland, 466 U.S. at 694. Importantly, this
standard “is not the same as, and should not be confused with, a requirement that a
defendant prove by a preponderance of the evidence that but for error[,] things would have
been different.” United States v. Dominguez Benitez, 542 U.S. 74, 83 n.9 (2004); see
Strickland, 466 U.S. at 693 (rejecting as too demanding a standard for prejudice in this
context that would require a defendant to show “that counsel’s deficient conduct more
likely than not altered the outcome in the case”).
In the context of counsel’s deficient performance that results in the Guidelines being
improperly calculated, the nature of the error that occurred, without more, “can, and most
often will, be sufficient” to satisfy the defendant’s burden as to prejudice. Molina-Martinez
v. United States, 136 S. Ct. 1338, 1345 (2016). This conclusion flows from the Supreme
Court’s holding in that case, which considered what evidence is necessary to demonstrate
prejudice in the context of plain-error review on direct appeal. 2 In Molina-Martinez, the
2
To be clear, plain-error review and Strickland ineffective assistance claims are not
identical. We have previously recounted their differences and explained why a defendant
may be entitled to relief under one inquiry, but not the other. E.g., United States v.
Carthorne, 878 F.3d 458, 465–66 (4th Cir. 2017) (detailing the different standards and
explaining why they are not identical). But the two inquiries overlap in at least one respect
because “in both cases prejudice turns on whether substantial rights of the defendant were
affected,” United States v. Strickland, 245 F.3d 368, 380 (4th Cir. 2001), meaning that in
both instances the Court must consider whether a reasonable probability exists that the
result of the proceeding would have been different absent the complained-of error. See
Molina-Martinez, 136 S. Ct. at 1345 (using the “reasonable probability” language in the
plain-error context); United States v. Rangel, 781 F.3d 736, 745–46 (4th Cir. 2015)
6
Supreme Court reiterated that even though the Guidelines are advisory, they continue to
play a central role in federal sentencing. The Supreme Court held that “most often” errors
in calculating the Guidelines range will be prejudicial to a defendant. Id. at 1345.
Specifically, the Court stated: “[w]hen a defendant is sentenced under an incorrect
Guidelines range—whether or not the defendant’s ultimate sentence falls within the correct
range—the error itself can, and most often will, be sufficient to show a reasonable
probability of a different outcome absent the error.” Id.
In reaching its holding, the Supreme Court relied on “the essential framework the
Guidelines establish for sentencing proceedings,” id., observing that they serve “not only
[as] the starting point[,] . . . but also the lodestar,” id. at 1346; see also id. at 1345
(explaining that any “error related to the Guidelines can be particularly serious”). Because
of this central role that the Guidelines have in federal sentencing, the Supreme Court
rejected the Fifth Circuit’s requirement that a defendant must come forward with additional
evidence of harm—beyond an erroneously calculated Guidelines range—to prove that the
error affected his substantial rights when his actual sentence fell within both the improperly
calculated Guidelines range and the range that the court should have used. Id. at 1345.
Instead, the Supreme Court held that no “further showing of prejudice beyond the fact that
the erroneous, and higher, Guidelines range set the wrong framework for the sentencing
proceedings” is necessary for most defendants to demonstrate prejudice. Id.
(discussing the similarity between plain-error review’s effects on substantial rights analysis
and Strickland’s prejudice inquiry).
7
After explaining these principles, the Supreme Court observed that in the case
before it, after the Guidelines were improperly calculated, the parties’ § 3553(a) arguments
centered on where the defendant should be sentenced within that Guidelines range, with
the Government arguing in favor of the top of the Guidelines range. The Supreme Court
concluded that the district court’s selection of a sentence at the bottom of the Guidelines
range, despite the Government’s argument, indicated that it had intentionally selected a
sentence at what it believed to be the bottom of the proper range. Id. at 1347–48. Further,
the Supreme Court noted that the district court “said nothing to suggest that it would have
imposed [that] sentence regardless of the Guidelines range” or that its sentence was based
“on factors independent of the Guidelines,” either of which may have been sufficient to
demonstrate that the Guidelines calculation error did not prejudice the defendant. Id. at
1347–1348. For all these reasons, the Supreme Court held that Molina-Martinez had
demonstrated a “reasonable probability that the district court would have imposed a
different sentence under the correct range.” Id. at 1349.
Having reviewed the record related to Feldman’s sentencing, we conclude that his
case falls into the typical sentencing proceeding described in Molina-Martinez, in which
an error related to a defendant’s Guidelines calculation is sufficient to demonstrate a
reasonable probability that he received a higher sentence as a result of that error than he
would have received otherwise. As an initial matter, Feldman has shown that the
Guidelines range used during his sentence was higher than it should have been absent
counsel’s deficient performance in failing to object to the Guidelines calculation: 46 to 57
8
months’ imprisonment instead of 37 to 46 months’ imprisonment. 3 Under Molina-
Martinez, “[a]bsent unusual circumstances, [a defendant] will not be required to show
more” than this fact to satisfy his burden of demonstrating prejudice. 136 S. Ct. at 1347.
Additional aspects of the record further support the showing of prejudice because
they demonstrate the specific role the Guidelines played in Feldman’s sentencing. For
example, in discussing the appropriate sentence under § 3553(a), the Government argued
in support of a sentence at the bottom of the Guidelines range and Feldman argued for a
downward variance. As Molina-Martinez noted was usually the case, the appropriateness
of the Guidelines was the focal point of the parties’ arguments concerning the § 3553(a)
factors. Id. at 1345–47. Then, after these arguments, the district court sentenced Feldman
to the bottom-of-the-Guidelines term of imprisonment of 46 months. The district court’s
selection of that sentence—despite Feldman’s arguments for a downward variance—and
the Government’s argument for the bottom-of-the-Guidelines sentence “conspicuous[ly]”
demonstrate that this was “the lowest sentence within what the District Court believed to
be the applicable range,” just like the sentence at issue in Molina-Martinez. Id. at 1347.
This is not a case where the district court imposed a variant sentence upward or downward
based on an explanation of why the Guidelines range did not adequately reflect the sentence
needed to carry out the § 3553(a) objectives. What’s more, although the district court
3
The district court found that had Feldman’s counsel objected to the incorrect base
offense level, the court would have used the offense level corresponding to fraud involving
more than $100,000,000 and less than $150,000,000. See J.A. 57. A base offense level of
24 would have resulted in an adjusted level of 21, which when combined with a criminal
history category of I, yielded a Guidelines range of 37 to 46 months’ imprisonment.
9
discussed all of the § 3553(a) factors, it reiterated that this sentence was imposed only after
having “carefully calculated” Feldman’s Guidelines range and with the Guidelines
“calculation very much in mind.” J.A. 134. Indeed, the district court specifically observed
that the Guidelines’ fraud loss table was particularly useful in determining an appropriate
sentence “when the size of the fraud [is] as enormous as” in Feldman’s case because “the
factor that is most important in deciding what the sentence should be is the sheer magnitude
of the crime and the size of the fraud that was perpetrated.” J.A. 134–35.
These comments by the district court reinforce the particular role the Guidelines
played in its sentencing decision. And, notably, they serve as additional evidence in support
of a finding of prejudice in this case that was not present in Molina-Martinez. This is so
because they expressly connect the district court’s discussion of all of the § 3553(a) factors
not just to the appropriateness of the Guidelines range as one of the factors, but specifically
to the aspect of the Guidelines calculation that was erroneous (the fraud loss amount).
Lastly, nothing in the district court’s explanation of its sentence “suggest[s] that it would
have imposed [the same] sentence regardless of the Guidelines range.” Molina-Martinez,
136 S. Ct. at 1348. At best, “the record is silent as to what the district court might have
done had it considered the correct Guidelines range.” Id. at 1347. 4
4
The Government points out that because the judge who denied Feldman’s § 2255
motion was the same judge who originally sentenced Feldman, the court’s comments
during the § 2255 proceedings concluding that the error would have made no difference to
the sentence serve as further proof that Feldman was not prejudiced by counsel’s deficient
performance. That argument focuses on the incorrect timeframe. Our prejudice inquiry
looks to objective proof in the record at the time of the defendant’s sentencing, not to
statements of this sort made during the later § 2255 proceeding.
10
Each of these factors demonstrates the centrality of the Guidelines range to the
specific sentence Feldman received and that the improperly calculated Guidelines range
clearly influenced the district court’s consideration of the § 3553(a) factors. All told,
nothing in this record moves Feldman’s case away from the usual case where a Guidelines
calculation error “can, and most often will” mean that the defendant has satisfied his burden
of showing a “reasonable probability that the district court would have imposed a different
sentence under the correct range.” Id. at 1345, 1349. As such, we conclude that Feldman
has satisfied his burden of demonstrating prejudice under Strickland.
III.
For the reasons set out above, the district court erred in denying Feldman’s § 2255
motion. Therefore, we reverse the judgment of the district court, vacate Feldman’s
sentence, and remand the case for Feldman to be resentenced. 5
REVERSED, VACATED,
AND REMANDED
5 In doing so, we offer no view as to the proper sentence, as that is a function for
the district court in the first instance.
11