United States Court of Appeals For the First Circuit
No. 23-1087
UNITED STATES,
Appellee,
v.
MANISH KUMAR,
Defendant, Appellant.
APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MASSACHUSETTS
[Hon. Mark L. Wolf, U.S. District Judge]
Before
Kayatta, Howard, and Rikelman, Circuit Judges.
Edward Crane for appellant.
Donald C. Lockhart, Assistant United States Attorney, with whom Joshua S. Levy, Acting United States Attorney, was on brief, for appellee.
August 12, 2024 HOWARD, Circuit Judge. Manish Kumar brings a procedural
challenge to an 87-month sentence imposed after he pled guilty to
conspiring to smuggle misbranded prescription drugs and controlled
substances into the United States and making false statements. He
argues that the sentencing court erred in (1) applying a particular
fraud cross-reference in the Sentencing Guidelines and (2)
accepting the presentence investigation report (PSR) estimate as
to the loss amount involved in his offense. We affirm.
I.
A.
We briefly summarize the factual background of Kumar's
case, drawing on the change-of-plea colloquy, the revised PSR, and
the transcript of the sentencing hearing. See United States v.
Ihenacho, 716 F.3d 266, 269 (1st Cir. 2013).
From at least March 2015 until August 2019, Kumar
participated in an operation selling generic versions of
prescription drugs and controlled substances to customers in the
United States. Kumar, who is an Indian national, was one of at
least four partners in Mihu -- a company based in New Delhi that
functioned as the parent corporation of several subsidiaries that
assisted in the venture. The pills involved were primarily
generic versions of Viagra and Cialis, but many were also
controlled substances such as Adderall and tramadol (an opioid).
None were produced in formulations approved by the FDA or sold
- 2 - with proper prescriptions. Their importation thus violated the
Food, Drug, and Cosmetic Act and the Controlled Substances Act.
See 21 U.S.C. § 331(a) (prohibiting "[t]he introduction or
delivery for introduction into interstate commerce of any . . .
drug . . . that is adulterated or misbranded); 1 21 U.S.C.
§ 841(a)(1) ("[I]t shall be unlawful for any person knowingly or
intentionally . . . to manufacture, distribute, or dispense, or
possess with intent to manufacture, distribute, or dispense, a
controlled substance.").
Kumar oversaw call centers in India where
representatives targeted customers in the United States as part of
this operation. In those sales calls, the representatives would
make a variety of false statements to potential purchasers,
including that the representatives were located in the United
States, that they were calling from a pharmacy, that no
prescriptions were needed for the drugs, and that the drugs were
approved by the FDA. Each call center had a manager who reported
directly to Kumar, providing him with copies of drug orders and
audio recordings of sales calls. Kumar gave direction to these
managers about strategies for the calls and also played a role in
A prescription drug is "misbranded" if it is "dispensed" 1
without "a written prescription of a practitioner licensed by law to administer such drug." 21 U.S.C. § 353(b)(1).
- 3 - shipping the pills into the United States, taking various steps to
avoid detection by U.S. authorities and financial institutions.
In August 2019, Kumar was arrested at JFK Airport on
federal identity theft charges pending in Rhode Island. He pled
guilty to those charges, which were not directly related to this
case. After serving several months in prison, Kumar was briefly
released to immigration custody, where he was arrested in May 2021
on the charges in this case. The indictment, which was filed in
Massachusetts, contained three counts: (1) conspiracy under 18
U.S.C. § 371 to smuggle misbranded drugs and controlled substances
into the United States in violation of 18 U.S.C. § 545, 21 U.S.C.
§ 331(a), and 21 U.S.C. § 841(a)(1); (2) conspiracy to distribute
controlled substances in violation of 21 U.S.C. § 846; and (3)
false statements2 in violation of 18 U.S.C. § 1001(a)(2). Kumar
pled guilty to all three counts without a plea agreement in October
2022.
B.
Because Kumar challenges only his sentence on appeal, we
recount in some detail the post-guilty-plea stages of the
proceedings, although we save a more nuanced discussion of the
Sentencing Guidelines for later.
2 During a period in which Kumar was cooperating with federal authorities after his arrest on the Rhode Island charges, he falsely told investigators that he did not sell controlled substances.
- 4 - The Probation Officer filed an initial PSR for Kumar on
December 19, 2022. In calculating Kumar's base offense level, the
PSR applied the fraud cross-reference in U.S.S.G. §2N2.1,3 which
directs to §2B1.1. The base offense level was then adjusted
upward, based in large part on applying the loss table in
§2B1.1(b)(1) to the estimated amount that consumers paid for the
pills that Kumar had conspired to smuggle -- i.e., his revenue.
The initial PSR estimated that this amount was approximately
$400,000. But it also cautioned that Kumar's base offense level
could still be increased pending the government seeking further
clarification from its analysts about their estimates.
Kumar and the government subsequently exchanged
sentencing memoranda and replies. In its memorandum, the
government described Kumar's participation in the drug scheme,
which it alleged generated upward of $3.5 million in revenue. To
further illustrate Kumar's business practices, the government
provided multiple spreadsheets (together spanning close to 100
pages) that Kumar had maintained to track the operation's drug
shipments.4
3 All citations to the Sentencing Guidelines are to the 2021 Manual that was in effect at the time of Kumar's sentencing. 4 The government attached additional spreadsheets to its reply to Kumar's sentencing memorandum.
- 5 - The government also described how it reached its revenue
estimate. It acknowledged that such estimation was difficult due
to the fact that Kumar's sales spreadsheets contained limited
information on the prices that customers had paid for the pills,
but it explained how it settled on a $1-per-pill estimate for the
most commonly sold drugs after reviewing Kumar's data as well as
contemporaneous internet prices. The sentencing memorandum
additionally noted that the government's analysis of Kumar's sales
data was not yet complete. In its objections to the initial PSR,
the government expounded on that analysis by describing how it had
used the Wayback Machine (an internet archive) to research
historical prices for India-sourced pharmaceuticals during the
period when Kumar was operating. To demonstrate its work, the
government provided an extensive sample of that research.
The Probation Officer thereafter filed a revised PSR.
Adopting the government's updated estimate, the revised PSR
contained a significantly higher loss amount: approximately $3.8
million, up from approximately $400,000 in the initial PSR. The
updated estimate was summarized in a chart that detailed the number
of pills that Kumar had conspired to sell, the estimated price per
pill, and the total revenue for each year between 2015 and 2019.
The increase in the loss amount resulted in the recommended
Guidelines range increasing from 46–57 months in the initial PSR
to 87–108 months in the revised PSR.
- 6 - At his January 2023 sentencing hearing, Kumar raised two
objections: (1) the fraud cross-reference in U.S.S.G. §2N2.1 did
not apply to his case; and (2) even if it did, there was
insufficient evidence to support the government's estimate of loss
amount. The sentencing court ruled against him on both
objections. With regard to Kumar's first objection, the court
noted that the indictment alleged fraudulent conduct in the false
statements made by call center representatives to customers. In
response to the second objection, the sentencing judge determined
that the government's estimate was "a very thorough, careful and
conservative calculation of the loss."
The government then recommended a sentence of 87 months,
and Kumar recommended a time-served sentence of 20 months. After
hearing an apology from Kumar that it deemed insincere, the court
sentenced Kumar to 87 months of incarceration followed by 36 months
of supervised release.
Kumar timely filed this appeal, in which he generally
renews his objections made at sentencing.
II.
Kumar and the government agree that applying the
Guidelines to Kumar's case takes one at least as far as U.S.S.G.
§2N2.1. But once there, Kumar challenges the sentencing court's
decision to invoke a cross-reference in the section, which states:
"If the offense involved fraud, apply §2B1.1." U.S.S.G.
- 7 - §2N2.1(c)(1). We work on a fresh slate in considering his
challenge, because "'[a]rguments that the sentencing court erred
in interpreting or applying the guidelines' are reviewed de novo."
United States v. Ramirez-Frechel, 23 F.4th 69, 77 (1st Cir. 2022)
(quoting United States v. Leahy, 668 F.3d 18, 21 (1st Cir. 2012)).5
The Guidelines define "offense" as "the offense of
conviction and all relevant conduct." U.S.S.G. §1B1.1 cmt.
n.1(I). Thus, the cross-reference in §2N2.1(c)(1) should apply
if Kumar's offense of conviction or any "relevant conduct" involved
fraud. See United States v. Castillo, 981 F.3d 94, 100–01 (1st
Cir. 2020) (outlining this approach for a different
cross-reference). Most salient here, the Guidelines define
"relevant conduct" in a conspiracy to include "all acts and
omissions of others that were -- (i) within the scope of the
jointly undertaken criminal activity, (ii) in furtherance of that
criminal activity, and (iii) reasonably foreseeable in connection
with that criminal activity." U.S.S.G. §1B1.3(a)(1)(B).
Applying that definition, we hold that "relevant
conduct" involved fraud in Kumar's case. Kumar does not deny that
5The government characterizes the sentencing court's determination that Kumar's offense "involved fraud" as a factual finding that should be reviewed for clear error. In support of its argument, the government cites United States v. Dyer, but we note that the sentencing court's application of the cross-reference at issue in that case was "heavily fact-dependent." 589 F.3d 520, 530 (1st Cir. 2009). The facts of Kumar's case, by contrast, are straightforward.
- 8 - he oversaw call centers in India where representatives targeted
customers in the United States. Audio recordings demonstrated
that those representatives made a variety of false statements to
customers. "Fraud" is "[a] knowing misrepresentation or knowing
concealment of a material fact made to induce another to act to
his or her detriment." Fraud, Black's Law Dictionary (11th ed.
2019). The false statements made by the call center
representatives assuredly qualify. Furthermore, Kumar gave
directions to his call center managers about customer contacts,
customer service, and the tone of conversations. And there is no
indication that the representatives ever obtained from the call
center's customers information about a valid prescription or a
prescribing physician.
These facts taken together satisfy the Guidelines' three
elements of "relevant conduct" in the case of a conspiracy.
Although the Guidelines caution that "the scope of the 'jointly
undertaken criminal activity' is not necessarily the same as the
scope of the entire conspiracy," U.S.S.G. §1B1.3 cmt. n.3(B), the
fraudulent statements of the call center representatives here were
well "within the scope" of Kumar's activity because they were made
under his management. Additionally, the fraudulent statements
were made "in furtherance" of the criminal activity because they
were intended to induce customers into ordering the drugs that
Kumar was conspiring to sell. Finally, they were "reasonably
- 9 - foreseeable" because Kumar was responsible for directing call
center operations.
Kumar attempts to undermine the conclusion that relevant
conduct involved fraud by focusing on the discrete act of
importation -- i.e., moving the pills across the border of the
United States. He argues that the fraudulent statements were not
made during, or in preparation for, the unlawful importation of
the drugs and therefore were not "relevant conduct." Setting
aside the questionable proposition that the fraudulent statements
were not made "in preparation for" the unlawful importation,
Kumar's argument fails because it overlooks the fact that he did
not plead guilty to smuggling misbranded drugs and controlled
substances into the United States -- he pled guilty to conspiring
to smuggle such drugs. That conspiracy lasted from at least March
2015 through August 2019. Thus, even if Kumar is correct that the
fraudulent statements made by representatives at the call centers
that he oversaw were not directly connected to the importation of
the pills, they were sufficiently connected to the conspiracy that
Kumar engaged in to qualify as relevant conduct. The sentencing
court therefore did not err in applying the cross-reference in
§2N2.1(c)(1).
- 10 - III.
Kumar's second argument concerns §2B1.1 of the
Guidelines, which the sentencing court turned to after correctly
applying the cross-reference discussed above. By way of
background, we note that §2B1.1 is considered to be the most
general fraud guideline. See generally Roger W. Haines, Jr. et
al., Federal Sentencing Guidelines Handbook: Text and Analysis
396–543 (2022–2023 ed.). Although it often arises in cases
involving mortgage fraud or tax fraud, see, e.g., United States v.
Jiménez, 946 F.3d 8, 12–13 (1st Cir. 2019); United States v. Akoto,
61 F.4th 36, 45 (1st Cir. 2023), the guideline also comes into
play when a defendant has sold misbranded drugs, see Ihenacho, 716
F.3d at 276.
Section 2B1.1(b)(1) contains a "loss table," which
directs a sentencing court to increase a defendant's offense level
based on the amount of loss attributable to the defendant's fraud.
In this case, that loss amount is essentially equal to the value
paid by customers for the pills that Kumar conspired to import.
U.S.S.G. §2B1.1 cmt. n.3(F)(v)–(vi). The government bears the
burden of proving the loss amount by a preponderance of the
evidence. United States v. Flete-Garcia, 925 F.3d 17, 28 (1st
Cir. 2019). "The sentencing court has considerable discretion in
determining what evidence should be regarded as reliable in making
- 11 - findings as to the amount of loss." Id. (citing United States v.
Sklar, 920 F.2d 107, 110 (1st Cir. 1990)). In making its finding,
"[t]he district court 'may rely on the [PSR], affidavits,
documentary exhibits, and submissions of counsel." United States
v. Curran, 525 F.3d 74, 78 (1st Cir. 2008) (quoting United States
v. Ranney, 298 F.3d 74, 81 (1st Cir. 2002)). And a court's "loss
calculation need not be precise: the sentencing court need only
make a reasonable estimate of the range of loss." Flete-Garcia,
925 F.3d at 28 (citing Curran, 525 F.3d at 78).
With that background in mind, we turn to the specifics
of Kumar's case. Basic arithmetic tells us that, in order to
calculate Kumar's loss amount, the sentencing court needed to
multiply: (1) the number of pills sold, by (2) the price charged
per pill. On each of these elements, the court adopted at
sentencing the estimates in the revised PSR that had been supplied
by the government and were summarized in a chart in the PSR. To
produce those estimates, the government explained that it started
with sales spreadsheets and emails that were saved on Kumar's
laptop detailing drug shipments to the United States. Those
sources contained a limited amount of information about the prices
for which the pills were sold, so the government supplemented those
sources with information from recorded sales calls located in
Kumar's email account, as well as research on the historical prices
of pharmaceuticals produced in India and sold online, using the
- 12 - Wayback Machine. This led the government to use an estimate of
$1 per pill for the drugs that Kumar sold most often. In the end,
the government estimated that Kumar conspired to sell 3,859,772
pills between 2015 and 2019. With an estimated price of $1 for
the vast majority of those pills, the government estimated (and
the sentencing court adopted) a loss amount of $3,839,144.55.
Kumar takes issue with the sentencing court's
calculation of loss amount. In particular, he challenges the
court's reliance on the summary chart in the revised PSR. Kumar
claims that there are three issues with the chart: namely that (1)
it is backed by insufficient evidence on the price of the pills
that Kumar sold; (2) it lists more pills than are included in the
spreadsheets that were attached as exhibits to the government's
sentencing memorandum and its reply to Kumar's sentencing
memorandum; and (3) it does not list any specific drug types.
Preliminarily, Kumar and the government tussle over the
standard of review that we should apply in considering these
claims. A sentencing court's findings as to loss amount are
typically subject to clear error review. See Akoto, 61 F.4th at
45. The government, however, contends that Kumar's arguments with
respect to the loss amount were not properly raised below and
should thus be subject to plain error review. Ultimately, we need
- 13 - not resolve this dispute because Kumar's arguments fail even under
the clear error standard that he seeks to have applied.
To begin with, the sentencing court did not clearly err
in its estimation of the amount charged per pill.
The government was transparent that it had limited
information on the prices of pills that Kumar conspired to sell
and about the resulting need to estimate. At the sentencing
hearing, Kumar pointed to the fact that some of the government's
data showed pills being sold for less than $1 per pill (the
estimate used for most of the pills), but the sentencing court
correctly noted that the data also included pills being sold for
more than that price. In its objections to the initial PSR, the
government also provided an extensive sample of its research on
the historical prices of pharmaceuticals produced in India and
sold online. Kumar never challenged that research, and we cannot
say that the sentencing court clearly erred in adopting it.6
C.
Neither did the sentencing court clearly err in
estimating the quantity of pills in its loss amount calculation.
Kumar directs our attention to the fact that the pill quantities
6 In any event, Kumar's argument on appeal as to this issue is perfunctory to the point of being waived. See United States v. Zannino, 895 F.2d 1, 17 (1st Cir. 1990).
- 14 - in the spreadsheets attached by the government to its sentencing
memorandum and reply add up to slightly more than 1 million,
whereas the chart in the revised PSR shows that Kumar conspired to
sell more than 3.8 million pills. But the government never
claimed -- and the sentencing court never held -- that the
spreadsheets that the government attached to its sentencing
memoranda represented all of the data it possessed linking Kumar
to drug sales. In fact, the government asserted the opposite.
Its sentencing memorandum made clear that the government's
assertions of the quantity of pills attributable to Kumar were
just estimates, that its analysis of Kumar's sales data was not
yet complete, and that the attached exhibits were merely meant to
"further illustrate Kumar's business practices." At sentencing,
the government offered to provide the court with more spreadsheets
and also described how it performed its loss amount calculations.
The court declined that offer, a decision that Kumar did not
challenge at the time.
Kumar cites no case holding that a sentencing court can
rely on a summary chart in a PSR only if all of the underlying
data is included in the exhibits that are attached to the
government's sentencing memoranda. Of course, "[t]he evidentiary
requirements that obtain at sentencing are considerably less
rigorous than those that obtain in criminal trials." United
States v. Cintrón-Echautegui, 604 F.3d 1, 6 (1st Cir. 2010). But
- 15 - even under the more demanding Federal Rules of Evidence, "[t]he
proponent may use a summary, chart, or calculation offered to prove
the content of voluminous writings . . . that cannot be
conveniently examined in court" so long as they "make the originals
or duplicates available for examination or copying, or both, by
other parties at a reasonable time and place." Fed. R. Evid.
1006. The underlying records must be admissible but need not be
introduced into evidence. See United States v. Milkiewicz, 470
F.3d 390, 396 (1st Cir. 2006). Here, Kumar does not contend that
any evidence underlying the chart would have been inadmissible at
sentencing, and at oral argument before us, his counsel
forthrightly clarified that Kumar does not allege that he was
denied access to the underlying data.7 Rather, Kumar's argument
is that the only acceptable way the government could have
established that the chart in the revised PSR was a reasonable
In a Rule 28(j) letter filed following oral argument, Kumar 7
cites two out-of-circuit decisions that he claims stand for the proposition that, under the here-inapplicable Federal Rule of Evidence 1006, "[t]he accuracy of the summary must be established by evidence that was already introduced into the record." See United States v. Bishop, 264 F.3d 535, 547 (5th Cir. 2001); United States v. Wainwright, 351 F.3d 816, 820–21 (8th Cir. 2023). However, Milkiewicz makes clear that this court does not require any of the underlying evidence to come in before a summary chart is admitted; our rule instead is that the underlying documents must be admissible and made available to the other party. See 470 F.3d at 396–97. In any event, we conclude that the sentencing court did not clearly err in determining that the spreadsheets submitted by the government, along with its other representations, supported an estimate that Kumar conspired to sell approximately 3.8 million pills.
- 16 - estimate was to have attached all of the underlying data (an amount
Kumar at sentencing acknowledged was "large" and a "hodgepodge")
to its sentencing memoranda. We decline to adopt such a rule.
The two cases on which Kumar relies do not compel a
different conclusion. In United States v. Collado, the Third
Circuit found that there was insufficient evidence for a sentencing
court to make a drug quantity calculation with respect to a
specific transaction, where the only evidence offered by the
government on the issue consisted of transcripts of two phone calls
that did not reference any amount of drugs. See 975 F.2d 985,
998–99 (3d Cir. 1992). And in United States v. Washington, the
Eleventh Circuit found that there was insufficient evidence for a
sentencing court to establish that the defendant's crime involved
more than 250 victims, where the only evidence offered was the
government's bare assertion that over 6,000 individuals had their
credit card numbers stolen. See 714 F.3d 1358, 1361–62 (11th Cir.
2013). The sentencing courts erred in both instances because they
allowed the government to "cross[] the line" from permissible
estimation to impermissible speculation. Collado, 975 F.2d at
998. The sentencing court did not make anything resembling that
mistake here. Instead, it relied on the government's detailed
explanation of its calculation and a number of sample spreadsheets
in concluding that the government had satisfied its burden of
- 17 - establishing the quantity of pills that Kumar conspired to import.
That reliance was not a clear error.
D.
Kumar's sole remaining attack on the loss amount chart
in the revised PSR is that it does not include any information on
the types of pills sold. But this argument is also unavailing.
It is not at all apparent why the type of pill would need to be in
the chart in order for the quantity of pills and price per
pill -- the factors that directly affect the loss amount
calculation -- to be reasonable estimates. And to the extent that
Kumar is suggesting that the government lacks the ability to link
the data in the chart to the specific types of pills that data
represents, he undermines that assertion later in his briefing
when he tallies up the number of various drugs contained in the
sample spreadsheets provided by the government -- an implicit
acknowledgement that the government can link the data in the
summary chart to specific types of drugs.
***
Regardless of whether one considers Kumar's arguments
under a "clearly erroneous" standard of review or the plain error
rubric more forgiving to the government, the arguments do not
- 18 - establish that the sentencing court's estimate as to the loss
amount was mistaken.8
For these reasons, we affirm Kumar's sentence.
8 As a final matter, we note that the sentencing court would have had to estimate a loss amount of less than $1.5 million for Kumar's 87-month sentence to have been above the Guidelines range. Kumar has not argued below or in front of us that the loss amount properly calculated should fall below that mark.
- 19 -