United States v. Harinder Singh

FOR PUBLICATION

UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT

UNITED STATES OF AMERICA, No. 18-50423
Plaintiff-Appellee,
D.C. No.
v. 2:14-cr-00648-
CAS-9
HARINDER SINGH, AKA Lnu, Sonu,
Defendant-Appellant. OPINION

Appeal from the United States District Court
for the Central District of California
Christina A. Snyder, District Judge, Presiding

Argued and Submitted November 9, 2020
Pasadena, California

Filed May 3, 2021

Before: Barrington D. Parker, * Paul J. Watford, and
Patrick J. Bumatay, Circuit Judges.

Opinion by Judge Parker;
Partial Concurrence and Partial Dissent by Judge Watford

*
The Honorable Barrington D. Parker, United States Circuit Judge
for the U.S. Court of Appeals for the Second Circuit, sitting by
designation.
2 UNITED STATES V. SINGH

SUMMARY **

Criminal Law

The panel affirmed Harinder Singh’s convictions and
sentence for conspiracy to launder money (18 U.S.C.
§ 1956(h)), conspiracy to operate an unlicensed money
transmitting business (18 U.S.C. § 371), and operating such
a business (18 U.S.C. § 1960), stemming from Singh’s
involvement in a hawala operation, a money transmitting
network that he and his coconspirators used to move drug
trafficking proceeds from Canada to the United States and
eventually to Mexico.

Rejecting Singh’s sufficiency-of-the-evidence challenge
to his § 1956 conviction, the panel held that a jury could have
reasonably concluded that Singh intended to conceal the
ownership and control of the drug proceeds, as required by
18 U.S.C. § 1956(a)(1)(B)(i).

The panel also rejected Singh’s sufficiency-of-the-
evidence challenge to his convictions under § 1960, which
provides that money transmitting “includes” transferring
funds on behalf of the public. Explaining that “includes”
deems what follows to be a non-exhaustive list of what the
statute covers, the panel held that “on behalf of the public”
is not a necessary element of § 1960. The panel disagreed
with Singh’s argument that because he did not advertise his
services or make them generally available to everyone, his
transactions were not “on behalf of the public.” The panel

**
This summary constitutes no part of the opinion of the court. It
has been prepared by court staff for the convenience of the reader.
 UNITED STATES V. SINGH 3

therefore concluded that Singh’s conduct triggered liability
under § 1960. The panel held that even if “on the behalf of
the public” were an element—which it is not—the
government proved it.

As to Singh’s contention that the government’s closing
arguments constructively amended the indictment’s § 1960
counts, the panel saw no plain error. The panel explained
that the indictment charges that Singh worked with others in
a money transmitting business based on the hawala network,
which is not “distinctly different” from charging Singh with
conducting his own money transmitting business, and that
the indictment was not substantially altered at trial.

The panel held that the district court did not violate the
Confrontation Clause, nor abuse its discretion, in limiting the
cross-examination of a cooperating witness.

Without resolving whether a clear and convincing
evidence standard or a preponderance of the evidence
standard should apply, the panel held that the record
supports, under either standard, the district court’s
application of an enhancement under U.S.S.G. § 2S1.1(b)(1)
based on Singh’s knowing that the laundered funds were
drug trafficking proceeds.

Judge Watford concurred in part and dissented in part.
He agreed with the majority that Singh’s conduct rendered
him guilty of operating an unlicensed money transmitting
business in violation of § 1960, but in his view, Singh’s
conduct did not amount to participation in a money
laundering conspiracy.
4 UNITED STATES V. SINGH

COUNSEL

Elizabeth Richardson-Royer (argued), San Francisco,
California, for Defendant-Appellant.

Elana Shavit Artson (argued) and Carol Alexis Chen,
Assistant United States Attorneys; L. Ashley Aull, Chief,
Criminal Appeals Section; Nicola T. Hanna, United States
Attorney; United States Attorney’s Office, Los Angeles,
California; for Plaintiff-Appellee.

OPINION

PARKER, Circuit Judge:

After a seven-day trial, a jury convicted Harinder Singh
(“Singh”) of one count of conspiracy to launder money (see
18 U.S.C. § 1956(h)), one count of conspiracy to operate an
unlicensed money transmitting business (see 18 U.S.C.
§ 371), and one count of operating such a business (see
18 U.S.C. § 1960). The convictions stemmed from Singh’s
activities as a participant in a money transmitting enterprise
which transferred and laundered drug trafficking proceeds. 1

On this appeal, Singh raises a number of contentions, but
principally argues that the government adduced insufficient
evidence to support his conviction. He also argues that the
government’s proof at trial and its closing argument
constructively amended the indictment and that the district
court erroneously limited cross-examination of a

1
The indictment, originally returned November 13, 2014, included
22 defendants. Most co-defendants entered pleas of guilty and did not
proceed to trial.
 UNITED STATES V. SINGH 5

government witness. Lastly, Singh argues that the court
below erred in adding a six-level sentencing enhancement
because he knew the laundered funds were drug proceeds.
See U.S.S.G. § 2S1.1(b)(1). Finding no merit to these
contentions, we affirm.

BACKGROUND

Singh’s convictions derive from his involvement in a
hawala operation, a money transmitting network that he and
his coconspirators used to move drug trafficking proceeds
from Canada to the United States and eventually to Mexico.
Considered in the light most favorable to the government, its
proof at trial established the following. In early 2012,
Gurkaran Singh Isshpunani began to work for Deepinder
“Pindi” Singh, a drug trafficker, to transfer drug proceeds
from Canada to the United States. As a hawala broker,
Isshpunani collected Canadian funds from Pindi and worked
with other hawala dealers (including the defendant) to
coordinate the transfer of equivalent U.S. funds to California
where they were used to pay Mexican drug suppliers. Singh
worked in California. His primary role in the conspiracy was
to deliver drug proceeds to various hawala brokers in
California and elsewhere who then orchestrated the delivery
of the funds to a Mexican drug cartel.

Hawala is a system designed to transfer funds from point
to point outside of formal money transmission channels
without the physical movement of money. Typically, the
system is used to transfer funds from one country to another
through hawala brokers. A broker in one country receives
money and then communicates with a broker in the country
receiving the transfer. The broker in the receiving country
then pays out an equivalent amount (deducting for fees) to
the recipient in the appropriate currency. Hawala
transactions are discreet. They typically involve minimal
6 UNITED STATES V. SINGH

record-keeping, are not subject to government regulation and
are premised on trust.

Hawala is widely used in communities that have limited
access to formal banking structures and, for example, is an
important vehicle for remittance payments from immigrants
to family members in their home countries. But because of
its informality, lack of record keeping and government
oversight, hawala may be used to transfer illegally derived
funds, as was the case here.

The government’s proof established that the network in
which Singh was involved transferred, over a considerable
period of time, large sums of Canadian dollars from Pindi’s
Canadian drug operation to Los Angeles. Isshpunani worked
with a broad network of hawala brokers, based in California
and in India, to orchestrate the delivery of funds which had
been sent to California to the Mexican cartel.

In spring of 2012, Singh was recruited into the operation
by his uncle, Sucha Singh, who ran a hawala business.
Initially, Singh worked for his uncle but later worked
independently. Singh’s primary responsibility was
collecting and distributing money to Pindi’s associates. The
government’s proof at trial established that Singh knew the
funds were drug proceeds. Sanjiv “Bobby” Wadhwa, a co-
defendant who later became a government witness, testified
at trial that he told Singh that the funds Singh moved were
drug proceeds.

Singh was a hard worker. In 2012, he completed 10–15
deliveries for Isshpunani of sums ranging from $100,000 to
about $800,000. He received $250 for each $100,000
delivered. Singh also worked directly with Wadhwa,
ultimately completing 30–40 money collections of amounts
 UNITED STATES V. SINGH 7

ranging from $50,000 to $150,000 between April and
October 2012.

These transactions were clothed in secrecy and a number
of steps, above and beyond those routinely used in hawala
transactions, were taken to hide the nature of the
transactions. Singh switched out his SIM card and phone
number every 20 to 25 days. Members of the conspiracy
used burner phones—disposable, prepaid, effectively
untraceable devices to communicate among themselves.
Transfers involved code words such as “shaman” and
“merchandise” to disguise the nature of the transactions. The
hawala merchants used serial numbers on dollar bills to
verify that the person who received the cash was the intended
recipient. Higher than usual fees were charged.

The government’s proof at trial included video
surveillance records that showed Singh making deliveries on
a number of occasions as well as Singh’s own ledger which
documented his activities and included cash amounts,
recipients, and serial numbers. Finally, the government
adduced evidence that Singh was stopped in October 2012
by a California Highway Patrol officer and told the officer
that bags found in the car carried his wife’s shoes, but the
bags actually contained cash that he was on his way to
deliver. After discovering the bags, the officer arrested
Singh. After the arrest, law enforcement officers searched
his home and seized large sums of cash as well as drug
ledgers.

In addition to arguing that this evidence was insufficient
to establish his guilt on the three counts on which he was
convicted, Singh argues that two errors by the trial court
require reversal. As noted, Sanjiv “Bobby” Wadhwa
testified for the government at trial as a cooperating witness.
At some point, defense counsel received information that the
8 UNITED STATES V. SINGH

FBI had investigated him based on an allegation that he had
planned to murder Taran Singh, another hawala dealer. Both
were alleged to be members of the conspiracy. The FBI
ultimately concluded that the allegation was unsubstantiated
and closed the case. At trial, defense counsel attempted to
cross-examine Wadhwa regarding his involvement in the
murder-for-hire plot, arguing that the evidence was relevant
to his credibility. Defense counsel also sought to have
recordings of Wadhwa speaking about the murder-for-hire
plans, including discussing a $30,000 payment, admitted
into evidence to refresh his recollection.

The court ruled that defense could inquire into whether
Wadhwa was involved in the murder-for-hire scheme but
that, citing Fed. R. Evid. 608(b), if Wadhwa denied his
involvement, the inquiry must end, and extrinsic evidence
could not be admitted to impeach him. When questioned,
Wadhwa disavowed any involvement in a murder-for-hire
scheme. The court explained that it limited cross-
examination in order to “prevent impeachment of [him] on a
collateral matter and to avoid a mini-trial on the issue of the
murder-for-hire plot[.]” The court also excluded the
recordings. Singh contends that these limitations violated the
Confrontation Clause, U.S. Const. amend. VI.

Next, Singh contends that he is entitled to reversal
because at trial the court permitted a constructive
amendment of the indictment. Singh contends that the
indictment charged only “a single, joint money transmitting
business consisting of the entire hawala network and the
various transactions . . . within it.” At trial, however, the
government offered proof and argued to the jury that Singh
operated a money transmitting business. This variance, he
contends, violated the Fifth Amendment, U.S. Const.
amend. V.
 UNITED STATES V. SINGH 9

Following Singh’s conviction, the Probation Office
calculated an offense level of 34. The components were a
base offense level of eight, an 18-level enhancement because
the amount of laundered funds was between $3.5 and
$9.5 million, a six-level enhancement because Singh knew
or believed the funds were related to drug trafficking and a
two-level enhancement because Singh was convicted under
18 U.S.C. § 1956. Based on a Criminal History Category of
I, these calculations yielded an advisory Guidelines range of
151–188 months. At sentencing, Singh objected to the six-
level enhancement, contending that there was a lack of clear
and convincing proof that he knew the funds were derived
from drugs. The court disagreed but sentenced him well
below his Guidelines range to 70 months. This appeal
followed. For the reasons that follow, we affirm.

DISCUSSION

I

Singh’s main arguments are that the government
adduced insufficient evidence of a purpose to conceal, as
required by 18 U.S.C. § 1956(a)(1)(B)(i), to support his
conviction for concealment money laundering under Count
I, and insufficient evidence of public involvement to support
his convictions for operating and conspiring to operate a
money transmitting business under Counts II and III, see
18 U.S.C. § 1960(b)(2). When evaluating a sufficiency
challenge, “the relevant question is whether, after viewing
the evidence in the light most favorable to the prosecution,
any rational trier of fact could have found the essential
elements of the crime beyond a reasonable doubt.” Jackson
v. Virginia, 443 U.S. 307, 319 (1979); Long v. Johnson, 736
F.3d 891, 895–56 (9th Cir. 2013). We review sufficiency of
evidence challenges de novo. See United States v. Corrales-
Vazquez, 931 F.3d 944, 947 (9th Cir. 2019).
10 UNITED STATES V. SINGH

As noted, Singh was convicted of conspiracy to launder
money in violation of 18 U.S.C. § 1956(a)(1)(B)(i). (Count
I). The substantive elements of that offense are: “(1) the
defendant conducted or attempted to conduct a financial
transaction; (2) the transaction involved the proceeds of
unlawful activity; (3) the defendant knew that the proceeds
were from unlawful activity; and (4) the defendant knew that
the transaction was designed in whole or in part—(i) to
conceal or disguise the nature, the location, the source, the
ownership, or the control of the proceeds of specified
unlawful activity.” United States v. Wilkes, 662 F.3d 524,
545 (9th Cir. 2011) (internal quotations and citations
omitted). On appeal, Singh only challenges the sufficiency
of the Government’s proof on the 4th element.

On this element, Singh argues there was insufficient
evidence that the transactions he participated in were
designed to conceal illicit drug money. His support for this
contention is Regalado Cuellar v. United States, 553 U.S.
550 (2008). There, the Supreme Court held that a conviction
under 18 U.S.C. § 1956(a)(2)(B)(i), the provision that
criminalizes transportation money laundering and is
analogous to the (a)(1) provision at issue in this case,
required the government to establish that “the purpose—not
merely the effect—of the transportation was to conceal or
disguise a listed attribute.” Cuellar, 553 U.S. at 566. In other
words, that a transaction is structured to hide its source is not
enough. The government must prove that the transaction had
the purpose of concealing the source. Id. at 566 (explaining
“how one moves the money is distinct from why one moves
the money.”).

Cuellar was a drug courier—a “mule”—who was
arrested after law enforcement officers discovered him
transporting $81,000 of drug proceeds to Mexico. They were
 UNITED STATES V. SINGH 11

covered in plastic bags and animal hair and hidden in a secret
compartment in his car. Cuellar, 553 U.S. at 552. The Court
held that although petitioner hid the proceeds to transport
them, the evidence showed that his ultimate purpose was to
“compensate the Mexican leaders of the operation,” not to
conceal the funds. Id. In other words, according to the Court,
Petitioner’s conduct was not designed to conceal an attribute
of the funds but simply to move them. For this reason, the
Court found the evidence insufficient and reversed the
conviction.

Singh argues that Cuellar requires reversal of his
conviction because the government adduced insufficient
evidence that the hawala transactions in which he
participated had a concealment purpose. The purpose,
according to him, was simply to pay Mexican drug suppliers.
In other words, Singh believes he and Cuellar were similarly
situated.

We are not persuaded. The money laundering statute is
violated if the transaction in question is “designed in whole
or in part” to conceal. 18 U.S.C. § 1956(a)(1)(B)(i)
(emphasis added). In Cuellar, the government proved that
the effect of the transportation was payment of the Mexican
drug suppliers, but there was no proof, or at least no
sufficient proof, of a concealment purpose.

We conclude, for a host of reasons, that the transactions
in question had (certainly in part) a concealment purpose.
First, Singh and his co-conspirators used the hawala system.
They could, theoretically, have saved themselves a good deal
of time and effort by using wire transfers or mailing checks:
procedures used countless times everywhere every day to
move funds quickly and efficiently. Instead of doing so, they
used a private system that involved informality,
confidentiality, and intricate pickup and delivery procedures
12 UNITED STATES V. SINGH

with person-to-person contact to move very large sums of
money. This system featured minimal record keeping and no
governmental regulation, oversight or reporting
requirements. While hawala is a system with legitimate users
and an ostensibly legitimate purpose, a jury could have
reasonably concluded from this evidence that Singh used it
for the purpose of concealing the location and ownership of
drug money.

Moreover, Singh did not simply use a basic hawala
system. He used a stepped up system that included a number
of concealment enhancing add-ons. He and his associates
used coded words for drug money (“saman”, “merchandise”)
to facilitate cash pick-ups and drop offs. Instead of using an
iPhone or an Android, he used burner phones which he
changed every 20 to 25 days. Burner phones obviously have
legitimate uses. But they are often used in connection with
drug transactions because there are no readily retrievable
records of who owns them, calls are difficult to trace and it
is considerably more difficult for law enforcement to get
wire-tap authorizations for them. He also used serial
numbers on currency, which were used to verify the identity
of the courier receiving funds. When cash was delivered, the
receiving courier was required to provide a serial number as
verification. Moreover, the hawala system Singh used
charged premium fees to move the Canadian money. Finally,
when Singh was arrested, he falsely stated that a bag in his
car that contained large amounts of drug proceeds held his
wife’s shoes. Based on this constellation of facts, a jury
could have reasonably concluded that Singh intended to
conceal the ownership and control of drug proceeds.

In United States v. Wilkes, the defendant was convicted
of concealment money laundering under § 1956(a)(1)(B)(i)
for payments and gifts to a California congressman in
 UNITED STATES V. SINGH 13

exchange for government contracts. 662 F.3d 524, 530 (9th
Cir. 2011). Wilkes transferred a $525,000 mortgage payment
to the congressman in exchange for a contract; instead of
transmitting the funds directly, Wilkes conducted a series of
transfers, moving the money between different bank
accounts. We concluded that the transactions, “which
provided additional buffers between the corrupt contract and
the payoff of [the congressman’s] mortgage” were intended
to conceal the source of the funds because, as here, they were
“convoluted” and not “simple transactions,” which were
intended to mask the link between the funds and their source.
Id. at 547.

Decisions from other courts reinforce our conclusion. In
United States v. Brown, 553 F.3d 786, 787 (5th Cir. 2008),
the Fifth Circuit found a concealment purpose where “the
defendants intended to and did make it more difficult for the
government to trace and demonstrate the nature of [] funds[,]
. . . the transactions were in cash [and] [m]ost deposits were
below ten thousand dollars” to dodge reporting regulations. 2
In Magluta v. United States, 660 Fed. App’x 803, 807–08
(11th Cir. 2016), the Eleventh Circuit found a concealment

2
Accord United States v. Diaz, 2008 WL 4387209, *1 (S.D.N.Y.
2008) (finding a concealment purpose from a “sophisticated and
complex financial scheme” that moved drug funds from New York to the
Dominican Republic); United States v. Spencer, 2008 WL 4104693, *4
(D. Minn. 2008) (concluding that “mak[ing] it harder to trace the source
of [] money” suggests a concealment purpose); but see United States v.
Garcia; 587 F.3d 509 (2d Cir. 2009) (declining to find a concealment
purpose where defendant secretly transported $2.2 million in drug
proceeds across the U.S.); United States v. Ness, 565 F.3d 73, 78 (2d Cir.
2009) (finding no concealment purpose where defendant transported
millions of dollars in drug proceeds abroad because there was only “an
intent to conceal the transportation, not an intent to transport in order to
conceal.”).
14 UNITED STATES V. SINGH

purpose where checks (derived from drug funds) given to
criminal defense lawyers had false payees and the funds
themselves were moved from “Miami to New York to Israel
and then back to Miami.” In sum, we conclude that the
Government adduced sufficient evidence of Singh’s
concealment purpose.

II

Next, Singh argues that the evidence introduced by the
Government was insufficient to support his conviction under
18 U.S.C. § 1960, which bars the operation of an unlicensed
money transmitting business. “Money transmitting” under
§ 1960(b)(2) “includes transferring funds on behalf of the
public by any and all means including but not limited to
transfers within this country or to locations abroad[.]”

“A money transmitting business receives money from a
customer and then, for a fee paid by the customer, transmits
that money to a recipient in a place that the customer
designates[.]” United States v. Velastegui, 199 F.3d 590, 592
(2d Cir. 1999). That is precisely what Singh did. The
government’s proof at trial established that Singh’s conduct
fit this definition.

Singh contends that “on behalf of the public” is an
essential element of § 1960 which the government failed to
prove beyond a reasonable doubt. The reasoning behind this,
Singh maintains, is that “includes” in the statute’s text
should be understood as signifying “means.” We disagree.
We believe that “includes” deems what follows to be read as
 UNITED STATES V. SINGH 15

a non-exhaustive list of what the statute covers. 3 Thus, we
hold that “on behalf of the public” is not a necessary element
of § 1960.

To address what constitutes “on behalf of the public:” we
believe that for money transmission to be conducted “on
behalf of the public” under § 1960, it must occur within a
transactional, business dealing or for a member of the
broader community rather than within a personal or close
relationship. See, e.g., United States v. $215,587.22 in U.S.
Currency, 306 F. Supp. 3d 213, 218 (D.D.C. 2018) (defining
“on behalf of the public” as a money transmission that is
“made for third-parties or customers as part of a commercial
or business relationship, instead of with one’s own money or
for family or personal acquaintances.”). That is what
occurred here.

Singh argues that because he did not advertise his
services or make them generally available to everyone, his
transactions were not “on behalf of the public.” We disagree.
We find it highly unlikely—indeed inconceivable—that
Congress intended to limit § 1960 to money transferring
businesses that used TV commercials, business cards or
billboards. For these reasons, we conclude that Singh’s
conduct triggered liability under § 1960.

However, even if “on behalf of the public” were an
element—which it is not—the government proved it. Given
the numerosity, scale, and frequency of Singh’s transactions,
a jury could reasonably have concluded that his conduct was
what Congress intended to proscribe and what the statute in

3
Cf. United States. v. Wyatt, 408 F.3d 1257, 1261 (9th Cir. 2005)
(interpreting the statutory definition of “includes” as “non-exhaustive
rather than exclusive.”).
16 UNITED STATES V. SINGH

fact proscribes. Singh, after all, was not a small-time hawala
courier who limited his dealings to a small circle of family
and friends: he was involved in dozens and dozens of
transactions. For example, he picked up hundreds of
thousands of dollars from Taran on 30–35 occasions, and he
made 10–15 deliveries on Isshpunani’s behalf in amounts
between $100,000 and $800,000. He also transacted with
various parties in parking lots, apartment complexes,
warehouses, electronics stores and elsewhere. These
activities were extensive, involving many people and lots of
money. Drawing all inferences in the government’s favor, it
was reasonable for the jury to conclude that Singh was
operating a sufficiently publicly oriented money
transmitting business to fall under § 1960. See S. Rep. No.
101-460, at 14 (1990), reprinted in 1990 U.S.C.C.A.N. 6645,
6658–59; United States v. $215,587.22 in U.S. Currency,
306 F. Supp. 3d 213, 218 (D.D.C. 2018); see also United
States v. Banki, 685 F.3d 99, 114 (2d Cir. 2012) (defining
“business” under § 1960 as “an enterprise that is carried on
for profit or financial gain”). In sum, the government
adduced sufficient evidence to support Singh’s convictions
under § 1960 (Counts II and III).

III

Next, Singh argues that the government’s closing
arguments constructively amended Counts II and III of the
indictment. He contends that the indictment charged a
“single, joint money transmitting business consisting of the
entire hawala network and the various transactions . . .
within it,” but the government argued at trial that he operated
a money transmitting business of his own. Because Singh
 UNITED STATES V. SINGH 17

failed to object at trial, we review for plain error. 4 We see
none.

A constructive amendment is an alteration to the
indictment’s terms “either literally or in effect, by the
prosecutor or a court after the grand jury has last passed upon
them.” Id. at 1182–83. We have identified two kinds of
constructive amendments: (1) those involving a “complex of
facts presented at trial distinctly different from those set
forth in the charging instrument” and (2) those where “the
crime charged in the indictment was substantially altered at
trial, so that it was impossible to know whether the grand
jury would have indicted for the crime actually proved.”
United States v. Davis, 854 F.3d 601, 603 (9th Cir. 2017).
Neither occurred here.

The facts the government presented at trial were not
“distinctly different” from those in the indictment. The
government’s proof established that the hawala network in
which Singh operated was an extensive one involving many
brokers and many transactions. Initially, Singh worked for
his uncle but, as time went on, he worked independently.
Further, the government’s trial arguments did not
substantially alter the indictment. Both the indictment and
the government’s proof at trial were directed at the same
offense: operating an unlicensed money transmitting
business. Whether he shared income with his uncle or kept
it for himself is of no moment. He was still operating an

4
Plain error occurs “if there has been (1) error; (2) that was plain;
(3) that affected substantial rights; and (4) that seriously affected the
fairness, integrity, or public reputation of the judicial proceedings.”
United States v. Mickey, 897 F.3d 1173, 1183 (9th Cir. 2018).
18 UNITED STATES V. SINGH

unlicensed business. Thus, we see no error and certainly no
plain error.

Singh seeks support from Stirone v. United States,
361 U.S. 212, 217 (1960) and United States v. Ward,
747 F.3d 1184, 1192 (9th Cir. 2014), both cases where the
courts found a constructive amendment. In Stirone, the
Supreme Court found a constructive amendment when the
indictment charged the defendant with unlawful interference
with the interstate movement of sand, while the trial court’s
instruction allowed the jury to convict for either unlawful
sand or steel shipments. The Court held that the indictment
could not “fairly be read” as containing the same charge as
the conviction. Stirone, 361 U.S. at 217. In Ward, this court
found a constructive amendment where there was ambiguity
around whether identity theft convictions were based on the
indictment’s charge or “uncharged conduct.” 747 F.3d
at 1191. In that case, the jury may have convicted the
defendant for aggravated identity theft against victims who
were not specified in the indictment. A constructive
amendment occurred because, since “the identity of the
victims was necessary to satisfy an element of the offense,”
the conviction was not unequivocally based on the
indictment’s charged conduct. Id. at 1192.

In contrast to these cases, the indictment charges that
Singh worked with others in a money transmitting business
based on the hawala network, which is not “distinctly
different” from charging Singh with conducting his own
money transmitting business and did not “substantially alter”
the charges Singh faced.

IV

Next, Singh argues that the trial court violated the
Confrontation Clause by limiting the cross-examination of
 UNITED STATES V. SINGH 19

Sanjiv “Bobby” Wadhwa, who testified at trial as a
cooperating witness. At some point, defense counsel
received information that the FBI had investigated Wadhwa
based on an allegation that he had planned to murder Taran
Singh, another hawala dealer. Both were alleged to be
members of the conspiracy. The FBI ultimately concluded
that the allegation was unsubstantiated and closed the case.
At trial, defense counsel attempted to cross-examine
Wadhwa regarding his involvement in the murder-for-hire
plot, arguing that the evidence was relevant to his credibility.
Defense counsel also sought to have recordings of Wadhwa
speaking about the murder-for-hire plans, including
discussing a $30,000 payment, admitted into evidence to
refresh his recollection.

The district court ruled that defense could inquire into
whether Wadhwa was involved in the murder-for-hire
scheme; but, citing Rule 608(b), if Wadhwa denied his
involvement, the inquiry would need to end and extrinsic
evidence could not be admitted to impeach him. When
questioned, Wadhwa disavowed any involvement in a
murder-for-hire scheme. The court explained that it limited
cross-examination in order to “prevent impeachment of
[him] on a collateral matter and to avoid a mini-trial on the
issue of the murder-for-hire plot[.]” The court also excluded
the recordings. Singh contends that these limitations violated
the Confrontation Clause, U.S. Const. amend. VI. This court
reviews Confrontation Clause-based challenges to a district
court’s limitations on cross-examination de novo. See United
States v. Larson, 495 F.3d 1094, 1101 (9th Cir. 2007).
However, this court will review “[a] challenge to a trial
court’s restrictions on the manner or scope of cross-
examination on non-constitutional grounds” for an abuse of
discretion. Id.
20 UNITED STATES V. SINGH

The Confrontation Clause secures a defendant’s right “to
be confronted with the witnesses against him.” U.S. Const.
amend. VI. The Clause also guarantees “the right of effective
cross-examination.” Larson, 495 F.3d at 1102. However, the
right to cross-examine is subject to very well-established
limitations that permeate the Federal Rules of Evidence.
“[T]rial judges retain wide latitude insofar as the
Confrontation Clause is concerned to impose reasonable
limits on such cross-examination based on concerns about
. . . harassment, prejudice, confusion of the issues. . . or
interrogation that is. . . only marginally relevant.” Id. at 1101
(citation omitted).

At trial, Singh made extensive use of his right to
“confront” Wadhwa. Wadhwa testified for approximately
two and a half hours, and he was cross-examined extensively
about meeting with his cellmate’s wife and one of her
associates and about whether, during that meeting, he agreed
to have Taran killed in exchange for a payment of $30,000.
The court below imposed limitations on cross-examination,
invoking Rules 608(b) and 403, but there are precious few
federal criminal trials in which limitations of one kind or
another on cross-examination are not imposed.

United States v. Mikhel, 889 F.3d 1003, 1048 (9th Cir.
2018), is our test for when restrictions on cross-examination
become sufficiently extensive to raise Confrontation Clause
concerns that may undermine the fairness of a trial. Under
Mikhel, the inquiry is “(1) whether the excluded evidence
was relevant; (2) whether there were other legitimate
interests outweighing the defendant’s interest in presenting
the evidence; and (3) whether the exclusion of evidence left
the jury with sufficient information to assess the witness’s
credibility.” Id. (citing Larson, 495 F.3d at 1103).
 UNITED STATES V. SINGH 21

Here, the relevance of the additional questioning Singh’s
counsel wished to pursue—about recordings of meetings
between Wadhwa and his cellmate and the cellmate’s wife
related to the murder-for-hire—was, as the district court
ruled, highly attenuated and convoluted. The line of
examination defense counsel wished to pursue “becomes a
he-said/he-said/he-said and then she-said/he-said . . . [i]t’s
confusing because there’s a lot of different versions.”
Moreover, the trial judge concluded that the line of cross-
examination in question was not sufficiently relevant to any
potential bias Wadhwa might harbor because it involved
events that were simply too peripheral.

Under Mikhel’s second prong, it was well within the trial
judge’s discretion to limit cross-examination to prevent “a
trial-within-a-trial.” 889 F.3d at 1048. The trial judge did just
that, explaining “[w]e are not here to try Mr. Wadhwa for a
plot to murder another witness. It is collateral . . . we are not
trying the murder for hire case. We are trying the hawala
money laundering case.”

Lastly, the exclusion in question certainly left the jury
with enough evidence to assess Wadhwa’s credibility. The
jury already knew that Wadhwa had pleaded guilty, that the
government first approached him about testifying against
Singh while Wadhwa was in prison after sentencing, and that
Wadhwa was seeking a lower sentence. Moreover, the trial
judge did not completely exclude any inquiry about the
murder-for-hire plot. He permitted a question as to whether
Wadhwa had been involved in the scheme. Wadhwa denied
his involvement, and under Rule 608(b), the trial court acted
well within its discretion in ending the matter there. The
court also invoked Rule 403: “I’m not going to have a trial
on whether there was, in fact, a murder-for-hire plot and all
the meetings he may have had to effectuate those things
22 UNITED STATES V. SINGH

because I think that they are collateral, time-consuming, and
unfairly prejudicial, and they’re going to divert the jury from
this case.” Later, when denying defendant’s motion for a
new trial, the judge elaborated: “the probative value of
Wadhwa’s involvement in a murder-for-hire plot was
substantially outweighed by the danger of confusing the
issues before the jury and wasting time with a mini-trial
[especially considering] that the murder-for-hire allegations
against Wadhwa were found to be unsubstantiated.” We see
no Confrontation Clause violation and no abuse of discretion
in these rulings.

V

Finally, Singh challenges the district court’s application
of a six-level sentencing enhancement under USSG
§ 2S1.1(b)(1) because Singh knew that the laundered funds
were drug trafficking proceeds. Under Count I, the
government was required to prove, and did prove, that the
funds in question were derived from illegal activity but was
not required to prove that the funds were drug proceeds. The
parties disagree over the proper standard of proof the district
court should have applied to establish the facts supporting
the enhancement. Singh, relying on United States v. Staten,
466 F.3d 708 (9th Cir. 2006), argues that a clear and
convincing evidence standard should apply because the
application of the enhancement produces a disproportionate
impact on the sentence compared to the offense of
conviction. The government argues that the preponderance
of the evidence standard should apply. It reasons that once
the Guidelines became permissive, and not mandatory, the
binary approach to uncharged enhancements under Staten
was no longer appropriate and that this case should become
the vehicle for the Circuit to revisit the decision.
 UNITED STATES V. SINGH 23

We are not required to resolve this issue because the
record supports the application of the enhancement under
either standard of proof. The government’s proof at trial that
the funds were derived from drug trafficking and that Singh
knew that source was overwhelming. The entirety of Singh’s
seven-day trial centered around drug money. In fact, the
government’s only theory of illegality was that the funds
were the proceeds of drug trafficking. Moreover, the
government proved Singh knew the funds were drug
proceeds. Wadhwa testified that he told Singh that the
hawala money was from “davai” or drugs. Sucha also made
statements during a telephone call that was introduced into
evidence that strongly suggest Singh knew about the funds
were related to drug trafficking. On the strength of this
record, the district court concluded—quite correctly in our
view—that there was “substantial evidence that defendant
knew that the proceeds and the laundered funds were
connected to drug activity.”

Finally, we note the district court ultimately imposed a
sentence of 70 months, which is well below Singh’s
Guidelines range of 151–188 months. For these reasons, we
see no merit to Singh’s challenge to his sentence. 5

5
Moreover, even under this court’s disproportionate impact test in
United States v. Gonzalez, the clear and convincing evidence standard
would not apply. 492 F.3d 1031, 1039 (9th Cir. 2007); see also United
States v. Johansson, 249 F.3d 848 (9th Cir. 2001); United States v.
Jordan, 256 F.3d 922 (9th Cir. 2001). Gonzalez lists the six factors that
comprise the disproportionate impact test: “1. Does the enhanced
sentence fall within the maximum sentence for the crime alleged in the
indictment? 2. Does the enhanced sentence negate the presumption of
innocence or the prosecution's burden of proof for the crime alleged in
the indictment? 3. Do the facts offered in support of the enhancement
create new offenses requiring separate punishment? 4. Is the increase in
24 UNITED STATES V. SINGH

CONCLUSION

The judgment of the District Court is AFFIRMED.

WATFORD, Circuit Judge, concurring in part and
dissenting in part:

Harinder Singh helped transfer drug proceeds from a
drug trafficker in Canada to drug suppliers in Los Angeles.
I agree with my colleagues that this conduct rendered Singh
guilty of operating an unlicensed money transmitting
business in violation of 18 U.S.C. § 1960. In my view,
however, Singh’s conduct did not amount to participation in
a money laundering conspiracy. I therefore join Parts II
through IV of the majority opinion but am unable to join
Parts I and V.

I will be the first to concede that, on the surface, using a
hawala network to transfer hundreds of thousands of dollars

sentence based on the extent of a conspiracy? 5. Is the increase in the
number of offense levels less than or equal to four? 6. Is the length of the
enhanced sentence more than double the length of the sentence
authorized by the initial sentencing guideline range in a case where the
defendant would otherwise have received a relatively short sentence?”
The enhanced sentence of 151–188 months falls within the maximum
sentence (20 years) and the enhanced sentence does not negate the
presumption of innocence or the government’s burden of proof.
Moreover, the enhancement facts do not create a new offense and the
sentence increase is not derived from the extent of a conspiracy. While
the offense level increase (six) is greater than four and the enhanced
sentence length (151 to 188 months) more than doubles the length based
on the initial guidelines range (78 to 97 months), these factors,
considered in the aggregate, do not require application of a clear and
convincing evidence standard.
 UNITED STATES V. SINGH 25

in drug proceeds from Canada to Los Angeles certainly
seems like it should violate 18 U.S.C. § 1956(a)(1)(B)(i), the
statutory provision at issue here. The provision prohibits
engaging in a “financial transaction” involving the proceeds
of unlawful activity “knowing that the transaction is
designed in whole or in part . . . to conceal or disguise the
nature, the location, the source, the ownership, or the control
of the proceeds of specified unlawful activity.” Using a
hawala network to transfer money undoubtedly qualifies as
a “financial transaction” as that term is defined.
§ 1956(c)(3)–(4). In addition, transfers through a hawala
network unquestionably have the effect of concealing the
flow of money; they are far less transparent from law
enforcement’s perspective than, say, wire transfers through
a bank. While hawala brokers may keep informal ledgers
recording the senders, recipients, and amounts transferred,
they do not maintain the kind of detailed transactional
records that banks and other financial institutions must. And
it’s a safe bet that hawala brokers do not alert the government
to suspicious transactions involving large amounts of cash,
as banks and other financial institutions are required to do.

But does that mean anyone who uses a hawala network
to transfer illicit funds from point A to point B is guilty of
money laundering? The Supreme Court’s decision in
Regalado Cuellar v. United States, 553 U.S. 550 (2008),
suggests that the answer is no.

In Cuellar, the Court reviewed a defendant’s conviction
for transporting $81,000 in drug proceeds to Mexico. The
conviction arose under a neighboring provision of the money
laundering statute that prohibits transporting, transmitting,
or transferring proceeds of unlawful activity into or out of
the United States “knowing that such transportation,
transmission, or transfer is designed in whole or in part . . .
26 UNITED STATES V. SINGH

to conceal or disguise the nature, the location, the source, the
ownership, or the control of the proceeds of specified
unlawful activity.” 18 U.S.C. § 1956(a)(2)(B)(i). As one
can see, this provision directly parallels the provision at issue
in our case, § 1956(a)(1)(B)(i). Both prohibit engaging in
conduct with proceeds of unlawful activity for any of the
same forbidden purposes. One simply targets financial
transactions involving illicit funds, while the other targets
transporting, transmitting, or transferring such funds.
Because the “designed . . . to conceal or disguise” clause of
the two provisions is identically worded, lower courts have
held that Cuellar’s holding applies with equal force to
§ 1956(a)(1)(B)(i). See, e.g., United States v. Brown,
553 F.3d 768, 786 n.56 (5th Cir. 2008); United States v.
Huezo, 546 F.3d 174, 179 (2d Cir. 2008).

The Court said two things in Cuellar that are of prime
importance to the analysis in our case. First, the Court
interpreted the statute’s use of the term “design” to mean
“purpose or plan; i.e., the intended aim of the
transportation.” Cuellar, 553 U.S. at 563. Thus, a
conviction under § 1956(a)(2)(B)(i) “requires proof that the
purpose—not merely effect—of the transportation was to
conceal or disguise a listed attribute” of the funds. Id. at 567.
The Court stressed the distinction between purpose and
effect because in that case there was no question that the
effect of the transportation was to make it harder for law
enforcement to track the location and control of the funds.
Rather than sending the money by wire transfer, the
defendant tried to transport $81,000 in cash to Mexico in a
Volkswagen Beetle. He went to considerable lengths to
conceal the fact that he was transporting the money across
the border. Officers found the cash hidden in a secret
compartment beneath the car’s rear floorboard, bundled in
plastic bags and duct tape. Animal hair had been spread over
 UNITED STATES V. SINGH 27

the secret compartment, presumably to mask the smell of
marijuana emanating from the money. And someone had
taken steps to cover up the recent creation of the secret
compartment. Id. at 554.

Notwithstanding this evidence of a concealment effect,
the Court reversed the defendant’s conviction because the
evidence did not establish a concealment purpose. The
government’s expert testified that the purpose of
transporting the cash to Mexico was to pay the leaders of the
drug-trafficking organization located there. Id. at 566 & n.7.
In other words, the “intended aim” of the transportation was
simply to move the money from point A to point B. The
government did not prove that, in addition, the transportation
was designed to conceal or disguise a listed attribute of the
funds. Such a purpose might have been shown if, for
example, the defendant had transported the funds to Mexico
so that they could be buried in the desert, thereby concealing
their location from authorities. See id. at 558–59, 565.

Second, the Court drew a distinction between proof
concerning how the funds were transported and proof
concerning why they were transported. The concealment
evidence the government offered related to “the manner in
which [the transportation] was carried out.” Id. at 564. The
Court noted that the elaborate steps the defendant took to
conceal his transportation of the funds could serve as
circumstantial evidence that transporting the cash was
designed in part to conceal a listed attribute of the funds.
But, the Court held, evidence concerning how the defendant
moved the money was not sufficient on its own to prove why
he moved the money. Id. at 566. As far as the government’s
evidence showed, the “why” was simply to pay the leaders
of the drug-trafficking organization in Mexico, nothing
more.
28 UNITED STATES V. SINGH

The government’s evidence in our case suffers from the
same deficiencies the Court identified in Cuellar. To be
sure, the government proved that the financial transactions
at issue—transferring the funds through a hawala network
rather than by wire transfer or check—had the effect of
making it harder for law enforcement to track the location
and control of the funds. But just as in Cuellar, the
government’s proof did not establish that the “intended aim”
of the hawala transfers was to conceal or disguise a listed
attribute of the funds. Id. at 563. The government’s expert
in this case, too, testified that the purpose of the hawala
transfers was simply to pay off debts owed to the drug
suppliers in Los Angeles. In other words, just as in Cuellar,
the government proved only that the intended aim of the
financial transactions was to move drug proceeds from point
A to point B.

The majority suggests that this case involves something
more than using ordinary hawala transfers to move illicit
funds from one location to another. It relies on evidence that
the defendants tried to conceal the hawala transfers by using
code words, burner phones, and serial numbers on the
currency to verify the identity of the recipient—what the
majority refers to as “concealment enhancing add-ons.”
Maj. op. at 12. But the use of code words, burner phones,
and serial numbers during the hawala transactions is
equivalent to the efforts to prevent detection of the funds
during transportation that the Supreme Court found
insufficient to prove purpose in Cuellar. 553 U.S. at 563,
566. The evidence cited by the majority relates to the
manner in which the hawala transfers were carried out, not
why they were carried out. As noted, when the government’s
expert addressed the “why” question, he testified that the
purpose of the hawala transfers was to pay debts owed to the
leaders of the drug-trafficking organization in Los Angeles.
 UNITED STATES V. SINGH 29

The government introduced no other evidence concerning
the purpose of the transfers, so Singh’s conviction cannot be
saved by resorting to the statute’s “designed in whole or in
part” language.

The majority states that our decision in United States v.
Wilkes, 662 F.3d 524 (9th Cir. 2011), and cases from other
courts support its conclusion that these transactions evince a
concealment purpose, even under Cuellar. But our case
lacks what was critical in each of those other cases: evidence
of unnecessarily complex transactions. In Wilkes, for
example, the defendant moved funds intended as a bribe
through a series of “convoluted” transactions rather than
transmitting the money directly to the recipient of the bribe.
Id. at 547. Because the transactions between various
accounts were unnecessary, the evidence supported the
conclusion that the “dominant, if not the only, purpose” of
these transactions was to conceal the source and ownership
of the money. Id. Here, by contrast, there is no evidence
that the defendants carried out superfluous transactions or
that any of the transactions were intended to create a buffer
between the source and recipient of the funds.

Nor did the funds in our case travel a circuitous route to
their destination, as in Magluta v. United States, 660 F.
App’x 803 (11th Cir. 2016). In Magluta, the defendant
transferred funds from Miami to New York to Israel;
deposited cash in a bank account in Israel under a false name;
and then issued checks from that sham account to pay his
lawyers back in Miami. Id. at 807. The court held that this
evidence “would permit the jury to infer that Magluta’s
intent in paying his attorneys was at least in part to cover up
the fact that the payments derived from Magluta’s drug
proceeds.” Id. at 808. Here, the defendants moved money
directly from the drug trafficker in Canada to the drug
30 UNITED STATES V. SINGH

suppliers in Los Angeles. They did not engage in
unnecessarily convoluted transactions from which one could
infer an intent to conceal a listed attribute of the funds.

The facts of our case are far more similar to those in
United States v. Garcia, 587 F.3d 509 (2d Cir. 2009). There,
the Second Circuit reversed a defendant’s conviction for
conspiracy to commit money laundering in violation of
18 U.S.C. § 1956(a)(1)(B)(i). The financial transaction at
issue involved transferring $2.2 million in cash by truck
from the East Coast to California or Texas to pay a debt
owed to the drug supplier. The defendant was the truck
driver hired to make the trip. Relying on Cuellar, the court
found insufficient proof that a purpose of the transaction was
to conceal a listed attribute of the funds. 587 F.3d at 518–
19. The court rejected the government’s argument that such
a purpose could be inferred from the chosen method of
transfer (one that left no paper trail) and the steps taken by
the defendant to conceal the transaction from the authorities.
“At bottom,” the court concluded, “the purpose of the
transaction here, as in Cuellar, was merely to pay for
narcotics.” Id. at 519; see also United States v. Ness,
565 F.3d 73, 76–78 (2d Cir. 2009).

I would reach the same conclusion in this case. Because
the government failed to prove that the hawala transfers were
designed to conceal or disguise a listed attribute of the funds,
Singh’s conviction for conspiracy to commit money
laundering should be reversed.