In the
United States Court of Appeals
For the Seventh Circuit
____________________
No. 16-3572
UNITED STATES OF AMERICA,
Plaintiff-Appellee,
v.
CARNELL KING,
Defendant-Appellant.
____________________
Appeal from the United States District Court for the
Northern District of Illinois, Eastern Division.
No. 14 CR 353 — Robert W. Gettleman, Judge.
____________________
ARGUED MAY 31, 2017 — DECIDED JUNE 30, 2017
____________________
Before KANNE, SYKES, and HAMILTON, Circuit Judges.
HAMILTON, Circuit Judge. Defendant Carnell King appeals
his below-guideline sentence. Since he pled guilty and the dis-
trict court’s guideline calculation was admittedly correct, it is
not surprising that we affirm the sentence. We issue a prece-
dential opinion in the case, however, because King has raised
a novel argument about the relationship between the Sentenc-
ing Guidelines and the statute instructing sentencing judges
2 No. 16-3572
on what to consider in making their decisions, 18 U.S.C.
§ 3553(a).
The district judge did exactly what he was supposed to do
in this case: calculate the correct offense level and criminal
history category under the Guidelines, then step back and use
his independent judgment under § 3553(a) to impose a sen-
tence tailored to the individual offender and his crimes. See
Gall v. United States, 552 U.S. 38, 49–50 (2007). King argues,
however, that the “parsimony principle” in § 3553(a), which
instructs the court to impose a sentence “sufficient, but not
greater than necessary,” to serve the statutory purposes of
sentencing, requires an adjustment of the applicable guideline
calculations themselves. In support, he cites a tentative sug-
gestion from a non-precedential Sixth Circuit decision. We re-
ject his argument, which would make post-Booker federal sen-
tencing even more complex than it already is, but without
gaining any apparent benefit in terms of more just sentences.
I. The Crimes and the Punishment
King is an unsuccessful fraudster. Between 2012 and 2014,
he obtained personal identifying information (e.g., addresses,
dates of birth, and Social Security numbers) for more than 100
people, including the Director of the National Security
Agency. King used this information to create and use, or to
attempt to use, 185 access devices (e.g., credit and debit
cards). He also prepared and submitted 62 false tax refund
claims in names other than his own. Thankfully, King was not
very successful in his endeavors. Reported actual losses from
his crimes totaled only $10,980 ($367 from credit cards and
$10,613 in false tax returns).
No. 16-3572 3
King was first arrested in June 2014 and charged with ac-
cess device fraud. He was not detained before trial, but one
condition of his release was that he not commit a federal
crime. In November 2014, though, King was arrested again
and charged with wire fraud. King had resumed his activities
and attempted to use credit card accounts of various individ-
uals less than two months after his initial arrest. His pretrial
release was revoked and he was detained. King eventually
pled guilty to five counts, including a charge of aggravated
identity theft under 18 U.S.C. § 1028A(a)(1), which requires a
minimum sentence of 24 months in prison consecutive to any
other sentence. The district court sentenced King to concur-
rent terms of 24 and 30 months in prison on three access de-
vice fraud counts under 18 U.S.C. § 1029(a) and § 1029(b)(1)
and the fraudulent tax refund count under 18 U.S.C. § 287.
That sentence was below the applicable guideline range. The
court added the mandatory consecutive 24 months for aggra-
vated identity theft for a total sentence of 54 months.
On appeal, King challenges his sentence on the three ac-
cess device fraud counts and the fraudulent tax refund count.
We start with the guideline calculation. Because all the crimes
were so closely related, they were grouped together under
U.S.S.G. § 3D1.2 for guideline calculation purposes. Under
§ 2B1.1, the Guideline for fraud offenses, the base offense level
was six, and two levels were added based on the number of
victims. The Guideline for fraud offenses makes the amount
of loss a major factor in the calculation. The general rule is that
the amount of loss is the greater of actual or intended loss.
§ 2B1.1 note 3. For the multiple crimes in King’s case, the ac-
tual loss was $10,980. The additional intended loss, however,
was $195,948 from 62 fraudulent tax refund claims and
$92,500 based on possession of 185 fake credit and debit cards,
4 No. 16-3572
derived from § 2B1.1 note 3(F)(i), which instructs that the loss
amount shall be not less than $500 per device. 1
The total loss for guideline purposes was $288,448. Be-
cause that was more than $250,000 but less than $550,000,
twelve levels were added to King’s offense level.
§ 2B1.1(b)(1)(G). King has not contested the accuracy of the
loss calculation in the district court or on appeal. With other
adjustments for obstruction of justice and King’s criminal his-
tory of Category II, the final guideline range for King was 46
to 57 months in prison, plus the mandatory consecutive 24
months for aggravated identity theft. 2
1 The $500 loss amount in note 3(F)(i) was inserted into the Guidelines
in 2000. U.S.S.G. supp. app. C amend. 596. The rule was created as a re-
sponse to the Wireless Telephone Protection Act (WTPA), which in-
structed the Sentencing Commission to “provide an appropriate penalty
for offenses involving the cloning of wireless telephones,” Pub. L. No. 105-
172, § 2(e)(1), 112 Stat. 53, 55 (1998), as well as a response to the Identity
Theft and Assumption Deterrence Act, Pub. L. No. 105-318, 112 Stat. 3007
(1998), which has overlapping statutory definitions with the WTPA.
U.S.S.G. supp. app. C, amend. 596. In response, the Commission reviewed
the loss amounts created by identity theft crimes and determined an ade-
quate measure for establishing penalties. The Commission reasoned that
its “research and data supported increasing the minimum loss amount …
from $100 to $500 per access device.” Id. The prior minimum loss amount
of $100 per “stolen credit card[]” had been included in the first iteration of
the Guidelines. U.S.S.G. §2B1.1 note 4 (1987).
2 If only the actual loss amount of $10,980 had been used to calculate
the offense level, there would have been a two-level increase for a loss
exceeding $6,500 but less than $15,000. U.S.S.G. § 2B1.1(b)(1)(B). Without
the extra $92,500 loss calculation for the access devices, the intended loss
of the IRS claims alone would have been $195,948, resulting in a ten-level
increase. U.S.S.G. § 2B1.1(b)(1)(F).
No. 16-3572 5
The district judge concluded, however, that the guideline
range overstated King’s culpability. The judge noted in partic-
ular the defense argument that so much of the loss amount
was driven by “theoretical” rather than actual loss. The judge
provided a thoughtful explanation of the sentence that con-
sidered a variety of aggravating and mitigating factors, in-
cluding King’s continued crimes while on pretrial release, the
harm to the victims, the conditions of King’s pretrial deten-
tion, and his family situation. In the end, the court imposed a
below-guideline sentence of 30 months plus the mandatory
consecutive 24 months for a total sentence of 54 months in
prison.
II. The Relationship Between the Guidelines and § 3553(a)
So far, then, it is hard to see what King is appealing. The
district court calculated the guideline range accurately, as
King agrees, and then used its power and responsibility un-
der § 3553(a) to impose a below-guideline sentence. That pro-
cedure follows the guidance of the Supreme Court, this court,
and the Sentencing Commission. See Gall, 552 U.S. at 49–50;
Rita v. United States, 551 U.S. 338, 351 (2007); United States v.
Pennington, 667 F.3d 953, 956 (7th Cir. 2012); U.S.S.G. § 1B1.1.
And substantively, the result was a below-guideline sentence
that we must presume was not unreasonably harsh. See Rita,
551 U.S. at 347.
King argues, however, that the guideline calculation in
this case violates the statute, § 3553(a), and its parsimony
principle. To understand the argument, it may help to point
out what King is not arguing. He is not arguing that the dis-
trict court erred in calculating the guideline range according
to the Guidelines themselves. He also is not arguing that he
was prevented from arguing in the district court for a below-
6 No. 16-3572
guideline variance under § 3553(a). He made that argument,
and the court agreed.
King is arguing instead for a more subtle nuance: that the
guideline instruction to use the $500 minimum loss per access
device is contrary to the statutory parsimony principle. He ar-
gues that the statutory parsimony principle should allow a
defendant to argue that the guideline calculation itself should
be lowered before the court goes on to the § 3553(a) analysis. He
wants this new intermediate step added to the sentencing
process, requiring the court to consider the parsimony princi-
ple twice, first in considering whether to modify an otherwise
correct guideline calculation itself, and second in exercising
its final sentencing authority under § 3553(a).
King’s proposal has no apparent basis in the statute. To
support his argument, he relies on a suggestion in United
States v. Lyles, 506 Fed. Appx. 440, 445–46 (6th Cir. 2012). That
case also involved a loss calculation based on the $500 per ac-
cess device minimum loss amount in note 3(F)(i). The Sixth
Circuit affirmed the defendant’s sentence based on the $500
minimum but noted: “Theoretically, the $500 fictional amount
should have to pass muster under the parsimony provision of
18 U.S.C. § 3553(a) … . A rule requiring courts to automati-
cally double a sentence based on a fictional loss multiplier is
a rule that may well produce a sentence greater than neces-
sary to achieve punishment’s aims.” Id. at 445. Because the
point had not been argued or briefed, the Sixth Circuit noted
the “problem” but did not decide it. Id. at 446. Judge
McKeague dissented from that dictum. Id. at 454. More re-
cently, the Sixth Circuit seems to have rejected the suggestion
in a precedential decision, United States v. Moon, 808 F.3d 1085,
1092–93 (6th Cir. 2015), where the court affirmed a sentence
No. 16-3572 7
based on the $500 per access device minimum loss, finding
that the minimum as applied in Moon was not substantively
unreasonable.
Our disagreement with the Lyles dictum focuses on one
word: “automatically.” The panel majority wrote: “A rule re-
quiring courts to automatically double a sentence based on a
fictional loss multiplier is a rule that may well produce a sen-
tence greater than necessary … .” 506 Fed. Appx. at 445. As a
general rule, we might be inclined to agree with that state-
ment, but the critical point is that nothing in the Guidelines
requires anything “automatically.” They are advisory. See
Beckles v. United States, 580 U.S. —, 137 S. Ct. 886 (2017); United
States v. Booker, 543 U.S. 220 (2005). The sentencing judge must
evaluate the advice of the Guidelines and then must use his
or her independent judgment under § 3553(a). In fact, the sen-
tencing judge is not even allowed to presume that a guideline
sentence is a reasonable one. Rita, 551 U.S. at 351, citing Booker,
543 U.S. at 259–60. The parsimony principle in the statute—
“impose a sentence sufficient, but not greater than neces-
sary”—applies when the sentencing judge reaches the final
decision under § 3553(a). It does not provide a basis, at least
in an individual sentencing proceeding, for modifying the
guideline calculation itself.
To be clear, a defendant is always free to argue that the
Guidelines, taken as a whole or when particular provisions
are examined, recommend an unduly harsh sentence in his
case. That is as true for the $500 per access device minimum
loss as for any other provision in the Guidelines. The appro-
priate point to address such arguments is after the district
court has applied the Guidelines to determine a final guide-
line range, using the court’s discretion to impose a different
8 No. 16-3572
(“variant”) sentence under § 3553(a). The parsimony principle
does not require the district judge to consider the same argu-
ment twice, once in a novel adjustment to the guideline calcu-
lation itself and again under § 3553(a).
King’s proposal for further consideration of the parsimony
principle would also conflict with a long line of our post-
Booker decisions holding that a district judge is not required
to duplicate the efforts of the Sentencing Commission. We
have repeatedly held that when a defendant argues that a par-
ticular guideline provision is unduly harsh as a categorical
matter (crack-powder cocaine ratios and numerous child por-
nography factors are common targets), a judge who chooses
to follow the Guidelines is not required, in effect, to repeat the
work of the Sentencing Commission by evaluating the overall
policy issues and general reasonableness of the provision.
E.g., United States v. Freeman, 843 F.3d 315, 317–18 (7th Cir.
2016); United States v. Rosales, 813 F.3d 634, 637–38 (7th Cir.
2016); United States v. Estrada-Mederos, 784 F.3d 1086, 1088 (7th
Cir. 2015) (collecting cases); United States v. Schmitz, 717 F.3d
536, 542 (7th Cir. 2013) (collecting cases). The judge may
choose to address and accept such arguments, of course, as
the Supreme Court made clear in Kimbrough v. United States,
552 U.S. 85, 109–10 (2007), relying upon the parsimony prin-
ciple of § 3553(a), but the judge need not do so.
The parsimony principle in § 3553(a) is an important and
binding instruction from Congress. A sentencing court takes
it into account sufficiently when the court considers whether
and to what extent to accept the advice provided by the Sen-
tencing Guidelines in a particular case. Judge Gettleman did
so here and imposed a sentence that was thoughtful and
sound. The judgment of the district court is AFFIRMED.