KIMBROUGH v. UNITED STATES

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
85 OCTOBER TERM, 2007
Syllabus
KIMBROUGH v. UNITED STATES
certiorari to the united states court of appeals for
the fourth circuit
No. 06–6330. Argued October 2, 2007—Decided December 10, 2007
Under the statute criminalizing the manufacture and distribution of co
caine, 21 U. S. C. § 841, and the relevant Federal Sentencing Guidelines,
a drug trafficker dealing in crack cocaine is subject to the same sentence
as one dealing in 100 times more powder cocaine. Petitioner Kim
brough pleaded guilty to four offenses: conspiracy to distribute crack
and powder; possession with intent to distribute more than 50 grams of
crack; possession with intent to distribute powder; and possession of a
firearm in furtherance of a drug-trafficking offense. Under the relevant
statutes, Kimbrough’s plea subjected him to a minimum prison term of
15 years and a maximum of life. The applicable advisory Guidelines
range was 228 to 270 months, or 19 to 22.5 years. The District Court
found, however, that a sentence in this range would have been greater
than necessary to accomplish the purposes of sentencing set forth in 18
U. S. C. § 3553(a). In making that determination, the court relied in
part on its view that Kimbrough’s case exemplified the “disproportion
ate and unjust effect that crack cocaine guidelines have in sentencing.”
The court noted that if Kimbrough had possessed only powder cocaine,
his Guidelines range would have been far lower: 97 to 106 months.
Concluding that the statutory minimum sentence was long enough to
accomplish § 3553(a)’s objectives, the court sentenced Kimbrough to 15
years, or 180 months, in prison. The Fourth Circuit vacated the sen
tence, finding that a sentence outside the Guidelines range is per se
unreasonable when it is based on a disagreement with the sentencing
disparity for crack and powder offenses.
Held:
1. Under United States v. Booker, 543 U. S. 220, the cocaine Guide
lines, like all other Guidelines, are advisory only, and the Fourth Circuit
erred in holding the crack/powder disparity effectively mandatory. A
district judge must include the Guidelines range in the array of factors
warranting consideration, but the judge may determine that, in the par
ticular case, a within-Guidelines sentence is “greater than necessary” to
serve the objectives of sentencing, § 3553(a). In making that determi
nation, the judge may consider the disparity between the Guidelines’
treatment of crack and powder offenses. Pp. 94–110.
(a) Crack and powder cocaine have the same physiological and psy
chotropic effects, but are handled very differently for sentencing pur

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
86 KIMBROUGH v. UNITED STATES
Syllabus
poses. The relevant statutes and Guidelines employ a 100-to-1 ratio
that yields sentences for crack offenses three to six times longer than
those for offenses involving equal amounts of powder. Thus, a major
supplier of powder may receive a shorter sentence than a low-level
dealer who buys powder and converts it to crack. Pp. 94–100.
(1) The crack/powder disparity originated in the Anti-Drug
Abuse Act of 1986 (1986 Act), which created a two-tiered scheme of five
and ten-year mandatory minimum sentences for drug manufacturing
and distribution offenses. Congress apparently adopted the 100-to-1
ratio because it believed that crack, a relatively new drug in 1986, was
significantly more dangerous than powder. Thus, the 1986 Act’s five
year mandatory minimum applies to any defendant accountable for 5
grams of crack or 500 grams of powder, and its ten-year mandatory
minimum applies to any defendant accountable for 50 grams of crack or
5,000 grams of powder. In developing Guidelines sentences for cocaine
offenses, the Sentencing Commission employed the statute’s weight
driven scheme, rather than its usual empirical approach based on past
sentencing practices. The statute itself specifies only two quantities of
each drug, but the Guidelines used the 100-to-1 ratio to set sentences
for a full range of drug quantities. Pp. 95–97.
(2) Based on additional research and experience with the 100-to-1
ratio, the Commission later determined that the crack/powder differen
tial does not meet the objectives of the Sentencing Reform Act and the
1986 Act. The Commission also found the disparity inconsistent with
the 1986 Act’s goal of punishing major drug traffickers more severely
than low-level dealers, and furthermore observed that the differential
fosters a lack of confidence in the criminal justice system because of a
perception that it promotes an unwarranted divergence based on race.
Pp. 97–99.
(3) The Commission has several times sought to achieve a reduc
tion in the crack/powder ratio. Congress rejected a 1995 amendment
to the Guidelines that would have replaced the 100-to-1 ratio with a
1-to-1 ratio, but directed the Commission to propose revision of the ratio
under the relevant statutes and Guidelines. Congress took no action
after the Commission’s 1997 and 2002 reports recommended changing
the ratio. The Commission’s 2007 report again urged Congress to
amend the 1986 Act, but the Commission also adopted an ameliorating
change in the Guidelines. The modest amendment, which became effec
tive on November 1, 2007, yields sentences for crack offenses between
two and five times longer than sentences for equal amounts of powder.
The Commission thus noted that it is only a partial remedy to the prob
lems generated by the crack/powder disparity. Pp. 99–100.

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
87 Cite as: 552 U. S. 85 (2007)
Syllabus
(b) The federal sentencing statute, as modified by Booker, requires
a court to give respectful consideration to the Guidelines, but “permits
the court to tailor the sentence in light of other [§ 3553(a)] concerns as
well,” 543 U. S., at 245–246. The Government contends that the Guide
lines adopting the 100-to-1 ratio are an exception to this general free
dom and offers three arguments in support of its position, each of which
this Court rejects. Pp. 100–108.
(1) The Government argues that the 1986 Act itself prohibits the
Commission and sentencing courts from disagreeing with the 100-to-1
ratio. This position lacks grounding in the statute, which, by its terms,
mandates only maximum and minimum sentences: A person convicted
of possession with intent to distribute five grams or more of crack must
be sentenced to a minimum of 5 years and a maximum of 40. A person
with 50 grams or more of crack must be sentenced to a minimum of ten
years and a maximum of life. The statute says nothing about appro
priate sentences within these brackets, and this Court declines to read
any implicit directive into the congressional silence. See Jama v. Im
migration and Customs Enforcement, 543 U. S. 335, 341. Drawing
meaning from silence is particularly inappropriate here, because Con
gress knows how to direct sentencing practices in express terms. See,
e. g., 28 U. S. C. § 994(h). This cautious reading of the 1986 Act draws
force from Neal v. United States, 516 U. S. 284, which involved different
methods of calculating lysergic acid diethylamide (LSD) weights: The
method applicable in determining statutory minimum sentences com
bined the weight of the pure drug and its carrier medium, while the one
controlling the calculation of Guidelines ranges presumed a lower
weight for the carrier medium. This Court rejected the argument that
the Guidelines and the statute should be interpreted consistently, with
the Guidelines’ presumptive-weight method controlling the mandatory
minimum calculation. Were the Government’s current position correct,
the Guidelines involved in Neal would be in serious jeopardy. The
same reasons alleged to justify reading into the 1986 Act an implicit
command to the Commission and sentencing courts to apply the 100-to-1
ratio to all crack quantities could be urged in support of an argument
that the 1986 Act requires the Commission to include the full weight of
the carrier medium in calculating LSD weights. Yet Neal never ques
tioned the Guidelines’ validity, and in fact endorsed the Commission’s
freedom to adopt a new method. If the 1986 Act does not require the
Commission to adhere to the 1986 Act’s method for determining LSD
weights, it does not require the Commission—or, after Booker, sentenc
ing courts—to adhere to the 100-to-1 ratio for crack quantities other
than those triggering the statutory mandatory minimum sentences.
Pp. 102–105.

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
88 KIMBROUGH v. UNITED STATES
Syllabus
(2) The Government also argues that Congress made clear, in
disapproving the Commission’s 1995 proposed Guidelines amendment,
that the 1986 Act required the Commission and courts to respect the
100-to-1 ratio. But nothing in Congress’ 1995 action suggested that
crack sentences must exceed powder sentences by a ratio of 100 to 1.
To the contrary, Congress required the Commission to recommend a
revision of the ratio. The Government argues that, by calling for rec
ommendations to change both the statute and the Guidelines, Congress
meant to bar any Guidelines alteration in advance of congressional ac
tion. But the more likely reading is that Congress sought proposals to
amend both the statute and the Guidelines because the Commission’s
criticisms of the 100-to-1 ratio concerned the exorbitance of the crack/
powder disparity in both contexts. Moreover, as a result of the 2007
amendment, which Congress did not disapprove or modify, the Guide
lines now deviate from the statute’s 100-to-1 ratio, advancing a ratio
that varies (at different offense levels) between 25 to 1 and 80 to 1.
Pp. 105–106.
(3) Finally, the Government argues that if district courts are
free to deviate from the Guidelines based on disagreements with the
crack/powder ratio, “unwarranted sentence disparities,” 18 U. S. C.
§ 3553(a)(6), will ensue. The Government claims that, because sentenc
ing courts remain bound by the 1986 Act’s mandatory minimum sen
tences, deviations from the 100-to-1 ratio could result in sentencing
“cliffs” around quantities triggering the mandatory minimums. For ex
ample, a district court could grant a sizable downward variance to a
defendant convicted of distributing 49 grams of crack, but would be
required by the statutory minimum to impose a much higher sentence
for only 1 additional gram. The LSD Guidelines approved in Neal,
however, create a similar risk of sentencing “cliffs.” The Government
also maintains that, if district courts are permitted to vary from the
Guidelines based on their disagreement with the crack/powder disparity,
defendants will receive markedly different sentences depending on the
particular judge drawn for sentencing. While uniformity remains an
important sentencing goal, Booker recognized that some departures
from uniformity were a necessary cost of the remedy that decision
adopted. And as to crack sentences in particular, possible variations
among district courts are constrained by the 1986 Act’s mandatory mini
mums. Moreover, to the extent that the Government correctly identi
fies risks of “unwarranted sentence disparities” within the meaning of
§ 3353(a)(6), the proper solution is for district courts to take account of
sentencing practices in other courts and the “cliffs” resulting from the
statutory mandatory minimum sentences and weigh these disparities

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
89 Cite as: 552 U. S. 85 (2007)
Syllabus
against the other § 3553(a) factors and any unwarranted disparities cre
ated by the crack/powder ratio itself. Pp. 106–108.
(c) Booker rendered the Sentencing Guidelines advisory, 543 U. S.,
at 245, but preserved a key role for the Sentencing Commission. In the
ordinary case, the Commission’s recommendation of a sentencing range
will “reflect a rough approximation of sentences that might achieve
§ 3553(a)’s objectives.” Rita v. United States, 551 U. S. 338, 350. The
sentencing judge, on the other hand, is “in a superior position to find
facts and judge their import under § 3553(a)” in each particular case.
Gall v. United States, ante, at 51 (internal quotation marks omitted).
In light of these discrete institutional strengths, a district court’s deci
sion to vary from the advisory Guidelines may attract greatest respect
when the sentencing judge finds a particular case “outside the ‘heart
land’ to which the Commission intends individual Guidelines to apply.”
Rita, 551 U. S., at 351. On the other hand, while the Guidelines are no
longer binding, closer review may be in order when the sentencing judge
varies from the Guidelines based solely on the judge’s view that the
Guidelines range “fails properly to reflect § 3553(a) considerations” even
in a mine-run case. Ibid. The crack cocaine Guidelines, however,
present no occasion for elaborative discussion of this matter because
those Guidelines do not exemplify the Commission’s exercise of its char
acteristic institutional role. Given the Commission’s departure from its
empirical approach in formulating the crack Guidelines and its subse
quent criticism of the crack/powder disparity, it would not be an abuse
of discretion for a district court to conclude when sentencing a particular
defendant that the crack/powder disparity yields a sentence “greater
than necessary” to achieve § 3553(a)’s purposes, even in a mine-run
case. Pp. 108–110.
2. The 180-month sentence imposed on Kimbrough should survive ap
pellate inspection. The District Court began by properly calculating
and considering the advisory Guidelines range. It then addressed the
relevant § 3553(a) factors, including the Sentencing Commission’s re
ports criticizing the 100-to-1 ratio. Finally, the court did not purport
to establish a ratio of its own, but appropriately framed its final deter
mination in line with § 3553(a)’s overarching instruction to “impose a
sentence sufficient, but not greater than necessary,” to accomplish the
sentencing goals advanced in § 3553(a)(2). The court thus rested its
sentence on the appropriate considerations and “committed no proce
dural error,” Gall, ante, at 56. Kimbrough’s sentence was 4.5 years
below the bottom of the Guidelines range. But in determining that 15
years was the appropriate prison term, the District Court properly
homed in on the particular circumstances of Kimbrough’s case and ac
corded weight to the Sentencing Commission’s consistent and emphatic

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
90 KIMBROUGH v. UNITED STATES
Opinion of the Court
position that the crack/powder disparity is at odds with § 3553(a). Giv
ing due respect to the District Court’s reasoned appraisal, a reviewing
court could not rationally conclude that the 4.5-year sentence reduction
Kimbrough received qualified as an abuse of discretion. Pp. 110–111.
174 Fed. Appx. 798, reversed and remanded.
Ginsburg, J., delivered the opinion of the Court, in which Roberts,
C. J., and Stevens, Scalia, Kennedy, Souter, and Breyer, JJ., joined.
Scalia, J., filed a concurring opinion, post, p. 112. Thomas, J., post, p. 114,
and Alito, J., post, p. 116, filed dissenting opinions.
Michael S. Nachmanoff argued the cause for petitioner.
With him on the briefs were Frances H. Pratt, Geremy C.
Kamens, and Kenneth P. Troccoli.
Deputy S olici tor General D reeben argued the cause
for the United States. With him on the brief were Solici
tor General Clement, Assistant Attorney General Fisher,
Kannon K. Shanmugam, Nina Goodman, and Jeffrey P.
Singdahlsen.*
Justice Ginsburg delivered the opinion of the Court.
This Court’s remedial opinion in United States v. Booker,
543 U. S. 220, 244 (2005), instructed district courts to read
the United States Sentencing Guidelines as “effectively advi
sory,” id., at 245. In accord with 18 U. S. C. § 3553(a), the
Guidelines, formerly mandatory, now serve as one factor
among several courts must consider in determining an appro
priate sentence. Booker further instructed that “reason
*Briefs of amici curiae urging reversal were filed for the American
Civil Liberties Union by Adam B. Wolf, Graham A. Boyd, and Steven R.
Shapiro; for Federal Public and Community Defenders et al. by Mark
Osler, Carlos A. Williams, Henry J. Bemporad, Brett G. Sweitzer, and
David L. McColgin; for the NAACP Legal Defense and Educational Fund,
Inc., by Ian Heath Gershengorn, Theodore M. Shaw, Jacqueline A. Ber
rien, Christina Swarns, and Johanna Steinberg; for the National Associa
tion of Criminal Defense Lawyers by Miguel A. Estrada, David Debold,
and Peter Goldberger; and for the Sentencing Project et al. by Matthew
M. Shors and Pammela Quinn.

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
Cite as: 552 U. S. 85 (2007) 91
Opinion of the Court
ableness” is the standard controlling appellate review of the
sentences district courts impose.
Under the statute criminalizing the manufacture and dis
tribution of crack cocaine, 21 U. S. C. § 841, and the relevant
Guidelines prescription, § 2D1.1, a drug trafficker dealing in
crack cocaine is subject to the same sentence as one dealing
in 100 times more powder cocaine. The question here pre
sented is whether, as the Court of Appeals held in this case,
“a sentence . . . outside the guidelines range is per se unrea
sonable when it is based on a disagreement with the sentenc
ing disparity for crack and powder cocaine offenses.” 174
Fed. Appx. 798, 799 (CA4 2006) (per curiam). We hold that,
under Booker, the cocaine Guidelines, like all other Guide
lines, are advisory only, and that the Court of Appeals erred
in holding the crack/powder disparity effectively mandatory.
A district judge must include the Guidelines range in the
array of factors warranting consideration. The judge may
determine, however, that, in the particular case, a within-
Guidelines sentence is “greater than necessary” to serve the
objectives of sentencing. 18 U. S. C. § 3553(a) (2000 ed. and
Supp. V). In making that determination, the judge may
consider the disparity between the Guidelines’ treatment of
crack and powder cocaine offenses.
I
In September 2004, petitioner Derrick Kimbrough was in
dicted in the United States District Court for the Eastern
District of Virginia and charged with four offenses: conspir
acy to distribute crack and powder cocaine; possession with
intent to distribute more than 50 grams of crack cocaine; pos
session with intent to distribute powder cocaine; and posses
sion of a firearm in furtherance of a drug-trafficking offense.
Kimbrough pleaded guilty to all four charges.
Under the relevant statutes, Kimbrough’s plea subjected
him to an aggregate sentence of 15 years to life in prison: 10
years to life for the three drug offenses, plus a consecutive

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
92 KIMBROUGH v. UNITED STATES
Opinion of the Court
term of 5 years to life for the firearm offense.1 In order
to determine the appropriate sentence within this statutory
range, the District Court first calculated Kimbrough’s sen
tence under the advisory Sentencing Guidelines.2 Kim
brough’s guilty plea acknowledged that he was accountable
for 56 grams of crack cocaine and 92.1 grams of powder co
caine. This quantity of drugs yielded a base offense level of
32 for the three drug charges. See United States Sentenc
ing Commission, Guidelines Manual § 2D1.1(c) (Nov. 2004)
(USSG). Finding that Kimbrough, by asserting sole culpa
bility for the crime, had testified falsely at his codefendant’s
trial, the District Court increased his offense level to 34.
See § 3C1.1. In accord with the presentence report, the
court determined that Kimbrough’s criminal history cate
gory was II. An offense level of 34 and a criminal history
category of II yielded a Guidelines range of 168 to 210
months for the three drug charges. See id., ch. 5, pt. A,
Sentencing Table. The Guidelines sentence for the firearm
offense was the statutory minimum, 60 months. See
§ 2K2.4(b). Kimbrough’s final advisory Guidelines range
was thus 228 to 270 months, or 19 to 22.5 years.
A sentence in this range, in the District Court’s judgment,
would have been “greater than necessary” to accomplish the
1 The statutory range for possession with intent to distribute more than
50 grams of crack is ten years to life. See 21 U. S. C. § 841(b)(1)(A)(iii)
(2000 ed. and Supp. V). The same range applies to the conspiracy offense.
See § 846 (2000 ed.). The statutory range for possession with intent to
distribute powder cocaine is 0 to 20 years. See § 841(b)(1)(C) (Supp. V).
Finally, the statutory range for possession of a firearm in furtherance
of a drug-trafficking offense is five years to life. See 18 U. S. C.
§ 924(c)(1)(A)(i). The sentences for the three drug crimes may run con
currently, see § 3584(a), but the sentence for the firearm offense must be
consecutive, see § 924(c)(1)(A).
2 Kimbrough was sentenced in April 2005, three months after our deci
sion in United States v. Booker, 543 U. S. 220 (2005), rendered the Guide
lines advisory. The District Court employed the version of the Guidelines
effective November 1, 2004.

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
Cite as: 552 U. S. 85 (2007) 93
Opinion of the Court
purposes of sentencing set forth in 18 U. S. C. § 3553(a).
App. 72. As required by § 3553(a), the court took into ac
count the “nature and circumstances” of the offense and Kim
brough’s “history and characteristics.” Id., at 72–73. The
court also commented that the case exemplified the “dispro
portionate and unjust effect that crack cocaine guidelines
have in sentencing.” Id., at 72. In this regard, the court
contrasted Kimbrough’s Guidelines range of 228 to 270
months with the range that would have applied had he been
accountable for an equivalent amount of powder cocaine: 97
to 106 months, inclusive of the 5-year mandatory minimum
for the firearm charge, see USSG § 2D1.1(c); id., ch. 5, pt. A,
Sentencing Table. Concluding that the statutory minimum
sentence was “clearly long enough” to accomplish the objec
tives listed in § 3553(a), the court sentenced Kimbrough to
15 years, or 180 months, in prison plus 5 years of supervised
release. App. 74–75.3
In an unpublished per curiam opinion, the Fourth Circuit
vacated the sentence. Under Circuit precedent, the Court
of Appeals observed, a sentence “outside the guidelines
range is per se unreasonable when it is based on a disagree
ment with the sentencing disparity for crack and powder co
caine offenses.” 174 Fed. Appx., at 799 (citing United States
v. Eura, 440 F. 3d 625, 633–634 (CA4 2006)).
We granted certiorari, 551 U. S. 1113 (2007), to determine
whether the crack/powder disparity adopted in the United
States Sentencing Guidelines has been rendered “advisory”
by our decision in Booker.4
3 The prison sentence consisted of 120 months on each of the three drug
counts, to be served concurrently, plus 60 months on the firearm count, to
be served consecutively.
4 This question has divided the Courts of Appeals. Compare United
States v. Pickett, 475 F. 3d 1347, 1355–1356 (CADC 2007) (District Court
erred when it concluded that it had no discretion to consider the crack/
powder disparity in imposing a sentence), and United States v. Gunter,
462 F. 3d 237, 248–249 (CA3 2006) (same), with United States v. Leatch,
482 F. 3d 790, 791 (CA5 2007) (per curiam) (sentencing court may not

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
94 KIMBROUGH v. UNITED STATES
Opinion of the Court
II
We begin with some background on the different treat
ment of crack and powder cocaine under the federal sentenc
ing laws. Crack and powder cocaine are two forms of the
same drug. Powder cocaine, or cocaine hydrochloride, is
generally inhaled through the nose; it may also be mixed
with water and injected. See United States Sentencing
Commission, Special Report to Congress: Cocaine and Fed
eral Sentencing Policy 5, 12 (Feb. 1995), available at http://
www.ussc.gov/crack/exec.htm (hereinafter 1995 Report).
(All Internet materials as visited Dec. 7, 2007, and included
in Clerk of Court’s case file.) Crack cocaine, a type of co
caine base, is formed by dissolving powder cocaine and bak
ing soda in boiling water. Id., at 14. The resulting solid is
divided into single-dose “rocks” that users smoke. Ibid.
The active ingredient in powder and crack cocaine is the
same. Id., at 9. The two forms of the drug also have the
same physiological and psychotropic effects, but smoking
crack cocaine allows the body to absorb the drug much faster
than inhaling powder cocaine, and thus produces a shorter,
more intense high. Id., at 15–19.5
Although chemically similar, crack and powder cocaine are
handled very differently for sentencing purposes. The 100
to-1 ratio yields sentences for crack offenses three to six
times longer than those for powder offenses involving equal
amounts of drugs. See United States Sentencing Commis
sion, Report to Congress: Cocaine and Federal Sentencing
impose a sentence outside the Guidelines range based on its disagreement
with the crack/powder disparity), United States v. Johnson, 474 F. 3d 515,
522 (CA8 2007) (same), United States v. Castillo, 460 F. 3d 337, 361 (CA2
2006) (same), United States v. Williams, 456 F. 3d 1353, 1369 (CA11 2006)
(same), United States v. Miller, 450 F. 3d 270, 275–276 (CA7 2006) (same),
United States v. Eura, 440 F. 3d 625, 633–634 (CA4 2006) (same), and
United States v. Pho, 433 F. 3d 53, 62–63 (CA1 2006) (same).
5 Injecting powder cocaine produces effects similar to smoking crack co
caine, but very few powder users inject the drug. See 1995 Report 18.

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
95 Cite as: 552 U. S. 85 (2007)
Opinion of the Court
Policy iv (May 2002), available at http://www.ussc.gov/
r_congress/02crack/2002crackrpt.pdf (hereinafter 2002 Re
port).6 This disparity means that a major supplier of pow
der cocaine may receive a shorter sentence than a low-level
dealer who buys powder from the supplier but then converts
it to crack. See 1995 Report 193–194.
A
The crack/powder disparity originated in the Anti-Drug
Abuse Act of 1986 (1986 Act), 100 Stat. 3207. The 1986 Act
created a two-tiered scheme of five- and ten-year mandatory
minimum sentences for drug manufacturing and distribution
offenses. Congress sought “to link the ten-year mandatory
minimum trafficking prison term to major drug dealers and
to link the five-year minimum term to serious traffickers.”
1995 Report 119. The 1986 Act uses the weight of the drugs
involved in the offense as the sole proxy to identify “major”
and “serious” dealers. For example, any defendant respon
sible for 100 grams of heroin is subject to the five-year man
datory minimum, see 21 U. S. C. § 841(b)(1)(B)(i) (2000 ed. and
Supp. V), and any defendant responsible for 1,000 grams of
heroin is subject to the ten-year mandatory minimum, see
§ 841(b)(1)(A)(i).
Crack cocaine was a relatively new drug when the 1986
Act was signed into law, but it was already a matter of great
public concern: “Drug abuse in general, and crack cocaine
in particular, had become in public opinion and in mem
bers’ minds a problem of overwhelming dimensions.” 1995
Report 121. Congress apparently believed that crack was
significantly more dangerous than powder cocaine in that:
(1) crack was highly addictive; (2) crack users and dealers
were more likely to be violent than users and dealers of
6 As explained in Part II–C, infra, the Sentencing Commission amended
the Guidelines and reduced sentences for crack offenses effective Novem
ber 1, 2007. Except as noted, this opinion refers to the 2004 Guidelines
in effect at the time of Kimbrough’s sentencing.

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
96 KIMBROUGH v. UNITED STATES
Opinion of the Court
other drugs; (3) crack was more harmful to users than pow
der, particularly for children who had been exposed by their
mothers’ drug use during pregnancy; (4) crack use was espe
cially prevalent among teenagers; and (5) crack’s potency and
low cost were making it increasingly popular. See 2002
Report 90.
Based on these assumptions, the 1986 Act adopted a “100
to-1 ratio” that treated every gram of crack cocaine as the
equivalent of 100 grams of powder cocaine. The 1986 Act’s
five-year mandatory minimum applies to any defendant ac
countable for 5 grams of crack or 500 grams of powder, 21
U. S. C. § 841(b)(1)(B)(ii), (iii); its ten-year mandatory mini
mum applies to any defendant accountable for 50 grams of
crack or 5,000 grams of powder, § 841(b)(1)(A)(ii), (iii).
While Congress was considering adoption of the 1986 Act,
the Sentencing Commission was engaged in formulating the
Sentencing Guidelines.7 In the main, the Commission devel
oped Guidelines sentences using an empirical approach based
on data about past sentencing practices, including 10,000 pre
sentence investigation reports. See USSG § 1A.1, intro.
comment., pt. A, ¶ 3. The Commission “modif[ied] and ad
just[ed] past practice in the interests of greater rationality,
avoiding inconsistency, complying with congressional in
structions, and the like.” Rita v. United States, 551 U. S.
338, 349 (2007).
The Commission did not use this empirical approach in de
veloping the Guidelines sentences for drug-trafficking of
fenses. Instead, it employed the 1986 Act’s weight-driven
scheme. The Guidelines use a drug quantity table based on
drug type and weight to set base offense levels for drug
trafficking offenses. See USSG § 2D1.1(c). In setting of
7 Congress created the Sentencing Commission and charged it with
promulgating the Guidelines in the Sentencing Reform Act of 1984, 98
Stat. 1987, 18 U. S. C. § 3551 et seq. (2000 ed. and Supp. V), but the first
version of the Guidelines did not become operative until November 1987,
see 1995 Report iii–iv.

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
97 Cite as: 552 U. S. 85 (2007)
Opinion of the Court
fense levels for crack and powder cocaine, the Commission,
in line with the 1986 Act, adopted the 100-to-1 ratio. The
statute itself specifies only two quantities of each drug, but
the Guidelines “go further and set sentences for the full
range of possible drug quantities using the same 100-to-1
quantity ratio.” 1995 Report 1. The Guidelines’ drug
quantity table sets base offense levels ranging from 12, for
offenses involving less than 250 milligrams of crack (or 25
grams of powder), to 38, for offenses involving more than 1.5
kilograms of crack (or 150 kilograms of powder). USSG
§ 2D1.1(c).8
B
Although the Commission immediately used the 100-to-1
ratio to define base offense levels for all crack and powder
offenses, it later determined that the crack/powder sentenc
ing disparity is generally unwarranted. Based on additional
research and experience with the 100-to-1 ratio, the Commis
sion concluded that the disparity “fails to meet the sentenc
ing objectives set forth by Congress in both the Sentencing
Reform Act and the 1986 Act.” 2002 Report 91. In a series
of reports, the Commission identified three problems with
the crack/powder disparity.
First, the Commission reported, the 100-to-1 ratio rested
on assumptions about “the relative harmfulness of the two
drugs and the relative prevalence of certain harmful conduct
associated with their use and distribution that more recent
research and data no longer support.” Ibid.; see United
States Sentencing Commission, Report to Congress: Cocaine
and Federal Sentencing Policy 8 (May 2007), available at
http://www.ussc.gov/r_congress/cocaine2007.pdf (hereinafter
2007 Report) (ratio Congress embedded in the statute far
“overstate[s]” both “the relative harmfulness” of crack co
8 An offense level of 12 results in a Guidelines range of 10 to 16 months
for a first-time offender; an offense level of 38 results in a range of 235 to
293 months for the same offender. See USSG ch. 5, pt. A, Sentencing
Table.

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
98 KIMBROUGH v. UNITED STATES
Opinion of the Court
caine, and the “seriousness of most crack cocaine offenses”).
For example, the Commission found that crack is associated
with “significantly less trafficking-related violence . . . than
previously assumed.” 2002 Report 100. It also observed
that “the negative effects of prenatal crack cocaine exposure
are identical to the negative effects of prenatal powder co
caine exposure.” Id., at 94. The Commission furthermore
noted that “the epidemic of crack cocaine use by youth never
materialized to the extent feared.” Id., at 96.
Second, the Commission concluded that the crack/powder
disparity is inconsistent with the 1986 Act’s goal of punishing
major drug traffickers more severely than low-level dealers.
Drug importers and major traffickers generally deal in pow
der cocaine, which is then converted into crack by street
level sellers. See 1995 Report 66–67. But the 100-to-1
ratio can lead to the “anomalous” result that “retail crack
dealers get longer sentences than the wholesale drug distrib
utors who supply them the powder cocaine from which their
crack is produced.” Id., at 174.
Finally, the Commission stated that the crack/powder sen
tencing differential “fosters disrespect for and lack of confi
dence in the criminal justice system” because of a “widely
held perception” that it “promotes unwarranted disparity
based on race.” 2002 Report 103. Approximately 85 per
cent of defendants convicted of crack offenses in federal
court are black; thus the severe sentences required by the
100-to-1 ratio are imposed “primarily upon black offend
ers.” Ibid.
Despite these observations, the Commission’s most recent
reports do not urge identical treatment of crack and powder
cocaine. In the Commission’s view, “some differential in the
quantity-based penalties” for the two drugs is warranted, id.,
at 102, because crack is more addictive than powder, crack
offenses are more likely to involve weapons or bodily injury,
and crack distribution is associated with higher levels of
crime, see id., at 93–94, 101–102. But the 100-to-1 crack/

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
99 Cite as: 552 U. S. 85 (2007)
Opinion of the Court
powder ratio, the Commission concluded, significantly over
states the differences between the two forms of the drug.
Accordingly, the Commission recommended that the ratio be
“substantially” reduced. Id., at viii.
C
The Commission has several times sought to achieve a re
duction in the crack/powder ratio. In 1995, it proposed
amendments to the Guidelines that would have replaced the
100-to-1 ratio with a 1-to-1 ratio. Complementing that
change, the Commission would have installed special en
hancements for trafficking offenses involving weapons or
bodily injury. See Amendments to the Sentencing Guide
lines for United States Courts, 60 Fed. Reg. 25075–25077
(1995). Congress, acting pursuant to 28 U. S. C. § 994(p),9
rejected the amendments. See Pub. L. 104–38, § 1, 109 Stat.
334. Simultaneously, however, Congress directed the Com
mission to “propose revision of the drug quantity ratio of
crack cocaine to powder cocaine under the relevant statutes
and guidelines.” § 2(a)(2), id., at 335.
In response to this directive, the Commission issued re
ports in 1997 and 2002 recommending that Congress change
the 100-to-1 ratio prescribed in the 1986 Act. The 1997 Re
port proposed a 5-to-1 ratio. See United States Sentencing
Commission, Special Report to Congress: Cocaine and Fed
eral Sentencing Policy 2 (Apr. 1997), http://www.ussc.gov/
r_congress/newcrack.pdf. The 2002 Report recommended
lowering the ratio “at least” to 20 to 1. 2002 Report viii.
Neither proposal prompted congressional action.
The Commission’s most recent report, issued in 2007, again
urged Congress to amend the 1986 Act to reduce the 100-to-1
ratio. This time, however, the Commission did not sim
ply await congressional action. Instead, the Commission
9 Subsection 994(p) requires the Commission to submit Guidelines
amendments to Congress and provides that such amendments become ef
fective unless “modified or disapproved by Act of Congress.”

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
100 KIMBROUGH v. UNITED STATES
Opinion of the Court
adopted an ameliorating change in the Guidelines. See 2007
Report 9. The alteration, which became effective on No
vember 1, 2007, reduces the base offense level associated
with each quantity of crack by two levels. See Amendments
to the Sentencing Guidelines for United States Courts, 72
Fed. Reg. 28571–28572 (2007).10 This modest amendment
yields sentences for crack offenses between two and five
times longer than sentences for equal amounts of powder.
See ibid.11 Describing the amendment as “only . . . a partial
remedy” for the problems generated by the crack/powder
disparity, the Commission noted that “[a]ny comprehensive
solution requires appropriate legislative action by Con
gress.” 2007 Report 10.
III
With this history of the crack/powder sentencing ratio in
mind, we next consider the status of the Guidelines tied
to the ratio after our decision in United States v. Booker,
543 U. S. 220 (2005). In Booker, the Court held that the
mandatory Sentencing Guidelines system violated the Sixth
Amendment. See id., at 226–227. The Booker remedial
opinion determined that the appropriate cure was to sever
and excise the provision of the statute that rendered the
10 The amended Guidelines still produce sentencing ranges keyed to the
mandatory minimums in the 1986 Act. Under the pre-2007 Guidelines,
the 5- and 50-gram quantities that trigger the statutory minimums
produced sentencing ranges that slightly exceeded those statutory mini
mums. Under the amended Guidelines, in contrast, the 5- and 50
gram quantities produce “base offense levels corresponding to guideline
ranges that include the statutory mandatory minimum penalties.” 2007
Report 9.
11 The Commission has not yet determined whether the amendment will
be retroactive to cover defendants like Kimbrough. Even under the
amendment, however, Kimbrough’s Guidelines range would be 195 to 218
months—well above the 180-month sentence imposed by the District
Court. See Amendments to the Sentencing Guidelines for United States
Courts, 72 Fed. Reg. 28571–28572 (2007); USSG ch. 5, pt. A, Sentencing
Table.

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
101 Cite as: 552 U. S. 85 (2007)
Opinion of the Court
Guidelines mandatory, 18 U. S. C. § 3553(b)(1) (2000 ed., Supp.
IV).12 This modification of the federal sentencing statute,
we explained, “makes the Guidelines effectively advisory.”
543 U. S., at 245.
The statute, as modified by Booker, contains an over
arching provision instructing district courts to “impose a
sentence sufficient, but not greater than necessary,” to ac
complish the goals of sentencing, including “to reflect the
seriousness of the offense,” “to promote respect for the law,”
“to provide just punishment for the offense,” “to afford ade
quate deterrence to criminal conduct,” and “to protect the
public from further crimes of the defendant.” 18 U. S. C.
§ 3553(a) (2000 ed. and Supp. V). The statute further pro
vides that, in determining the appropriate sentence, the
court should consider a number of factors, including “the na
ture and circumstances of the offense,” “the history and
characteristics of the defendant,” “the sentencing range es
tablished” by the Guidelines, “any pertinent policy state
ment” issued by the Sentencing Commission pursuant to its
statutory authority, and “the need to avoid unwarranted sen
tence disparities among defendants with similar records who
have been found guilty of similar conduct.” Ibid. In sum,
while the statute still requires a court to give respectful con
sideration to the Guidelines, see Gall v. United States, ante,
at 46, 49, Booker “permits the court to tailor the sentence
in light of other statutory concerns as well,” 543 U. S.,
at 245–246.
The Government acknowledges that the Guidelines “are
now advisory” and that, as a general matter, “courts may
vary [from Guidelines ranges] based solely on policy consid
erations, including disagreements with the Guidelines.”
Brief for United States 16; cf. Rita, 551 U. S., at 351 (a dis
12 The remedial opinion also severed and excised the provision of the
statute requiring de novo review of departures from the Guidelines, 18
U. S. C. § 3742(e), because that provision depended on the Guidelines’ man
datory status. Booker, 543 U. S., at 245.

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
102 KIMBROUGH v. UNITED STATES
Opinion of the Court
trict court may consider arguments that “the Guidelines sen
tence itself fails properly to reflect § 3553(a) considerations”).
But the Government contends that the Guidelines adopting
the 100-to-1 ratio are an exception to the “general freedom
that sentencing courts have to apply the [§ 3553(a)] fac
tors.” Brief for United States 16. That is so, according to
the Government, because the ratio is a “specific policy deter
minatio[n] that Congress has directed sentencing courts to
observe.” Id., at 25. The Government offers three argu
ments in support of this position. We consider each in turn.
A
As its first and most heavily pressed argument, the Gov
ernment urges that the 1986 Act itself prohibits the Sentenc
ing Commission and sentencing courts from disagreeing with
the 100-to-1 ratio.13 The Government acknowledges that
the “Congress did not expressly direct the Sentencing Com
mission to incorporate the 100:1 ratio in the Guidelines.”
Brief for United States 33 (brackets and internal quotation
marks omitted). Nevertheless, it asserts that the Act “[i]m
plicit[ly]” requires the Commission and sentencing courts to
apply the 100-to-1 ratio. Id., at 32. Any deviation, the
Government urges, would be “logically incoherent” when
combined with mandatory minimum sentences based on the
100-to-1 ratio. Id., at 33.
This argument encounters a formidable obstacle: It lacks
grounding in the text of the 1986 Act. The statute, by its
terms, mandates only maximum and minimum sentences: A
person convicted of possession with intent to distribute five
grams or more of crack cocaine must be sentenced to a mini
13 The Government concedes that a district court may vary from the
100-to-1 ratio if it does so “based on the individualized circumstance[s]”
of a particular case. Brief for United States 45. But the Government
maintains that the 100-to-1 ratio is binding in the sense that a court may
not give any weight to its own view that the ratio itself is inconsistent
with the § 3553(a) factors.

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
Cite as: 552 U. S. 85 (2007) 103
Opinion of the Court
mum of 5 years and the maximum term is 40 years. A per
son with 50 grams or more of crack cocaine must be sen
tenced to a minimum of ten years and the maximum term is
life. The statute says nothing about the appropriate sen
tences within these brackets, and we decline to read any im
plicit directive into that congressional silence. See Jama v.
Immigration and Customs Enforcement, 543 U. S. 335, 341
(2005) (“We do not lightly assume that Congress has omitted
from its adopted text requirements that it nonetheless in
tends to apply . . . .”). Drawing meaning from silence is
particularly inappropriate here, for Congress has shown that
it knows how to direct sentencing practices in express terms.
For example, Congress has specifically required the Sentenc
ing Commission to set Guidelines sentences for serious recid
ivist offenders “at or near” the statutory maximum. 28
U. S. C. § 994(h). See also § 994(i) (“The Commission shall
assure that the guidelines specify a sentence to a substantial
term of imprisonment” for specified categories of offenders.).
Our cautious reading of the 1986 Act draws force from
Neal v. United States, 516 U. S. 284 (1996). That case
involved different methods of calculating lysergic acid di
ethylamide (LSD) weights, one applicable in determining
statutory minimum sentences, the other controlling the
calculation of Guidelines ranges. The 1986 Act sets manda
tory minimum sentences based on the weight of “a mixture
or substance containing a detectable amount” of LSD. 21
U. S. C. § 841(b)(1)(A)(v), (B)(v). Prior to Neal, we had inter
preted that language to include the weight of the carrier me
dium (usually blotter paper) on which LSD is absorbed even
though the carrier is usually far heavier than the LSD itself.
See Chapman v. United States, 500 U. S. 453, 468 (1991).
Until 1993, the Sentencing Commission had interpreted the
relevant Guidelines in the same way. That year, however,
the Commission changed its approach and “instructed courts
to give each dose of LSD on a carrier medium a constructive
or presumed weight of 0.4 milligrams.” Neal, 516 U. S.,

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
104 KIMBROUGH v. UNITED STATES
Opinion of the Court
at 287 (citing USSG § 2D1.1(c), n. (H) (Nov. 1995)). The
Commission’s change significantly lowered the Guidelines
range applicable to most LSD offenses, but defendants re
mained subject to higher statutory minimum sentences
based on the combined weight of the pure drug and its car
rier medium. The defendant in Neal argued that the re
vised Guidelines and the statute should be interpreted con
sistently and that the “presumptive-weight method of the
Guidelines should also control the mandatory minimum cal
culation.” 516 U. S., at 287. We rejected that argument,
emphasizing that the Commission had not purported to in
terpret the statute and could not in any event overrule our
decision in Chapman. See 516 U. S., at 293–295.
If the Government’s current position were correct, then
the Guidelines involved in Neal would be in serious jeopardy.
We have just recounted the reasons alleged to justify reading
into the 1986 Act an implicit command to the Commission
and sentencing courts to apply the 100-to-1 ratio to all quan
tities of crack cocaine. Those same reasons could be urged
in support of an argument that the 1986 Act requires the
Commission to include the full weight of the carrier medium
in calculating the weight of LSD for Guidelines purposes.
Yet our opinion in Neal never questioned the validity of the
altered Guidelines. To the contrary, we stated: “Entrusted
within its sphere to make policy judgments, the Commission
may abandon its old methods in favor of what it has deemed
a more desirable ‘approach’ to calculating LSD quantities.”
Id., at 295.14 If the 1986 Act does not require the Commis
14 At oral argument, the Government sought to distinguish Neal v.
United States, 516 U. S. 284 (1996), on the ground that the validity of the
amended Guidelines was not before us in that case. See Tr. of Oral
Arg. 25. That is true, but only because the Government did not challenge
the amendment. In fact, the Government’s brief appeared to acknowl
edge that the Commission may legitimately deviate from the policies and
methods embodied in the 1986 Act, even if the deviation produces some
inconsistency. See Brief for United States in Neal v. United States, O. T.
1995, No. 94–9088, p. 26 (“When the Commission’s views about sentencing

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
105 Cite as: 552 U. S. 85 (2007)
Opinion of the Court
sion to adhere to the Act’s method for determining LSD
weights, it does not require the Commission—or, after
Booker, sentencing courts—to adhere to the 100-to-1 ratio
for crack cocaine quantities other than those that trigger the
statutory mandatory minimum sentences.
B
In addition to the 1986 Act, the Government relies on Con
gress’ disapproval of the Guidelines amendment that the
Sentencing Commission proposed in 1995. Congress “not
only disapproved of the 1:1 ratio,” the Government urges; it
also made clear “that the 1986 Act required the Commission
(and sentencing courts) to take drug quantities into account,
and to do so in a manner that respects the 100:1 ratio.”
Brief for United States 35.
It is true that Congress rejected the Commission’s 1995
proposal to place a 1-to-1 ratio in the Guidelines, and that
Congress also expressed the view that “the sentence im
posed for trafficking in a quantity of crack cocaine should
generally exceed the sentence imposed for trafficking in a like
quantity of powder cocaine.” Pub. L. 104–38, § 2(a)(1)(A),
109 Stat. 334. But nothing in Congress’ 1995 reaction to the
Commission-proposed 1-to-1 ratio suggested that crack sen
tences must exceed powder sentences by a ratio of 100 to 1.
To the contrary, Congress’ 1995 action required the Commis
sion to recommend a “revision of the drug quantity ratio of
crack cocaine to powder cocaine.” § 2(a)(2), id., at 335.
The Government emphasizes that Congress required the
Commission to propose changes to the 100-to-1 ratio in both
policy depart from those of Congress, it may become difficult to achieve
entirely consistent sentencing, but that is a matter for Congress, not the
courts, to address.”). Moreover, our opinion in Neal assumed that the
amendment was a legitimate exercise of the Commission’s authority. See
516 U. S., at 294 (noting with apparent approval the Commission’s posi
tion that “the Guidelines calculation is independent of the statutory
calculation”).

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
106 KIMBROUGH v. UNITED STATES
Opinion of the Court
the 1986 Act and the Guidelines. This requirement, the
Government contends, implicitly foreclosed any deviation
from the 100-to-1 ratio in the Guidelines (or by sentencing
courts) in the absence of a corresponding change in the stat
ute. See Brief for United States 35–36. But it does not
follow as the night follows the day that, by calling for recom
mendations to change the statute, Congress meant to bar
any Guidelines alteration in advance of congressional action.
The more likely reading is that Congress sought proposals
to amend both the statute and the Guidelines because the
Commission’s criticisms of the 100-to-1 ratio, see Part II–B,
supra, concerned the exorbitance of the crack/powder dispar
ity in both contexts.
Moreover, as a result of the 2007 amendment, see supra,
at 99–100, the Guidelines now advance a crack/powder ratio
that varies (at different offense levels) between 25 to 1 and
80 to 1. See Amendments to the Sentencing Guidelines for
United States Courts, 72 Fed. Reg. 28571–28572. Adopting
the Government’s analysis, the amended Guidelines would
conflict with Congress’ 1995 action, and with the 1986 Act,
because the current Guidelines ratios deviate from the 100
to-1 statutory ratio. Congress, however, did not disapprove
or modify the Commission-initiated 2007 amendment. Ordi
narily, we resist reading congressional intent into congres
sional inaction. See Bob Jones Univ. v. United States, 461
U. S. 574, 600 (1983). But in this case, Congress failed to act
on a proposed amendment to the Guidelines in a high-profile
area in which it had previously exercised its disapproval
authority under 28 U. S. C. § 994(p). If nothing else, this
tacit acceptance of the 2007 amendment undermines the
Government’s position, which is itself based on implications
drawn from congressional silence.
C
Finally, the Government argues that if district courts are
free to deviate from the Guidelines based on disagreements

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
107 Cite as: 552 U. S. 85 (2007)
Opinion of the Court
with the crack/powder ratio, unwarranted disparities of two
kinds will ensue. See 18 U. S. C. § 3553(a)(6) (sentencing
courts shall consider “the need to avoid unwarranted sen
tence disparities”). First, because sentencing courts remain
bound by the mandatory minimum sentences prescribed in
the 1986 Act, deviations from the 100-to-1 ratio could result
in sentencing “cliffs” around quantities that trigger the man
datory minimums. Brief for United States 33 (internal quo
tation marks omitted). For example, a district court could
grant a sizable downward variance to a defendant convicted
of distributing 49 grams of crack but would be required by
the statutory minimum to impose a much higher sentence on
a defendant responsible for only 1 additional gram. Second,
the Government maintains that, if district courts are permit
ted to vary from the Guidelines based on their disagreement
with the crack/powder disparity, “defendants with identical
real conduct will receive markedly different sentences, de
pending on nothing more than the particular judge drawn
for sentencing.” Id., at 40.
Neither of these arguments persuades us to hold the crack/
powder ratio untouchable by sentencing courts. As to the
first, the LSD Guidelines we approved in Neal create a
similar risk of sentencing “cliffs.” An offender who pos
sesses LSD on a carrier medium weighing ten grams is
subject to the ten-year mandatory minimum, see 21 U. S. C.
§ 841(b)(1)(A)(v), but an offender whose carrier medium
weighs slightly less may receive a considerably lower sen
tence based on the Guidelines’ presumptive-weight method
ology. Concerning the second disparity, it is unquestioned
that uniformity remains an important goal of sentencing.
As we explained in Booker, however, advisory Guidelines
combined with appellate review for reasonableness and ongo
ing revision of the Guidelines in response to sentencing prac
tices will help to “avoid excessive sentencing disparities.”
543 U. S., at 264. These measures will not eliminate varia
tions between district courts, but our opinion in Booker rec

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
108 KIMBROUGH v. UNITED STATES
Opinion of the Court
ognized that some departures from uniformity were a neces
sary cost of the remedy we adopted. See id., at 263 (“We
cannot and do not claim that use of a ‘reasonableness’ stand
ard will provide the uniformity that Congress originally
sought to secure [through mandatory Guidelines].”). And as
to crack cocaine sentences in particular, we note a congres
sional control on disparities: possible variations among dis
trict courts are constrained by the mandatory minimums
Congress prescribed in the 1986 Act.15
Moreover, to the extent that the Government correctly
identifies risks of “unwarranted sentence disparities” within
the meaning of 18 U. S. C. § 3553(a)(6), the proper solution is
not to treat the crack/powder ratio as mandatory. Section
3553(a)(6) directs district courts to consider the need to
avoid unwarranted disparities—along with other § 3553(a)
factors—when imposing sentences. See Gall, ante, at 50,
n. 6, 54. Under this instruction, district courts must take
account of sentencing practices in other courts and the
“cliffs” resulting from the statutory mandatory minimum
sentences. To reach an appropriate sentence, these dispari
ties must be weighed against the other § 3553(a) factors and
any unwarranted disparity created by the crack/powder
ratio itself.
IV
While rendering the Sentencing Guidelines advisory,
Booker, 543 U. S., at 245, we have nevertheless preserved a
key role for the Sentencing Commission. As explained in
Rita and Gall, district courts must treat the Guidelines as
the “starting point and the initial benchmark,” Gall, ante,
at 49. Congress established the Commission to formulate
and constantly refine national sentencing standards. See
Rita, 551 U. S., at 347–350. Carrying out its charge, the
15 The Sentencing Commission reports that roughly 70 percent of crack
offenders are responsible for drug quantities that yield base offense levels
at or only two levels above those that correspond to the statutory mini
mums. See 2007 Report 25.

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
109 Cite as: 552 U. S. 85 (2007)
Opinion of the Court
Commission fills an important institutional role: It has the
capacity courts lack to “base its determinations on empirical
data and national experience, guided by a professional staff
with appropriate expertise.” United States v. Pruitt, 502
F. 3d 1154, 1171 (CA10 2007) (McConnell, J., concurring); see
supra, at 96.
We have accordingly recognized that, in the ordinary case,
the Commission’s recommendation of a sentencing range will
“reflect a rough approximation of sentences that might
achieve § 3553(a)’s objectives.” Rita, 551 U. S., at 350. The
sentencing judge, on the other hand, has “greater familiarity
with . . . the individual case and the individual defendant
before him than the Commission or the appeals court.” Id.,
at 357–358. He is therefore “in a superior position to find
facts and judge their import under § 3553(a)” in each particu
lar case. Gall, ante, at 51 (internal quotation marks omit
ted). In light of these discrete institutional strengths, a dis
trict court’s decision to vary from the advisory Guidelines
may attract greatest respect when the sentencing judge
finds a particular case “outside the ‘heartland’ to which the
Commission intends individual Guidelines to apply.” Rita,
551 U. S., at 351. On the other hand, while the Guidelines
are no longer binding, closer review may be in order when
the sentencing judge varies from the Guidelines based solely
on the judge’s view that the Guidelines range “fails properly
to reflect § 3553(a) considerations” even in a mine-run case.
Ibid. Cf. Tr. of Oral Arg. in Gall v. United States, O. T.
2007, No. 06–7949, pp. 38–39.
The crack cocaine Guidelines, however, present no occa
sion for elaborative discussion of this matter because those
Guidelines do not exemplify the Commission’s exercise of its
characteristic institutional role. In formulating Guidelines
ranges for crack cocaine offenses, as we earlier noted, the
Commission looked to the mandatory minimum sentences set
in the 1986 Act, and did not take account of “empirical data
and national experience.” See Pruitt, 502 F. 3d, at 1171

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
110 KIMBROUGH v. UNITED STATES
Opinion of the Court
(McConnell, J., concurring). Indeed, the Commission itself
has reported that the crack/powder disparity produces dis
proportionately harsh sanctions, i. e., sentences for crack co
caine offenses “greater than necessary” in light of the pur
poses of sentencing set forth in § 3553(a). See supra, at
97–98. Given all this, it would not be an abuse of discretion
for a district court to conclude when sentencing a particular
defendant that the crack/powder disparity yields a sentence
“greater than necessary” to achieve § 3553(a)’s purposes,
even in a mine-run case.
V
Taking account of the foregoing discussion in appraising
the District Court’s disposition in this case, we conclude that
the 180-month sentence imposed on Kimbrough should sur
vive appellate inspection. The District Court began by
properly calculating and considering the advisory Guidelines
range. It then addressed the relevant § 3553(a) factors.
First, the court considered “the nature and circumstances”
of the crime, see § 3553(a)(1), which was an unremarkable
drug-trafficking offense. App. 72–73 (“[T]his defendant and
another defendant were caught sitting in a car with some
crack cocaine and powder by two police officers—that’s the
sum and substance of it—[and they also had] a firearm.”).
Second, the court considered Kimbrough’s “history and char
acteristics.” § 3553(a)(1). The court noted that Kimbrough
had no prior felony convictions, that he had served in combat
during Operation Desert Storm and received an honorable
discharge from the Marine Corps, and that he had a steady
history of employment.
Furthermore, the court alluded to the Sentencing Com
mission’s reports criticizing the 100-to-1 ratio, cf. § 3553(a)(5)
(2000 ed., Supp. V), noting that the Commission “recognizes
that crack cocaine has not caused the damage that the Jus
tice Department alleges it has.” App. 72. Comparing the
Guidelines range to the range that would have applied if
Kimbrough had possessed an equal amount of powder, the

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
111 Cite as: 552 U. S. 85 (2007)
Opinion of the Court
court suggested that the 100-to-1 ratio itself created an un
warranted disparity within the meaning of § 3553(a). Fi
nally, the court did not purport to establish a ratio of its own.
Rather, it appropriately framed its final determination in line
with § 3553(a)’s overarching instruction to “impose a sen
tence sufficient, but not greater than necessary,” to accom
plish the sentencing goals advanced in § 3553(a)(2). See
supra, at 101. Concluding that “the crack cocaine guide
lines [drove] the offense level to a point higher than is neces
sary to do justice in this case,” App. 72, the District Court
thus rested its sentence on the appropriate considerations
and “committed no procedural error,” Gall, ante, at 56.
The ultimate question in Kimbrough’s case is “whether the
sentence was reasonable—i. e., whether the District Judge
abused his discretion in determining that the § 3553(a) fac
tors supported a sentence of [15 years] and justified a sub
stantial deviation from the Guidelines range.” Ibid. The
sentence the District Court imposed on Kimbrough was 4.5
years below the bottom of the Guidelines range. But in de
termining that 15 years was the appropriate prison term, the
District Court properly homed in on the particular circum
stances of Kimbrough’s case and accorded weight to the Sen
tencing Commission’s consistent and emphatic position that
the crack/powder disparity is at odds with § 3553(a). See
Part II–B, supra. Indeed, aside from its claim that the
100-to-1 ratio is mandatory, the Government did not attack
the District Court’s downward variance as unsupported by
§ 3553(a). Giving due respect to the District Court’s rea
soned appraisal, a reviewing court could not rationally con
clude that the 4.5-year sentence reduction Kimbrough re
ceived qualified as an abuse of discretion. See Gall, ante, at
58–60; Rita, 551 U. S., at 358–360.
* * *
For the reasons stated, the judgment of the United States
Court of Appeals for the Fourth Circuit is reversed, and the

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
112 KIMBROUGH v. UNITED STATES
Scalia, J., concurring
case is remanded for further proceedings consistent with
this opinion.
It is so ordered.
Justice Scalia, concurring.
The Court says that “closer review may be in order when
the sentencing judge varies from the Guidelines based solely
on the judge’s view that the Guidelines range ‘fails properly
to reflect § 3553(a) considerations’ even in a mine-run case,”
but that this case “present[s] no occasion for elaborative dis
cussion of this matter.” Ante, at 109 (quoting Rita v. United
States, 551 U. S. 338, 351 (2007)). I join the opinion only
because I do not take this to be an unannounced abandon
ment of the following clear statements in our recent opinions:
“[Our remedial opinion] requires a sentencing court to
consider Guidelines ranges, . . . but it permits the court
to tailor the sentence in light of other statutory concerns
as well, see § 3553(a).
. . . . .
“[W]ithout this provision—namely, the provision that
makes ‘the relevant sentencing rules . . . mandatory
and impose[s] binding requirements on all sentencing
judges’—the statute falls outside the scope of Ap
prendi’s requirement. . . .
. . . . .
“The district courts, while not bound to apply the Guide
lines, must consult those Guidelines and take them into
account when sentencing.” United States v. Booker,
543 U. S. 220, 245–246, 259, 264 (2005).
“Under the system described in Justice Breyer’s
opinion for the Court in Booker, judges would no longer
be tied to the sentencing range indicated in the Guide
lines. But they would be obliged to ‘take account of ’
that range along with the sentencing goals Congress
enumerated in the [Sentencing Reform Act of 1984] at

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
113 Cite as: 552 U. S. 85 (2007)
Scalia, J., concurring
18 U. S. C. § 3553(a).” Cunningham v. California, 549
U. S. 270, 286–287 (2007).
“[The sentencing judge] may hear arguments by prose
cution or defense that the Guidelines sentence should
not apply, perhaps because (as the Guidelines them
selves foresee) the case at hand falls outside the
‘heartland’ to which the Commission intends individual
Guidelines to apply, USSG § 5K2.0, perhaps because
the Guidelines sentence itself fails properly to reflect
§ 3553(a) considerations, or perhaps because the case
warrants a different sentence regardless. See Rule
32(f). . . .
. . . . .
“A nonbinding appellate presumption that a Guide
lines sentence is reasonable does not require the sen
tencing judge to impose that sentence. Still less does
it prohibit the sentencing judge from imposing a sen
tence higher than the Guidelines provide for the jury
determined facts standing alone. As far as the law is
concerned, the judge could disregard the Guidelines and
apply the same sentence (higher than the statutory min
imum or the bottom of the unenhanced Guidelines range)
in the absence of the special facts (say, gun brandishing)
which, in the view of the Sentencing Commission, would
warrant a higher sentence within the statutorily permis
sible range.” Rita, supra, at 351, 353.
These statements mean that the district court is free to
make its own reasonable application of the § 3553(a) factors,
and to reject (after due consideration) the advice of the
Guidelines. If there is any thumb on the scales; if the
Guidelines must be followed even where the district court’s
application of the § 3553(a) factors is entirely reasonable;
then the “advisory” Guidelines would, over a large expanse
of their application, entitle the defendant to a lesser sentence

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
114 KIMBROUGH v. UNITED STATES
Thomas, J., dissenting
but for the presence of certain additional facts found by
judge rather than jury. This, as we said in Booker, would
violate the Sixth Amendment.
Justice Thomas, dissenting.
I continue to disagree with the remedy fashioned in
United States v. Booker, 543 U. S. 220, 258–265 (2005). The
Court’s post-Booker sentencing cases illustrate why the re
medial majority in Booker was mistaken to craft a remedy
far broader than necessary to correct constitutional error.
The Court is now confronted with a host of questions about
how to administer a sentencing scheme that has no basis in
the statute. Because the Court’s decisions in this area are
necessarily grounded in policy considerations rather than
law, I respectfully dissent.
In Booker, the Court held that the Federal Sentencing
Guidelines violate the Sixth Amendment insofar as they per
mit a judge to make findings that raise a sentence beyond
the level justified by the “ ‘facts reflected in the jury verdict
or admitted by the defendant.’ ” Id., at 232 (quoting Blakely
v. Washington, 542 U. S. 296, 303 (2004); emphasis deleted).
In my view, this violation was more suitably remedied by
requiring any such facts to be submitted to the jury. Booker,
543 U. S., at 323–325 (Thomas, J., dissenting in part). That
approach would have been consistent with our longstanding
presumption of the severability of unconstitutional applica
tions of statutory provisions. Id., at 322–323. And it
would have achieved compliance with the Sixth Amendment
while doing the least amount of violence to the mandatory
sentencing regime that Congress enacted. Id., at 324–326.
The Court, however, chose a more sweeping remedy. De
spite acknowledging that under the mandatory Guidelines
not “every sentence gives rise to a Sixth Amendment viola
tion,” the Court rendered the Guidelines advisory in their
entirety and mandated appellate review of all sentences for
“reasonableness.” Id., at 268. Because the Court’s “solu

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
115 Cite as: 552 U. S. 85 (2007)
Thomas, J., dissenting
tion fail[ed] to tailor the remedy to the wrong,” I dissented
from the remedial opinion. Id., at 313.
As a result of the Court’s remedial approach, we are now
called upon to decide a multiplicity of questions that have
no discernibly legal answers. Last Term, in Rita v. United
States, 551 U. S. 338 (2007), the Court held that a Court of
Appeals may treat sentences within the properly calculated
Guidelines range as presumptively reasonable. Today, in
Gall v. United States, ante, p. 38, the Court holds that a
Court of Appeals may not require sentences that deviate
substantially from the Guidelines range to be justified by
extraordinary circumstances. And here the Court holds
that sentencing courts are free to reject the Sentencing
Guidelines’ 100-to-1 crack-to-powder ratio.
These outcomes may be perfectly reasonable as a matter
of policy, but they have no basis in law. Congress did not
mandate a reasonableness standard of appellate review—
that was a standard the remedial majority in Booker fash
ioned out of whole cloth. See 543 U. S., at 307–312 (Scalia,
J., dissenting in part). The Court must now give content to
that standard, but in so doing it does not and cannot rely on
any statutory language or congressional intent. We are
asked here to determine whether, under the new advisory
Guidelines regime, district courts may impose sentences
based in part on their disagreement with a categorical policy
judgment reflected in the Guidelines. But the Court’s an
swer to that question necessarily derives from something
other than the statutory language or congressional intent be
cause Congress, by making the Guidelines mandatory, quite
clearly intended to bind district courts to the Sentencing
Commission’s categorical policy judgments. See 18 U. S. C.
§ 3553(b) (2000 ed. and Supp. V) (excised by Booker). By
rejecting this statutory approach, the Booker remedial ma
jority has left the Court with no law to apply and forced it
to assume the legislative role of devising a new sentencing
scheme.

552US1 Unit: $$U7 [01-11-12 21:30:51] PAGES PGT: OPIN
116 KIMBROUGH v. UNITED STATES
Alito, J., dissenting
Although I joined Justice Scalia in Rita accepting the
Booker remedial opinion as a matter of “statutory stare deci
sis,” 551 U. S., at 368 (opinion concurring in part and concur
ring in judgment), I am now convinced that there is no prin
cipled way to apply the Booker remedy—certainly not one
based on the statute. Accordingly, I think it best to apply
the statute as written, including 18 U. S. C. § 3553(b), which
makes the Guidelines mandatory. Cf. Dickerson v. United
States, 530 U. S. 428, 465 (2000) (Scalia, J., dissenting).
Applying the statute as written, it is clear that the District
Court erred by departing below the mandatory Guidelines
range. I would therefore affirm the judgment of the Court
of Appeals vacating petitioner’s sentence and remanding
for resentencing.
Justice Alito, dissenting.
For the reasons explained in my dissent in Gall v. United
States, ante, p. 61, I would hold that, under the remedial
decision in United States v. Booker, 543 U. S. 220, 258–265
(2005), a district judge is still required to give significant
weight to the policy decisions embodied in the Federal Sen
tencing Guidelines. The Booker remedial decision, however,
does not permit a court of appeals to treat the Guidelines’
policy decisions as binding. I would not draw a distinction
between the Guideline at issue here and other Guidelines.
Accordingly, I would vacate the decision of the Court of Ap
peals and remand for reconsideration.