UNITED STATES of America, Plaintiff-Appellee, v. Kevin MARTIN, Defendant-Appellant, 438 F.3d 621

GILMAN, J., delivered the opinion of the court. MARTIN, J. (pp. 639 - 42), also delivered a separate concurring opinion. OPINION RONALD LEE GILMAN, Circuit Judge. In this case of first impression, we consider whether the United States Sentencing Commission failed to comply with Congress’s directive when it established ratios to estimate the amount of methamphetamine that can reasonably be manufactured from certain precursor chemicals.

The issue arises in the context of Kevin Martin’s challenge to the sentence imposed on him after he pled guilty to five counts relating to the manufacture of methamphetamine and the possession of pseu-doephedrine as a precursor chemical. Relying on a Presentence Report (PSR) to which Martin had objected on Sixth Amendment grounds, the district court sentenced Martin to 189 months of imprisonment on each count, with the sentences to run concurrently. This sentence was calculated using a 50% ratio for converting the amount of pseudoephedrine attributed to Martin into the corresponding quantity of methamphetamine.

The Sentencing Commission promulgated the conversion ratio in response to a statute enacted in 2000 that required it to establish a table of such ratios “based on scientific, law enforcement, and other data the Sentencing Commission considers appropriate.” Pub.L. No. 106-310, § 3651(b)(2), 114 Stat. 1238-39 (2000).

On appeal, Martin argues that (1) the ratio set forth in the Sentencing Guidelines commentary for converting the precursor chemical pseudoephedrine to methamphetamine is invalid both because it does not comply with the statutory mandate and because it is arbitrary and capricious, (2) the district court erred in calculating his criminal history category, and (3) the district court violated his Sixth Amendment rights as interpreted by United States v. Booker, 543 U.S. 220, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005).

Although we hold that the unambiguous language of the statute requires the Commission to utilize both scientific and law enforcement data, we ultimately conclude that Martin has not met his burden of showing that the Commission actually failed to base the conversion ratios on both types of data when it relied on a report prepared by the Drug Enforcement Administration (DEA). We also reject Martin’s challenge to the calculation of his criminal history category, but agree that the district court’s Booker error entitles him to resentencing.

Accordingly, we VACATE Martin’s sentence and REMAND the case for resentencing consistent with Booker. I. BACKGROUND A. Statutory and Guidelines framework Congress responded to growing concerns about a “methamphetamine epidemic in America,” United States v. Layne, 324 F.3d 464, 468 (6th Cir.2003) (quoting H.R. Rep. 106-878, at 22 (Sept. 21, 2000)), by passing the Methamphetamine and Club Drug Anti-Proliferation Act of 2000 (the Act), Pub.L. No. 106-310, §§ 3601-3673, 114 Stat. 1101, 1227-46 (2000).

Supplant ing the individualized determination of how much of a controlled substance .certain-chemicals would yield, see United States v. Hamilton, 81 F.3d 652, 653-54 (6th Cir.1996), the Act instructed the United States Sentencing Commission to (1) ... review and amend its guidelines to provide for increased penalties such that those penalties corresponded to the quantity of controlled substance that could reasonably have been manufactured using the quantity of ephedrine, phenylpropanolamine, or pseudoephed-rine possessed or distributed.

(2) CONVERSION RATIOS. For the purposes of the amendments made by this subsection, the quantity of controlled substance that could reasonably have been manufactured shall be determined by using a table of manufacturing conversion ratios for ephedrine, phenyl-propanolamine, and pseudoephedrine, which table shall be established by the Sentencing Commission based on scientific, law enforcement, and other data the Sentencing Commission considers appropriate. Pub.L. No. 106-310, § 3651(b), 114 Stat. 1238-39 (2000) (emphasis added). The Commission responded to the congressional directive by promulgating Amendment 611.

In relevant part, Amendment 611 provides a new chemical-quantity table for precursor chemicals like pseudoephedrine and a conversion table for those chemicals. See U.S. Sentencing Guidelines, App. C, Amendment 611 (Nov. 1, 2003). These tables adopt a 50% conversion ratio for pseudoephedrine, such that 2 grams of the chemical is equivalent to 1 gram of methamphetamine. That ratio was inserted into the tables in § 2D 1.1, cmt. n.10, which already established that 1 gram of methamphetamine is to be treated as the equivalent of 20 kilograms of marijuana for sentencing purposes.

Since the enactment of Amendment 611, therefore, 1 gram of pseudoephedrine is treated as the equivalent of 10 kilograms of marijuana. ■In adopting the 50% conversion ratio for pseudoephedrine, -the Commission relied on a report promulgated by the DEA’s Office of Diversion Control that was published on the website of the Office of National Drug' Control Policy (ONDCP).

That report “indicate[dj that the actual yield of methamphetamine from ephedrine and pseudoephedrine is typically in the range of 50 to 75 percent.” Proposed Amendments to the Sentencing Guidelines, 66 Fed.Reg. 7962, 7965 (Jan. 26, 2001) (citation omitted); see also U.S. Sentencing Guidelines, App. C, Amendment 611 (“This yield is based on information provided by the Drug Enforcement Agency (DEA) that the typical yield of these substances for clandestine laboratories is 50 to 75 percent.”).

The DEA report on which the Commission based its conversion ratio is no longer available on the DEA website and is not part of the record in the present case. A document that appears to be the report, however, remains available through another publically accessible website. See Gene Haislip, Methamphetamine Precursor Chemical Control in the 1990’s (1996), http://www.erowid.org/ archive/rhodium/chemistry/ dojmeth3.txt (stating that “[ajctual yield in clandestine labs is typically in the range of 50 to 75 percent”) (last visited Feb. 14, 2006).

When a defendant was convicted of an offense relating to the manufacture of methamphetamine from precursor chemicals like pseudoephedrine prior to the passage of the Act in 2000, the district court would rely on expert testimony to approximate the amount of methamphetamine that could be produced from various precursor chemicals under differing laboratory conditions. Experts would testify as to how. much methamphetamine the precursor chemicals would yield in a given situation, and the court would base its sentence on that yield calculation. See U.S.

Sentencing Guidelines § 2D1.1, cmt. 12 (2003) (“Where there is no drug seizure or the amount seized does not reflect the scale of offense, the court shall approximate the quantity of the controlled substance.”); see also, e.g., United States v. Brannon, 7 F.3d 516, 520 (6th Cir.1993) (upholding the district court’s estimate of the quantity of methamphetamine, which was based on the affidavits and testimony from a DEA agent and a chemist); United States v.

Eschman, 227 F.3d 886, 891 (7th Cir.2000) (rejecting the district court’s use of a one-to-one conversion rate for pseu-doephedrine where expert testimony contradicted the reliability of that rate). This court required that the findings of the testifying experts be “particularized to individual laboratories,” Hamilton, 81 F.3d at 654, a practice that the uniform conversion rate imposed by the Guidelines provision rendered unnecessary. B. Crimes and conviction The government charged Martin with five offenses relating to the production of methamphetamine, all of which stem from two incidents in late 2003.

On November 24th of that year, Martin, who was on supervised release for previous methamphetamine offenses, rented a hotel room outside of Chattanooga, Tennessee. Law enforcement officers learned of a possible methamphetamine laboratory at the hotel and arrested five people found in a room rented by Christi Kinsey. Among those people was Genea Davis, Martin’s girlfriend, who consented to a search of the room that she was sharing with Martin.

In that room, the officers discovered a hot plate, Coleman fuel, jars with multi-lay-ered liquids, antifreeze, and a 1,000 milliliter flask — all materials commonly found in methamphetamine labs. Subsequent tests confirmed that at least some of the materials had been used to produce methamphetamine. The government learned from Davis that several of the articles recovered at the hotel belonged to Rory Shankles, another known formulator of methamphetamine. Davis then led police to Shankles’s residence, a location where Davis reported having seen Shankles and Martin “cooking” methamphetamine two days earlier.

Police obtained a warrant and searched the premises, recovering 24 empty bottles of pseudoephedrine, an over-the-counter decongestant that is also a raw material used to manufacture methamphetamine. Those bottles, when full, contained approximately 51 grams of pseudoephedrine. The second incident occurred on December 2, 2003, when police searched an abandoned pickup truck that Martin had borrowed from his friends. Officers found items in the truck similar to those previously discovered during the search of the hotel room in November, as well as 1.1 grams of methamphetamine residue attached to coffee filters.

After Martin’s codefendants pled guilty to various methamphetamine-related charges, a grand jury returned a superseding indictment charging Martin with five counts stemming from the production of methamphetamine and the use of a hotel room as a methamphetamine lab. Martin entered a guilty plea to all five counts on the day before his trial was scheduled to begin. C. Sentencing proceedings The final PSR attributed to Martin 51 grams of pseudoephedrine and 1.1 grams of seized methamphetamine. Martin objected to the calculation of the drug quantity, arguing that the Supreme Court’s intervening decision in Blakely v.

Washington, 542 U.S. 296, 124 S.Ct. 2531, 159 L.Ed.2d 403 (2004), barred the district court from accepting a quantity that had neither been found by a jury beyond a reasonable doubt nor admitted by Martin in his guilty plea. The government responded by offering the testimony of Mitchell Smith, a member of the DEA task force who worked on the case, as well as that of Christi Kinsey, one of Martin’s coconspirators. These witnesses explained Martin’s ties to the laboratory found in the hotel room and to the pseu-doephedrine bottles found at Shankles’s residence.

The district court credited this testimony in ruling that the amount from the empty bottles had properly been attributed to Martin. In calculating Martin’s criminal history category, the PSR assessed one point for each of four car thefts occurring between November 11 and December 4, 2000. Martin objected to these assessments both in writing and at the sentencing hearing, arguing that he had engaged in “a string of thefts or a common scheme to steal autos,” all of which were “related” offenses for which he should have been assessed only one point under U.S. Sentencing Guidelines Manual § 4A1.2(a)(2).

The district court rejected Martin’s argument, observing that the offenses had taken place at different times and at different locations, and that the state courts had not entered an order consolidating the convictions. With Martin’s offense level and criminal history category yielding a Guidelines range of 168 to 210 months, he was sentenced to 189 months in prison and 6 years of supervised release. This timely appeal followed. II. ANALYSIS A.

This court has the discretion to address Martin’s challenge to the validity of the Guidelines drug-conversion ratios even though he did not raise the issue before the district court The government’s initial argument is that Martin has waived any challenge to the validity of the Guidelines provisions by failing to raise the issue before the district court. Citing United States v.

Chesney, 86 F.3d 564, 567-68 (6th Cir.1996), the government contends that all challenges not presented to the district court are waived “unless the case is exceptional or where the rule would produce a plain miscarriage of justice.” The general principle “that courts of appeals do not consider claims or arguments that were not raised before the district court,” however, “is a prudential rule, not a jurisdictional one.” United States v. Hayes, 218 F.3d 615, 619 (6th Cir.2000); see also United States v.

Suarez, 263 F.3d 468, 487 (6th Cir.2001) (Boggs, J., dissenting in part) (“[Tjhis rule is one of prudence, not jurisdiction, and the court has discretion to examine the issue under certain conditions.”). Moreover, this court has frequently addressed belated challenges based on both constitutional and statutory grounds where “the question [is] a purely legal one that has been fully briefed by both parties.” Id.; accord United States v.

Pickett, 941 F.2d 411, 415 (6th Cir.1991) (addressing the sentencing challenges of a defendant who failed to raise the issues below where “[b]oth parties have extensively briefed the issues at stake, and ... the issues are wholly legal”); cf. Mayhew v. Allsup, 166 F.3d 821, 823 (6th Cir.1999) (holding that the court of appeals may address an issue not presented to the district court where “the issue allegedly forfeited is purely a matter of law, is dispositive, and, if applied, will result in reversal”).

In the present case, Martin’s challenges to the validity of the Guidelines provisions governing methamphetamine-related offenses are purely legal in nature, and both parties have fully briefed the issue. The specific challenges that Martin raises are also certain to recur, as the government acknowledges in its brief by citing to three consolidated cases raising analogous challenges that are currently pending before another panel of this court.

Because the government has fully briefed the relevant issue and will therefore suffer no prejudice from our decision to address it at this time, we will exercise our discretion and take up Martin’s challenge despite his failure to raise it before the district court. B. Martin has not demonstrated that the Commission failed to follow the Congressional command in formulating the 50% ratio Martin presents two challenges to the validity of the pseudoephedrine-conversion ratio adopted by the Commission.

First, he argues that “the Commission engaged in an unauthorized delegation” by failing to follow Congress’s statutory command. His second contention is that the use of the DEA report as the exclusive basis for the conversion ratio was “arbitrary and capricious” and therefore unlawful. Even though the Supreme Court has declared that the Sentencing Guidelines are no longer mandatory, United States v.

Booker, 543 U.S. 220, 245, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005), Martin’s challenge to the validity of the conversion ratios remains consequential because the district court is still required to “consult [the] Guidelines and take them into account when sentencing.” Id. at 263, 125 S.Ct. 738 (Opinion of Breyer, J.). We must therefore consider Martin’s argument regarding the proper application of the Guidelines in order to correctly instruct the district court on remand. See, e.g., United States v.

Trammel, 404 F.3d 397, 403 (6th Cir.2005) (“While remand for resentencing is required under Booker, we consider Trammel’s remaining claim under the Guidelines because the district court will need to consider the correct Guidelines-recommended sentence in fashioning its own post -Booker sentence on remand.”) (citation omitted). Because Martin did not present his statutory argument to the district court, we will apply the “plain error” standard of review set forth in Rule 52(

b) of the Federal Rules of Criminal Procedure. See Johnson v. United States, 520 U.S. 461, 467-70, 117 S.Ct. 1544, 137 L.Ed.2d 718 (1997) (applying the four-pronged plain-error analysis to a challenge first raised on appeal); see also Hayes, 218 F.3d at 622 (assuming that the plain-error standard was appropriate in reviewing a Confrontation Clause challenge presented for the first time on appeal); Suarez, 263 F.3d at 487 (Boggs, J., dissenting in part) (following Hayes and applying plain-error analysis to a constitutional challenge of a federal criminal statute). This court has summarized the “four distinct analyses” of the plain-error inquiry under Rule 52(

b) as follows: First, we are to consider whether an error occurred in the district court. Absent any error, our inquiry is at an end. However, if an error occurred, we then consider if the error was plain. If it is, then we proceed to inquire whether the plain error affects substantial rights. Finally, even if all three factors exist, ... we must decide whether the plain error affecting substantial rights seriously affected the fairness, integrity or public reputation of judicial proceedings. United States v.

Thomas, 11 F.3d 620, 630 (6th Cir.1993) (interpreting the Supreme Court’s then-recent decision in United States v. Olano, 507 U.S. 725, 113 S.Ct. 1770, 123 L.Ed.2d 508 (1993)); see also United States v. Jones, 108 F.3d 668, 670 (6th Cir.1997) (en banc) (following Thomas’s division of the plain-error inquiry into “four distinct, though interrelated, analy-ses”). We must decide at the outset, then, whether the district court’s use of the conversion ratio established by Amendment 611 constituted an error.

See Olano, 507 U.S. at 733-34, 113 S.Ct. 1770 (“If a legal rule was violated ... and if the defendant did not waive the rule, then there has been an ‘error’ within the meaning of Rule 52(

b) despite the absence of a timely objection.”). 1. The plain language of the statute requires the Commission to base the conversion ratios, at a minimum, on both scientific and law enforcement data Martin first argues that the Commission failed to heed Congress’s command to base the determination of the quantity of methamphetamine that “could reasonably [be manufactured]” from precursor chemicals “on scientific, law enforcement, and other data the Sentencing Commission considers appropriate.” Pub.L. No. 106-310, § 3651(b)(2), 114 Stat. 1239 (2000).

This failure to adhere to the language of the statute renders the challenged provisions invalid, Martin contends, because the Commission’s “significant discretion ... must bow to the specific directives of Congress.” United States v. LaBonte, 520 U.S. 751, 757, 117 S.Ct. 1673, 137 L.Ed.2d 1001 (1997) (citation omitted); see also United States v. Butler, 207 F.3d 839, 850 (6th Cir.2000) (citing LaBonte for that proposition). At issue in LaBonte was the Commission’s

interpretation of statutory language that directed the Commission to “assure that the guidelines specify a sentence to a term of imprisonment at or near the maximum term authorized for categories of defendants.” 520 U.S. at 753, 117 S.Ct. 1673 (quoting 28 U.S.C. § 994(h)). The Commission had interpreted the phrase “at or near the maximum term authorized” as excluding potential enhancements for so-called career offenders, LaBonte, 520 U.S. at 754-55, 117 S.Ct. 1673, but the Supreme Court invalidated the Commission’s reading as inconsistent with the unambiguous statutory language. Id. at 762, 117 S.Ct. 1673.

In doing so, the Court started from the premise that “Congress said what it meant,” and examined whether the Guidelines commentary “accurately reflected] Congress’ intent” by “[g]iving the words used their ordinary meaning.” Id. at 757, 117 S.Ct. 1673 (citation omitted); see also Stinson v. United States, 508 U.S. 36, 38, 113 S.Ct. 1913, 123 L.Ed.2d 598 (1993) (deciding “that commentary in the Guidelines Manual that interprets or explains a guideline is authoritative unless it violates the Constitution or a federal statute”). This court’s decision in United States v.

Butler, 207 F.3d 839, 844 (6th Cir.2000), applied LaBonte to a provision of the Guidelines that instructed the sentencing court to increase by two the offense levels of all defendants, regardless of their age, who “use[] or attempt[] to use a person less than eighteen years of age to commit the offense dr assist in avoiding detection of, or apprehension for, the offense.” U.S. Sentencing Guidelines Manual § 3B1.4.

The Commission enacted the challenged Guideline in response to a statute that “directed the Commission to ‘promulgate guidelines or amend existing guidelines to provide that a defendant 21 years of age or older who has been convicted of an offense shall receive an appropriate sentence enhancement if the defendant involved a minor in the commission of the offense.’ ” Butler, 207 F.3d at 844 (quoting Violent Crime Control and Law Enforcement Act of 1994, Pub.L. No. 103-22, § 140008, 108 Stat.2033 (1994)) (emphasis added).

This court held that the Commission had effectively “overrul[ed] .an explicit Congressional declaration because it eliminated the age limit, lock, stock-and-barrel.” Butler, 207 F.3d at 850. “Looking at the face of both the [congressional] directive and the guideline,” the court concluded that the Commission had- “flatly ignored a clear Congressional directive.” Id. Because courts will not apply Guidelines “[w]hen the Commission’s

interpretation ... does not square with clear Congressional intent,” the Butler court declared the Guideline invalid and instead applied the age restriction set forth in the authorizing statute. Id. at 850-51; see also United States v. Price, 990 F.2d 1367, 1369 (D.C.Cir.1993) (explaining that whatever level of deference is owed to the Commission, “that deference does not extend to

interpretations in conflict with a clear determination of Congress”). Invoking LaBonte and Butler, Martin focuses on the plain text of the Act. Martin’s argument centers on the word “and,” which joins the types of data on which the Commission was to base the conversion ratios it added to § 2D1.1 of the Sentencing Guidelines. According to Martin, the use of “and” — which is normally conjunctive — required the Commission, at a minimum, to utilize both scientific data and law enforcement data.

The government, for its part, maintains that “Congress stated [that the conversion] ratio could be based upon law enforcement data or other data the Commission found to be appropriate.” (Emphasis added.) If the government’s disjunctive reading is correct, then Martin cannot prevail because reliance on a single DEA report would be in keeping with the Commission’s discretion to consider whatever sources it found “appropriate.” We think that the

interpretation advanced by Martin is the correct one. His reading of the statute can perhaps be illuminated by the following example: Assume that a state legislature directs a commission that oversees the sale and distribution of certain food and beverage products to • promulgate regulations governing the sale of “alcoholic, caffeinated, and other beverages that the commission considers to be addictive.” The most natural reading of such a statute is that the commission must regulate the sale of both alcoholic and caffeinated beverages.

The concluding phrase “and other beverages that the commission considers to be addictive” allows the commission to regulate the sale of additional beverages, but does not eliminate the commission’s obligation to regulate at least the two types of products specifically listed by the legislature. That is, the statute establishes two categories of beverages that the commission must regulate and a third that it may, in its discretion, also regulate. This reading is the most natural one because, as this court recently explained at length, “dictionary

definitions, legal usage guides and case law compel us to start from the premise that ‘and’ usually does not mean ‘or.’ ” OfficeMax, Inc. v. United States, 428 F.3d 583, 588 (6th Cir.2005); see also 1A Norman J.

Singer, Statutes and Statutory Construction § 21.14 at 183 (6th ed. 2000) (“The literal meaning of [‘and’ and ‘or’] should be followed unless it renders the statute inoperable or the meaning becomes questionable.”) (emphasis added); id. at 184-86 (“The [two] words are not interchangeable, and their strict meaning should be followed when their accurate reading does not render the sense of the statute confusing and there is no clear legislative intent to have the words not mean what they strictly should.”). In the context of evaluating an

interpretation by the Commission, the Supreme Court in LaBonte likewise reiterated that the words of statutes delegating power to the Commission should be given their “ordinary meaning,” 520 U.S. at 757, 117 S.Ct. 1673 (citation omitted), and the ordinary meaning of “and” is a conjunctive one. See also Woodall v. Fed. Bureau of Prisons, 432 F.3d 235, 245 (3d Cir.2005) (opining that “the use of the word ‘and’ before the final factor in [a] five-part list indicatefd] that Congress intended for the [Bureau of Prisons] to weigh all of the factors listed”) (emphasis in original).

In light of these interpretive principles, we find the government’s disjunctive reading of the statute unpersuasive. The government inaccurately summarizes the meaning of the statute by omitting the key word “scientific” and by replacing “and” with “or.” Its

interpretation also violates the grammatical rule known as the “rule of the last antecedent,” a principle that is consistent with the

interpretation of the Act that Martin advocates. Under that rule, “a limiting clause or phrase” — here, the phrase “[that] the Commission considers appropriate” — “should ordinarily be read as modifying only the noun or phrase that it immediately follows.” Barnhart v. Thomas, 540 U.S. 20, 26, 124 S.Ct. 376, 157 L.Ed.2d 333 (2003); 2A Singer, Statutes and Statutory Construction § 47.33 at 369 (“Referential and qualifying words and phrases, where no contrary intention appears, refer solely to the last antecedent.”); see also United States v.

Kerley, 416 F.3d 176, 180 n. 2 (2d Cir.2005) (acknowledging that the rule of the last antecedent is “not absolute,” but applying the rule where it led to an

interpretation of a statute “that not only comports with case law ... but also makes eminent sense”). The phrase that the limiting clause follows in the Act is “other data.” If we apply the last antecedent rule, we find a statute syntactically analogous to the beverages-commission example set forth above. That is, the Commission has been conferred the discretion to base its conversion ratios not just on scientific and law enforcement data, but also on additional sources that it “considers appropriate.” The Fourth Circuit in Commonwealth of Virginia v.

Browner, 80 F.3d 869, 877 (4th Cir.1996), faced an analogous limiting clause that followed the third category in a series of three. The language in the Clean Air Act at issue in Browner required state permit programs to contain “an opportunity for judicial review in State court of the final permit action by the applicant, any person who participated in the public comment process, and any other person who could obtain judicial review of that action under applicable law.” Id. (citing

section 502(b)(6) of the Clean Air Act, 42 U.S.C. § 7661a(b)(6)). Virginia argued that the limiting clause “who could obtain judicial review of that action under applicable law” applied to all three categories of people listed in the statute, such that states needed “to grant standing to participants in the public comment process only if those persons would otherwise have standing under existing law.” Id. The Fourth Circuit, however, refused to adopt that

interpretation. Instead, it applied the last-antecedent rule, reasoning that if the rule did not apply and Virginia’s reading was the correct one, “then there would have been no need for Congress to have included the first two categories.” Id .. Similarly, if the last-antecedent rule does not apply in the present case, Congress need not “have included the first two categories” listed in the Act — namely, scientific and law enforcement data.

Finally, the government points to the fact that “Congress has not acted to modify or vacate the Commission’s action,” something that it was permitted to do within 180 days after the Commission issued its proposed amendment. See 28 U.S.C. § 994(p). Congressional silence, the government urges, demonstrates legislative acquiescence in and implicit approval of the Commission’s work. This court, however, has expressly refused to rely on legislative silence that is “contrary to all other textual and contextual evidence of congressional intent.” Butler, 207 F.3d at 851 (quoting Burns v.

United States, 501 U.S. 129, 136, 111 S.Ct. 2182, 115 L.Ed.2d 123 (1991)). Accepting the government’s argument, as this court noted in Butler, would “dictate that all enacted guidelines inherently satisfied Congressional intent, and would eliminate [the courts’] vital role — described in LaBonte and other cases — of squaring the enacted guideline with the original statutory language.” Butler, 207 F.3d at 851. Congress’s refusal to exercise its power to amend or modify the challenged Guidelines, therefore, does not compel the conclusion that the Commission complied with the statutory text. 2.

The record does not support Martin’s contention that the Commission failed to rely on both scientific and law enforcement data in formulating the challenged conversion ratio Our conclusion that the statute unambiguously requires the Commission to base the conversion ratios on both scientific and law enforcement data, however, does not end our inquiry. Rather, we must still determine whether Martin has shown that the Commission did not actually rely on the requisite sources in promulgating the 50% conversion ratio. We begin by addressing the scope of our review of the Commission’s actions.

Unlike in LaBonte and Butler, the primary cases upon which Martin relies, we are not confronting a direct facial conflict between the substance of the Commission’s promulgated Guidelines commentary and the sentencing parameters set forth by federal statute.

See LaBonte, 520 U.S. at 762, 117 S.Ct. 1673 (declaring invalid Guidelines commentary that commanded a result contrary to “unambiguous” statutory language); Butler, 207 F.3d at 850 (concluding, after “[l]ooking at the face of both the [congressional] directive and the guideline,” that the enacted Guideline directly overruled “an explicit Congressional declaration” and was therefore invalid). Such a facial comparison is not possible in a case like the one before us, where the defendant’s challenge is to the internal procedures and methodology that the Commission followed in setting the relevant Guideline.

We have found no case that authorizes a searching judicial inquiry into the internal procedures by which the Commission implements Guidelines. To the contrary, the few courts to discuss the matter have suggested that investigation into the Commission’s internal processes and procedures may well be beyond the scope of judicial review. See United States v.

LeRoy, 984 F.2d 1095, 1098 & n. 3 (10th Cir.1993) (concluding that “[d]iseovery into the Guideline formulation process would be an intrusion into a quasi-legislative rulemak-ing function delegated by Congress solely to the Commission,” and observing that the Guidelines promulgation process “is not susceptible to discovery”); United States v. Lopez, 938 F.2d 1293, 1296-97 (D.C.Cir.1991) (holding that the Commission’s alleged failure to provide adequate reasons for limiting age-related departures was not subject to judicial review under the Administrative Procedures Act); United States v.

Vincent, 167 F.3d 428, 431 (8th Cir.1999) (stating that “the Sentencing Commission’s formulation of the Guidelines is not subject to judicial review unless the Commission oversteps constitutional bounds”). This court, following the D.C. Circuit’s decision in Lopez, issued an unpublished opinion agreeing with the view that courts of appeals “laek[ ] authority to review the Sentencing Commission’s rulemaking process” under the Administrative Procedures Act (APA). United States v. Tapert, No. 92-1628, 1993 WL 168923, at *9 (6th Cir. May 19,1993) (per curiam).

The Eleventh Circuit reached the same conclusion a few years later in United States v. Wimbush, 103 F.3d 968, 970 (11th Cir.1997) (per cu-riam) (“Federal courts do not have authority to review the Commission’s actions for compliance with APA provisions, at least insofar as the adequacy of the statement of the basis and purpose of an amendment is concerned.”). These cases rejected the contention that a court, pursuant to the APA, could invalidate otherwise lawful Guidelines provisions simply because the Commission had not explained its reasons for implementing the challenged provisions.

On the other hand, these precedents do not address a situation like the one at issue, where a defendant argues not that the Commission has failed to provide reasons for its actions, but instead that the reasons provided demonstrate that the Commission has not complied with a congressional directive. We have not previously articulated, at least in our published opinions, limitations on judicial review as broad as the ones announced in the cited cases.

See Butler, 207 F.3d at 851-52 (rejecting the broad conclusion of the Eighth Circuit in Vincent that the only limits on the Commission’s authority are ones of a constitutional dimension). The present case likewise does not require us to accept or reject such limitations. This is so because (1) even if we are permitted to delve into the Commission’s methodology as a general matter, the type of searching review that Martin asks us to conduct is impossible on the relatively thin record before us; and

(2) Martin’s arguments are not confined to the APA, but instead rely on the more general principle that courts “may set ¿side a sentencing guideline ... if it contravenes an unambiguously expressed intent of Congress or is unreasonable.” Tapert, 1993 WL 168923, at *5. " In essence, Martin would have us conclude that the Commission’s recognition in the public record that the chosen ratio mirrors the one found in a DEA report necessarily means that the ratio was based exclusively on law enforcement data.

But our analysis must focus not on the report, which is not in the record, but rather on the information that the Commission provided when it implemented the congressional directive. The Commission published notice of its proposed revised Guidelines and conversion table in the Federal Register on January 26, 2001, stating that [t]his table, which provides for a 50 percent conversion ratio for ephedrine, PPA, and pseudoephedrine, was developed using data from the Drug Enforcement Agency, Office of Diversion Control, as published on the web site of the Office of National Drug Control Policy (ONDCP).

These data indicate that the actual yield of methamphetamine from ephédrine and pseudoephedrine is “typically in the range of 50 to 75 percent[.]” 66 Fed.Reg. 7962, 7965 (Jan. 26, 2001). Martin also calls our attention to the June 6, 2001, Federal Register, which further describes the promulgation of the new Guidelines amendments as follows: The Commission held a public hearing on the proposed amendments in Washington, DC on March 19, 2001. After a review .of hearing testimony and additional public comment, the Commission promulgated the amendments set forth herein ....

On May 1, 2001, the Commission submitted these amendments to Congress and. specified an effective date of November 1, 2001. 66 Fed.Reg. 30512, 30513. These statements, along with the almost identical language published in the explanatory notes to Amendment 611, make up the sum total of the information that we have about the Commission’s creation of the conversion table. The record before us includes no testimony and no further documentation regarding the Commission’s method of selecting the new 50% conversion ratio.

No such evidence is contained in the district court record either, because Martin did not present his argument to that court. These scant indications, in our view, do not suffice to demonstrate that the Commission failed to base its ratio on both scientific and law enforcement data. Citing nothing more the Commission’s statement that the ratio was “developed using data from the Drug Enforcement Agency,” Martin asks us to infer that the DEA report consists solely of data gathered from law enforcement sources, to the exclusion of all scientific sources. We do not believe that such an inference is justified.

A particular document, regardless of its institutional author, might contain scientific data, law enforcement data, neither, or both. DEA reports could contain, for example, both scientific data regarding the controlled laboratory yield ratios for different conversion methods, as well as law enforcement statistics on which conversion methods were most popular among methamphetamine manufacturers. A single report could synthesize the data from various primary sources into a general conclusion more comprehensible to a layperson.

Nothing in the language of the enabling statute — which limits the Commission’s otherwise wide discretion, see LaBonte, 520 U.S. at 757, 117 S.Ct. 1673—prohibits the Commission from obtaining both its scientific and law enforcement data from the same source. Language in another federal statute, 28 U.S.C. § 995(c), suggests that Congress expected the Commission to rely on source material provided by other federal entities.

Congress instructed each federal agency, “[u]pon the request of the Commission, ... to make its services, equipment, personnel, facilities, and information available to the greatest practicable extent to the Commission in the execution of its functions.” Id. (emphasis added). The 2000 Act did not purport to negate this congressional authorization, nor did it impose any requirement that the information utilized by the Commission be raw data, as opposed to data previously collected and analyzed by other governmental bodies.

Furthermore, the presumption of regularity that attaches to the acts of government officials requires us to resolve in favor of the Commission any doubt as to the Commission’s compliance with the congressional mandate. As Judge Posner has observed, “[a] presumption of regularity attends the Commission’s doings, as it does that of other official bodies.” United States v. Tomasino, 206 F.3d 739, 742 (7th Cir.2000); see also U.S. Postal Serv. v.

Gregory, 534 U.S. 1, 10, 122 S.Ct. 431, 151 L.Ed.2d 323 (2001) (noting “that a presumption of regularity attaches to the actions of Government agencies”). “[I]n the absence of clear evidence to the contrary, courts presume that [public officials] have properly discharged their official duties,” United States v.

Chemical Foundation, Inc., 272 U.S. 1, 14-15, 47 S.Ct. 1, 71 L.Ed. 131 (1926), and Martin has not pointed to any evidence in the record — much less “clear evidence” — that can overcome the presumption that the Commission did consider the required variety of data in promulgating the pseudoephedrine-conversion ratio.

Notwithstanding our rejection of Martin’s challenge, we do not read the relevant authorities as holding that, to be successful, all challenges to Guidelines provisions or commentary must be of precisely the same ilk as the ones addressed in LaBonte and Butler — namely, that the substance of the provision directly conflicts with the authorizing statute.

We do believe, however, that there is an important difference between the contention that a Guidelines provision is substantively inconsistent with a clear statutory command and the assertion that the Commission has failed to follow the procedural path prescribed by Congress. The decisions in LaBonte and Butler signal that challenges of the former kind will likely succeed when the Commission acts contrary to the unambiguously expressed intent of Congress. On the other hand, challenges of the latter variety are likely to succeed only where

(1) Congress has set forth clear procedural requirements for the Commission to follow, and (2) the party advancing the challenge supplies evidence demonstrating that Commission did not comply with those procedures. In the present case, the first of these two conditions is satisfied because Congress required the Commission to base the 50% conversion ratio on at least two kinds of data — scientific and law enforcement. The second condition, as we have explained, has not been met.

Taking into account both the possibility that the DEA report in question contains the required types of data and the “presumption of regularity [that] attends the Commission’s doings,” Tomasino, 206 F.3d at 742, we conclude that Martin’s proof is insufficient to show that the Commission contravened the congressional' directive in promulgating the 50% conversion ratio. The district court therefore did not commit an error, plain or otherwise, in calculating Martin’s sentence using that ratio. C.

The 50% conversion ratio is neither “arbitrary and capricious” nor otherwise unconstitutional Martin also argues that the conversion ratio, beyond failing to conform to the statutory directive, is “arbitrary and capricious” because “DEA testimony before Congress and before federal courts was significantly different” from the ratio eventually adopted. The familiar “arbitrary and capricious” standard from § 706 of the Administrative Procedure Act (APA) is used to assess “the reasonableness of an agency’s actions pursuant to its governing statute.” Ohio Pub. Interest Research Group, Inc. v.

Whitman, 386 F.3d 792, 795 (6th Cir.2004). The court that deals most frequently with the APA, however, has held that the Sentencing Commission is generally not treated as an agency under that statute. See Wash. Legal Found. v. U.S. Sentencing Comm’n, 17 F.3d 1446, 1450 (D.C.Cir.1994) (holding that, unless specified by statute, the APA does “not apply to the Commission because it is a part of the judicial branch.”).

This court has not consistently applied the “arbitrary and capricious” standard to challenges to Guidelines provisions, instead asking “whether the guideline is sufficiently reasonable in light of congressional directions to the Sentencing Guideline Commission.” See United States v. Kincaid, 959 F.2d 54, 56 (6th Cir.1992); see also Butler, 207 F.3d at 850 (same); United States v. Kennedy, 32 F.3d 876, 889 (4th Cir.1994) (“The Sentencing Commission brings expertise to the implementation of its mandate, and we must accordingly defer to its

interpretation as long as it is sufficiently reasonable in light of the congressional directive.”) (citations and quotation marks omitted).

Nevertheless, because this court has not been consistent in articulating what standard of review applies to a challenge of this sort, we will assume for the sake of argument that the Commission’s selection of a conversion ratio of 50% could be struck down if such a choice were indeed “arbitrary and capricious.” The thrust of Martin’s argument is that expert testimony in reported federal court opinions and by DEA personnel before Congress conflicts with the Commission’s choice of 50% as the appropriate conversion ratio for pseudoephedrine.

But the sources that Martin cites reveal that, although yield rates are at times as low as 15%, they can also be as high as 85%. These sources — among them the so-called “Iowa Study” and expert testimony by a DEA chemist in United States v. Eschman, 227 F.3d 886, 889 (7th Cir.2000)— therefore reflect a “difference of opinion in the scientific community” as to yield rates. See United States v. Pickett, 941 F.2d 411, 418 (6th Cir.1991). This “difference of opinion” is similar to the type of ongoing debate that this court has found to be a rational basis for sentencing distinctions approved by Congress.

See id. (rejecting the argument that the 100-to-l crack-cocaine-to-powder ratio was unconstitutionally arbitrary, based in part on the “difference of opinion in the scientific community” with regard to whether crack was significantly more likely to lead to addiction than was cocaine). A yield rate of 50%, moreover, is not just a reasonable middle ground between two extremes, but is also borne out by cases predating the Act — cases in which this court endorsed the 50% rate as a valid approximation. See, e.g., United States v. Thompson, 86 Fed.Appx. 144, 146-47 (6th Cir.2004) (unpublished); United States v.

Oldham, 13 Fed.Appx. 221, 223 (6th Cir.2001) (unpublished); United States v. Pavlik, No. 93-2494, 1995 WL 59227, at *7-*8 (6th Cir. Feb.13, 1995) (unpublished). The disparities evident in the studies cited by Martin, combined with the pre-Act cases accepting the 50% yield rate, confirm that the Commission did not act “arbitrarily or irrationally” in adopting that percentage as its conversion ratio. See Pickett, 941 F.2d at 418. To the extent that we construe Martin’s “arbitrary and capricious” argument as a constitutional challenge, like the government does, Martin faces an uphill battle.

As an initial matter, neither this circuit nor any other has decided whether amendments to the Sentencing Guidelines are even subject to challenge as violating the substantive component of the Due Process Clause. See United States v. Fortney, 357 F.3d 818, 821 (8th Cir.2004) (assuming without deciding that such amendments were “subject to substantive due process challenge,” but calling the issue a “serious” one). This court has continually rejected substantive-due-process challenges to the Guidelines provisions themselves, see United States v.

Gaines, 122 F.3d 324, 329 (6th Cir.1997) (listing three cases that rejected substantive-due-process challenges to the crack-to-powder-cocaine sentencing ratio), without deciding what standard of review applies when the action challenged is one by the Commission rather than by Congress. See Fortney, 357 F.3d at 821 n. 4 (discussing the different standards of review applicable to substantive-due-process challenges to legislation and similar challenges to the conduct of executive officials).

As discussed above, if we apply to Martin’s contention the same standard used to evaluate substantive-due-process challenges to congressional action, then Martin has failed to show that the Commission “was acting arbitrarily or irrationally” in choosing the 50% ratio. Pickett, 941 F.2d at 418. Likewise, if Martin is challenging Congress’s directive to the Commission for formulating the conversion ratios, we believe that such a directive “was sufficiently rational to meet the demands of substantive due process.” United States v. Pruitt, 156 F.3d 638, 645 (6th Cir.1998) (citation omitted).

Because he cannot demonstrate that the Commission acted “arbitrarily or irrationally” in adopting the 50% yield rate, Martin cannot satisfy the more stringent standard of review that would apply if the Commission is treated as if it were an executive official accused of a substantive-due-process violation. See Fortney, 357 F.3d at 821 n. 4 (noting that “the conduct of executive officials must, at a minimum, ‘shock the conscience of federal judges’ to violate substantive due process”) (citing Collins v. City of Harker Heights, 503 U.S. 115, 126, 112 S.Ct. 1061, 117 L.Ed.2d 261 (1992)).

The Commission’s exclusive reliance on the DEA report does not “shock the conscience” simply because other available information was disregarded. Martin’s substantive-due-process claim therefore fails whether the Commission is held to the same standard applied to

an act of Congress or to the standard applied to executive officials. D. The district court did not commit clear error in determining Martin’s criminal history category Martin álso challenges the district court’s calculation of his criminal history category, arguing that the four car thefts he committed in November and December of 2000 are “related” offenses for which only one criminal history point should have been assessed. See U.S.

Sentencing Guidelines Manual § 4A1.2(a)(2) (establishing that prior sentences imposed in related cases are to be treated as one sentence for purposes of calculating a defendant’s criminal history category). As explained above, the district court on remand is required to “consult [the] Guidelines and take them into account when sentencing” even after Booker, 543 U.S. at 263 (Opinion of Breyer, J.), and we must therefore review that court’s application of the Guidelines. In doing so, “[w]e continue ... to apply the standards of review we applied prior to Booker,” United States v.

Davidson, 409 F.3d 304, 310 (6th Cir.2005), and thus will not overturn a district court’s determination as to whether offenses are related unless that court has committed a clear error. United States v. Horn, 355 F.3d 610, 613 (6th Cir.), cert. denied, 541 U.S. 1082, 124 S.Ct. 2436, 158 L.Ed.2d 998 (2004) (“[W]e must review - deferentially, that is, for clear error, the ... district court’s determination that Horn’s prior robbery convictions were not related.”). The Application Notes to § 4A1.2(a)(2) define related offenses as ones “that (

A) occurred on the same occasion, (

B) were part of a single common scheme or plan, or (

C) were consolidated for trial or sentencing.” U.S. Sentencing Guidelines Manual § 4A1.2, Application Note 3. Martin concedes that the thefts occurred on different occasions and that the state trial court did not consolidate the convictions.

He insists, however, that the thefts “were part of a single common scheme or plan” because they “were accomplished with the same modus operandi, specifically, that a fake drivers license was used to secure permission for a test drive.” Martin has the burden of proving that the four thefts at issue were part of a “single common scheme or plan,” a burden that he could satisfy by establishing that “his crimes were jointly planned or that the commission of one entailed the other.” United States v. Irons, 196 F.3d 634, 638-39 (6th Cir.1999).

Irons is this court’s most extensive treatment of the “single common scheme or plan” language in Application Note 3. The defendant in Irons was convicted of sending threatening letters to his ex-girlfriend. He claimed that his previous convictions for violating a temporary protective order and for robbing the victim’s home were “related” to each other and to the current crime as part of a common scheme or plan.

This court rejected his argument, holding instead that crimes are related “only if the offenses were jointly planned, or, at a minimum, the commission of one offense necessarily required the commission of another.” United States v. Carter, 283 F.3d 755, 758 (6th Cir.2002) (discussing the holdings in Irons and in United States v. Ali, 951 F.2d 827, 828 (7th Cir.1992)). Because the defendant in Irons could not prove that he had decided to break into the victim’s house at the same time that he violated the protective order, the offenses were held not to be part of single common scheme or plan. See 196 F.3d at 639.

Similarly, Martin introduced no evidence that he planned the final three car thefts at the time of the first theft, pointing only to the fact that all four crimes occurred within 23 days of one another. This court has observed, however, that “offenses are not necessarily related merely because they were committed within a short period of time.” Horn, 355 F.3d at 615; see also Irons, 196 F.3d at 640 (noting that “[njeither close geographic [n]or temporal proximity commands a finding that the defendant jointly planned the crimes”).

Moreover, “prior convictions are not ‘related’ merely because they are part of a crime spree.” Irons, 196 F.3d at 638; accord Horn, 355 F.3d at 614 (“This Court has further held that merely because crimes are part of a crime spree does not mean that they are related.”). The four car thefts for which Martin was convicted took place in two different states on four separate occasions, and Martin has presented no evidence establishing that the thefts were anything other than a crime spree that ended when he was arrested for his role in the thefts on December 4, 2000.

Martin’s argument that the common modus operandi rendered the crimes part of a single scheme or plan is likewise unavailing. This court has held that “the simple sharing of a modus operandi cannot alone convert [separate offenses] into one offense by virtue of their being a single common scheme or plan.” Horn, 355 F.3d at 614 (quoting United States v. Cowart, 90 F.3d 154, 160 (6th Cir.1996)).

To the contrary, similar substantive crimes committed on different dates involving different victims are not considered related even if each “was committed with the same purpose or common goal,” usually that of acquiring money. Irons, 196 F.3d at 639. Although Martin used the same tactics in stealing all four automobiles, the victim in each crime was different, and the commission of one theft did not necessarily entail committing the other thefts.

See Horn, 355 F.3d at 615 (holding that two offenses were not related where they were committed weeks apart, involved different victims, and the defendant operated with an accomplice during one crime but not during the other). The government persuasively argues that Irons, Carter, and Horn foreclose Martin’s argument. Those cases, the government contends, stand for the proposition that “the commission of a series of individual, similar crimes does not mean that the resulting multiple convictions are combined for criminal history purposes.” We agree.

As stated above, a defendant seeking to show that offenses are related must prove that the crimes were jointly planned or that commission of one crime entailed committing the other crime or crimes. See Irons, 196 F.3d at 639. Because Martin has not shown that the four car thefts were jointly planned or that committing one theft necessarily entailed the other three, the district court properly refused to treat the offenses as related for purposes of § 4A1.2(a)(2). E.

Under Booker and its progeny, Martin’s sentence must be vacated and his case remanded for resentencing Martin argues, and the government concedes, that this court’s post-Booker cases require us to vacate his sentence and remand the case to the district court for resentencing. See United States v.

Barnett, 398 F.3d 516, 529 (6th Cir.2005) (holding that sentencing under the mandatory Guidelines regime creates a presumption of prejudice that the government must rebut with “clear and specific evidence that the district court would not have ... sentenced the defendant to a lower sentence” if it had treated the Guidelines as advisory); United States v. Oliver, 397 F.3d 369, 378 (6th Cir.2005) (applying Booker to a defendant convicted of methamphetamine-related offenses whose sentence was increased by 12 months on the basis of facts found by the judge at sentencing).

The government has not attempted to rebut the presumption of prejudice in the present case. We therefore vacate Martin’s sentence and remand for resentencing consistent with Booker. III. CONCLUSION For all of the reasons set forth above, we hold that

(1) Martin has not demonstrated that the Sentencing -Commission failed to comply with the unambiguous statutory language when it promulgated the 50% conversion ratio for pseudoephed-rine now found at U.S. Sentencing Guidelines Manual § 2D1.1 cmt. n.10, and (2) the ratio is not “arbitrary and capricious” or otherwise unconstitutional. We also reject Martin’s challenge to the calculation of his criminal history category, but agree that the district court committed a Booker error in sentencing Martin under a mandatory Guidelines regime. Accordingly, we VACATE Martin’s sentence and REMAND the case to the district court for resentencing consistent with Booker.