LIPEZ, Circuit Judge.

Falcón Diómedes Andújar-Arias ("Andújar") claims that his sentence for illegal reentry after deportation was unlawful because the district court declined to account for "unwarranted" sentence disparities as required by 18 U.S.C. § 3553(a)(6). He claims that these "unwarranted" disparities result from the operation of fast-track sentencing programs in other districts. These programs allow some districts whose resources are strained by high immigration workloads to offer diminished charges or sentences in immigration cases in exchange for a defendant's agreement to waive certain procedural rights. In addition, Andújar argues that fast-track programs violate his constitutional right to equal protection. Finally, he argues that the district court erred in treating his prior convictions as sentencing factors. We reject each of these contentions and affirm his sentence.

I.

Andújar was born in the Dominican Re-public in 1968 and entered the United States when he was about eighteen years old to join his mother, who had already immigrated to Peabody, Massachusetts. Shortly thereafter, Andújar became addicted to cocaine and was convicted on cocaine distribution charges in 1993. He was deported in 1996, after serving his sentence.

In 1998, Andújar returned to the United States and assumed the name of "Dwight Braswell," an identity he maintained until the time of his arrest in this case. By 1999, he had become addicted to heroin and was convicted for three heroin offenses between April 1999 and November 2000. According to the Pre-Sentence Report ("PSR") prepared by the Probation Office in connection with the current illegal reentry charge, Andújar devoted himself to recovering from his addictions following his arrest in November 2000. He signed himself into a seven-day detoxification program and then joined an outpatient program with weekly meetings; he became an active member of the Church of the Living God in Woburn, Massachusetts; and he secured a job working for a fellow parishioner repairing floors.

Despite these positive developments, Andújar was convicted of indecent assault and battery in April 2002. Fingerprints taken in connection with that conviction revealed his true identity; as a result, he was subsequently indicted for illegal reentry into the United States after having been deported. See 8 U.S.C. §§ 1326(a)(1), (b)(2). Andújar entered a guilty plea to that charge in July 2005.

The PSR calculated a base offense level of "24," and recommended a three-level reduction for acceptance of responsibility. Using a criminal history category of "V," the PSR concluded that the applicable Guidelines sentencing range was 70-87 months of imprisonment. Andújar did not object to the PSR or its Guidelines calculations. However, he submitted a sentencing memorandum to the district court arguing that: (1) the court could not use prior convictions that were not charged in the indictment or admitted by him in determining his sentence; and (2) a below-Guidelines sentence of forty-eight months would be adequate to achieve the purposes of sentencing prescribed by statute. See 18 U.S.C. § 3553(a) (requiring the court to "impose a sentence sufficient, but not greater than necessary" to meet certain sentencing goals and which enumerates a number of factors that the court "shall consider" in arriving at that sentence). Andújar based his request on several grounds, including his "extraordinary" pre-arrest rehabilitation and the need to avoid unwarranted sentence disparities resulting from the absence of a fast-track program in the District of Massachusetts. Fast-track programs in other parts of the country might allow a defendant in Andújar's circumstances to receive a reduced sentence in exchange for a guilty plea to the charged immigration offense (or to a lesser reentry charge) and a waiver of certain procedural rights. We describe these programs in detail in Section II, infra.

At the sentencing hearing, the district court rejected all of Andújar's contentions. It explained that the Supreme Court's decision in Almendarez-Torres v. United States, 523 U.S. 224, 235, 118 S.Ct. 1219, 140 L.Ed.2d 350 (1998), allows prior felony convictions to be used as sentencing factors so long as they are proven to the court by a preponderance of the evidence. The court found that the Probation Office's investigation was sufficient to satisfy that standard. The court also rejected defendant's fast-track disparities argument, concluding, based on this court's previous decision in United States v. Martin, 221 F.3d 52 (1st Cir.2000), that the different sentencing standards of different districts "was not . . . [a] sufficient justification for a departure." Although the Court acknowledged Andújar's pre-2002 arrest self-rehabilitation, it found that his criminal history and persistent use of a false identity — even throughout his alleged rehabilitation" counterbalanced that rehabilitation and made a variance unwarranted. Considering all relevant factors, the court announced a sentence of 70 months of imprisonment — at the bottom of the Guide-lines range — to be followed by two years of supervised release.

Andújar now appeals, claiming that the court erred in refusing to consider sentence disparities engendered by the presence of fast-track programs in some districts and not others. By failing to consider these factors, Andújar argues, the district court committed an error of law, making his sentence per se unreasonable. He also argues, for the first time on appeal, that fast-track sentencing programs — as currently applied — violate his constitutional right to equal protection under the Fifth and Fourteenth Amendments. Finally, he argues that recent decisions by the Supreme Court cast doubt on the continuing validity of Almendarez-Torres.

II.

We review constitutional questions de novo. Goodrich v. Hall, 448 F.3d 45, 49 (1st Cir.2006). We also review claims of legal error in sentencing de novo, United States v. Wallace, 461 F.3d 15, 33 (1st Cir.2006), but we review the final sentence for reasonableness, regardless of whether it falls inside the Sentencing Guidelines range, as it does here, or outside. United States v. Martinez-Vives, 475 F.3d 48, 54 (1st Cir.2007).

Andújar contends that the district court committed legal error in rejecting his argument that the availability of fast-track sentencing programs in some districts creates "unwarranted" sentence disparities which compelled a below-Guidelines sentence in this case. Such inconsistencies must be taken into account, he argues, under 18 U.S.C. § 3553(a)(6), which requires courts to consider "the need to avoid unwarranted sentence disparities among defendants with similar records who have been found guilty of similar conduct." Andújar's unwarranted disparities argument has three components: first, that fast-track programs generally create unwarranted sentence disparities as defined by § 3553(a)(6); second, that the Attorney General's implementation of fast-track programs has fallen outside Congress's mandate, thereby creating unwarranted disparities; and third, that charge-bargaining fast-track programs, both inherently and as they have been implemented in districts around the country, contravene Congressional intent and create unwarranted disparities. Andújar asks us to conclude that his sentence was unreasonable as a matter of law because the district court failed to vary his sentence from the Guidelines range based on these disparities.

The government contends that the district court did consider Andújar's fast-track argument, but rejected it on the basis that such disparities are not "unwarranted." It further argues that disparities arising from Congressionally approved fast-track programs cannot be considered "unwarranted" under the statute, and that sentence variations arising from charge-bargaining systems are a product of executive discretion and, as such, provide no basis for varying from the Guidelines sentence.

Although Andújar says that he only seeks from us a directive to the district court that it may consider in its sentencing decision the unwarranted sentencing disparities created by the fast-track programs, that is not the logic of his arguments. Taken to their logical conclusion, Andújar's arguments would compel us to find that any sentence in an illegal reentry case that failed to account for fast-track disparities reflected an unwarranted sentencing disparity and hence was per se unreasonable. In essence, he asks us to find that district courts must, as a matter of law, account for such disparities.

We reach the opposite conclusion. We agree with the government that the limited disparities arising from approved fast-track programs reliant on downward sentence departures, as contemplated by Congress, are permissible. We also find that fast-track programs using charge-bargaining measures rather than downward sentence departures are within the discretion of the United States Attorneys and additionally have been authorized by the Attorney General. Therefore, the disparities resulting from fast-track programs, whether a product of downward departures or charge-bargaining programs, are "war-ranted" and may not be considered by a district judge in sentencing as a basis for a variance from a Guidelines sentence pursuant to § 3553(a)(6).

In this case, the district court did not commit legal error by declining to adjust Andújar's sentence on the basis of such disparities. We begin our analysis by briefly discussing the background of fast-track programs.

A. Origins of the Fast-Track Program

The first fast-track program was a product of prosecutorial discretion rather than Congressional legislation. See Rebecca Schendel Norris, Fast-Track Disparities in the Post-Booker World: Re-examining Illegal Reentry Sentencing Policies, 84 Wash. U.L.Rev. 747, 750 (2006) (hereinafter "Schendel Norris"). In the mid-1990s, the United States Attorney's Office for the Southern District of California allowed some offenders charged with illegal reentry under 8 U.S.C. § 1326(b)"which carried a maximum sentence of five or fifteen years, depending on the defendant's criminal history" to plead guilty to a lesser reentry offense, § 1326(a)"which carried only a two-year maximum sentence" if the offender agreed to waive indictment, plead guilty, waive appeal of all sentencing issues, stipulate to the two-year sentence, and further agreed not to seek downward adjustments or departures. Id.; see United States v. Estrada-Plata, 57 F.3d 757, 759 (9th Cir.1995). Similar programs were adopted by federal prosecutors in Texas, New Mexico, Arizona and in other California districts. United States v. Martinez-Flores, 428 F.3d 22, 25 (1st Cir. 2005).

In 2003, Congress passed the Prosecutorial Remedies and Other Tools to End the Exploitation of Children Today (PROTECT) Act, Pub.L. No. 108-21, 117 Stat. 650 (2003) (codified in sections 18, 28, and 42 of the U.S.Code), which explicitly authorizes fast-track programs. The Act seeks to "substantially reduce" the "incidence of downward departures" in sentencing, PROTECT Act, § 401(m)(2)(A), 117 Stat, at 675, and also creates an explicit, regulated exception for fast-track programs. Id. § 401(m)(2)(B), 117 Stat, at 675. To that end, the Act required the U.S. Sentencing Commission to promulgate a "policy statement authorizing a downward departure of not more than 4 levels if the Government files a motion for such departure pursuant to an early disposition program authorized by the Attorney General and the United States Attorney." Id. Thus, the language of the PROTECT Act suggests that Congress intended to authorize a narrow departure scheme for fast-track programs while otherwise restricting the use of downward departures.

There is additional evidence of Congress's intent to establish a narrow exception in a House of Representatives report issued prior to the enactment of the PROTECT Act, in connection with a companion bill, the Child Abduction Prevention Act of 2003 (also known as the "Feeney Amendment"). The House Report to the Feeney Amendment endorsed a formal program of limited departures from the Sentencing Guidelines for "those particular classes of offenses (such as illegal reentry) whose high incidence within the district has imposed an extraordinary strain on the resources of that district as compared to other districts." See 149 Cong. Rec. H2405, H2421 (daily ed. Mar. 27, 2003) (amendment offered by Rep. Feeney). The report explicitly acknowledged that sentence disparities would result between fast-track districts and non-fast track districts; however, it specified that such disparities "do not confer authority to depart downward on an ad hoc basis in individual cases." Id., at H2421; see, e.g., United States v. Mejia, 461 F.3d 158, 163 (2nd Cir.2006).

A contemporaneous report to Congress from the U.S. Sentencing Commission also noted the tension between § 3553(a) and the PROTECT Act, §§ 401(m)(2)(A) and (B), explicitly highlighting that sentencing disparities would result:

Defendants sentenced in districts without authorized early disposition programs . . . can be expected to receive longer sentences than similarly-situated defendants in districts with such programs. This type of geographical disparity appears to be at odds with the overall . . . goal of reducing unwarranted sentencing disparity. . . . Furthermore, sentencing courts in districts without early disposition programs, particularly those in districts that adjoin districts with such programs, may feel pressured to employ other measures" downward departures in particular" to reach similar sentencing outcomes for similarly situated defendants. This potential response by sentencing courts could undermine the goal of the PROTECT Act to reduce the incidence of downward departures.

U.S. Sentencing Commission, Report to Congress: Downward Departures from the Federal Sentencing Guidelines at 66-67 (Oct.2003), available at http://www.ussc. gov/departrpt03/departrpt03.pdf. The report concluded that review of fast-track program performance might be appropriate in the future to provide additional guidance to the courts on these issues. Id. at 67.

Shortly after Congress passed the PROTECT Act, and pursuant to the legislation, the Attorney General issued a memorandum outlining the elements required for any fast-track program and explaining the conditions necessary to obtain the Attorney General's authorization for such a program. A month later, twenty-six fast-track applications were submitted in fifteen districts; thirteen specifically addressed the crime of illegal reentry. Schendel Norris, supra, at 757. Those approved were: Arizona, Idaho, Nebraska, New Mexico, North Dakota, Oregon, all four California districts, the Southern and Western Districts of Texas, and the Western District of Washington. Id. at 757-58. At the time of Andújar's sentencing, Massachusetts had no fast-track program.

B. The Statutory Challenge

Whether sentence disparities attributable to the existence of fast-track programs are "unwarranted" is a question of first impression in this circuit. Every other circuit except the D.C. Circuit has addressed this issue, and each has reached the same conclusion. Because sentence disparities resulting from the presence of fast-track programs in some districts and the absence thereof in others are not un-warranted, a court's failure to adjust a sentence to compensate for such disparities does not make that sentence unreasonable. Mejia, 461 F.3d at 163; United States v. Vargas, 164 F.3d 94, 98-99 (3rd Cir.2007); United States v. Perez-Pena, 453 F.3d 236, 243 (4th Cir.2006); United States v. Aguirre-Villa, 460 F.3d 681, 683 (5th Cir.2006); United States v. Hernandez-Fierros, 453 F.3d 309, 314 (6th Cir.2006); United States v. Martinez-Martinez, 442 F.3d 539, 542 (7th Cir.2006); United States v. Sebastian, 436 F.3d 913, 916 (8th Cir.2006); United States v. Marcial-Santiago, 447 F.3d 715, 718 (9th Cir. 2006); United States v. Martinez-Trujillo, 468 F.3d 1266, 1268-69 (10th Cir.2006); United States v. Anaya Castro, 455 F.3d 1249, 1252 (11th Cir.2006). The Fourth Circuit has taken an additional step, finding that a district court's below-Guidelines sentence was unreasonable as a matter of law because the court's justification for such variance, namely a desire to avoid fast-track sentence disparities, was improper. Perez-Pena, 453 F.3d at 241 ("If a district court provides an inadequate statement of reasons or relies on improper factors in imposing a sentence outside the properly calculated advisory guideline range, the sentence will be found unreasonable and vacated."). In reaching these conclusions, courts have relied on the same two-step rationale: "Congress must have thought the disparity warranted when it authorized early disposition programs without altering § 3553(a)(6)," Aguirre-Villa, 460 F.3d at 683, and it is "within the province of the policymaking branches of government to determine that certain disparities are warranted, and thus need not be avoided." Sebastian, 436 F.3d at 916.

While we have not squarely addressed this issue before, we have twice expressed in dicta our view that sentence disparities arising from fast-track programs are not "unwarranted." See United States v. Jiménez-Beltre, 440 F.3d 514, 519 (1st Cir.2006) (en banc) ("[The existence of fast-track programs] certainly permits disparities but they are the result of a congressional choice made for prudential reasons, implicitly qualifying the general aim of equality."); United States v. Martinez-Flares, 428 F.3d 22, 30 n. 3 (1st Cir.2005) ("It is arguable that even post- Booker, it would never be reasonable to depart downward based on disparities between fast-track and non-fast-track jurisdictions given Congress' clear (if implied) statement in the PROTECT Act provision that such disparities are acceptable."). However, in each case, we did not have to reach the ultimate substantive issue. Jiménez-Beltre, 440 F.3d at 519 ("[T]he district court ruled that the defendant had not furnished a factual basis for assessing the extent of the disparities or provided a reason why to take them into account. . . . In declining to alter the sentence on this ground, the district court did not act unreasonably."); Martinez-Flores, 428 F.3d at 30 n. 3 (dismissing the argument on plain error grounds). We now join the other circuit courts and make that holding explicit. Because Congress has authorized fast-track programs with the understanding that such programs would create sentencing disparities, we find that such disparities are not unwarranted as a matter of law. Therefore, a sentence imposed without consideration of such disparities under § 3553(a) is not legally erroneous or, as defendant would put it, per se unreasonable. In passing the PROTECT Act, Congress determined that the benefits of maintaining high prosecution rates and relieving the case management burdens on many districts outweighed the costs of lower sentences. See, e.g., Martínez-Martínez, 442 F.3d at 542. As a court we are not in a position to second guess that determination.

Seeking to convince us otherwise, Andújar argues that whether disparities are warranted within the context of § 3553(a)(6) must depend only upon the nature of the offense and the offender, and that the needs of the government — including the burdens placed upon its resources by the high incidence of particular crimes — cannot serve as a legitimate basis for interdistrict disparities. Andújar supports this claim by pointing out that § 3553(a) itself, in listing the factors the court must consider when imposing a sentence, focuses on the nature of the offense and the offender and never mentions governmental resources. In addition, he argues that the legislative history behind the "unwarranted disparity" language in § 3553(a)(6) also indicates no expectation that scarce government resources could justify sentence disparities. However, this argument ignores Congress's authority to modify that focus in the PROTECT ACT itself. We agree with the Ninth Circuit that

when Congress passed the PROTECT Act, it did so with knowledge that 18 U.S.C. § 3553(a)(6) was directing sentencing courts to consider the need to avoid unwarranted sentencing disparities. By authorizing fast-track programs without revising the terms of § 3553(a)(6), Congress was necessarily providing that the sentencing disparities that result from these programs are warranted and, as such, do not violate § 3553(a)(6).

C. The As-Applied Challenge

Andújar also presents an as-applied challenge to fast-track programs, contending that even if disparities occasioned by Congressionally approved fast-track programs are "warranted," disparities arising from fast-track programs as they are currently implemented are "un-warranted" because they deviate from Congressional intent. Specifically, Andújar argues that the Attorney General's decisions to approve and deny fast-track applications in individual districts have defied Congressional intent. Because the Attorney General's actions are at odds with Congressional intent, Andújar concludes that the disparities resulting from this system have not in fact been "warranted" by Congress. To our knowledge no other circuit has squarely decided such a claim and the district court did not explicitly address the as-applied challenge raised below.

To support his argument that the Attorney General has defied Congressional intent in its implementation of the fast-track programs, Andújar marshals substantial statistical evidence, including data prepared by the United States Sentencing Commission Office of Policy Analysis, to show that the Attorney General has approved fast-track programs in districts where immigration cases make up a low percentage of the district's total caseload and has denied programs to at least one district where they make up a high percentage of the caseload. The Sentencing Commission Office's 2003 report states that, for that year, immigration-related sentences accounted for 21.9% of the national total of terms imposed. They accounted for 54% of the sentences handed down in the Southern District of California, 54% in the Southern District of Texas, and 53% in the District of Arizona. However, they accounted for only 6% of the sentences in the Western District of Washington and 11% in the District of Nebraska, both of which maintained fast-track programs. See United States Sentencing Commission, Office of Policy Analysis, 2003 Datafiles, generally available at http://www.ussc.gov/linktojp.htm (compiling annual sentencing statistics and reporting the number of immigration sentences in relation to the number of total sentences in Table 5 of each file). In addition, Andújar notes that the Attorney General denied a fast-track program to the District of Utah, where immigration cases comprised 27% of the sentences imposed in 2004. Id. According to Andújar, this data shows that these fast-track programs, as regulated by the Attorney General, have strayed from the intent of Congress as expressed by the House Report, which stated that fast-track programs be approved for "particular classes of offenses (such as illegal reentry) whose high incidence within [a] district has imposed an extraordinary strain on the resources of that district as compared to other districts." 149 Cong. Rec. at H2421 (daily ed. Mar. 27, 2003).

In a further attempt to support this argument, Andújar compares the number of immigration cases per Assistant United States Attorney ("AUSA") in each district, and finds that these numbers also fail to correlate with the incidence of fast-track approval. Based on these statistics, Andújar concludes that fast-track programs, as they are implemented currently, deviate from Congressional intent, and that they thereby create arbitrary sentence disparities across districts.

The government responds that the statistical relationships cited by Andújar present, at best, an incomplete picture of the resource demands imposed by illegal reentry cases on each of the districts studied. When Congress authorized the Attorney General to approve fast-track programs, it did not restrict the criteria for such programs to the mere proportion of each district's caseload comprised of immigration cases. Instead, it allowed the Attorney General to evaluate such factors as the number and type of immigration cases encountered by each district and the quantity and variety of resources each district could marshal in response. See, e.g., Ashcroft Memorandum, at *l-2.

Andújar's statistic of choice is the number of immigration cases as a percentage of total cases within each district. However, this measure of the resource burden resulting from immigration caseloads obscures the fact that different types of cases may require different types and amounts of resources. In addition, the statistics dividing immigration cases by the number of AUSAs lump together attorneys handling civil and criminal cases and do not account for AUSAs assigned to particular programs, such as the Organized Crime and Drug Enforcement Task Force. Even a district with significant personnel can find resources tight when only a small subset of its workforce is available for immigration cases. Likewise, a district with a high overall number of immigration cases may possess the resources to manage that caseload without incident, or it may determine that, though stretched, it would rather continue to employ tougher sentences than apply for a fast-track pro-gram. In addition, Andújar's statistics do not account for sudden influxes of immigration cases. The government notes that, while Nebraska is not a border state, it faced a sudden increase in its immigrant population over the relatively short period between 2000 and 2006. The government thus argues that if Andújar's statistics were adjusted to account for these fine-grained distinctions between districts, the seeming anomalies he relies on in labeling interdistrict disparities "unwarranted" would likely disappear.

We find the government's arguments compelling. The text of the Feeney Amendment requires that fast-track programs be established to address "extraordinary resource constraints, not typical of most districts, associated with the disproportionately high incidence of illegal reentry or other specific offenses within a particular district. . . ." 149 Cong. Rec. at H2420 (daily ed. Mar. 27, 2003). Although Andújar argues otherwise, such judgments regarding resource allocation can rarely be reduced to a single variable or calculation. If they could, Congress would have little reason to delegate such decisions to an executive agency, such as the Attorney General's Office. As a product of such delegation, the Attorney General's decisions regarding questions of implementation and resource allocation deserve significant deference by courts, especially when they address the enforcement of the nation's criminal laws. See, e.g., United States v. Armstrong, 517 U.S. 456, 464, 116 S.Ct. 1480, 134 L.Ed.2d 687 (1996). We therefore decline to second-guess the Attorney General's methods for determining which districts warrant fast-track programs.

D. Charge-Bargaining Programs

Andújar raises two additional fast-track claims, each challenging the use of charge-bargaining programs to obtain comparable sentence and efficiency results. Rather than relying on government initiated downward sentence departures, several districts engage in charge-bargaining practices, which allow U.S. Attorneys to reduce the charges filed against an illegal reentry defendant in return for a waiver of certain procedural rights. In the Southern District of California, where the first charge-bargaining program was established in the mid-1990s, the district's U.S. Attorneys could charge an illegal reentry defendant with one count under 8 U.S.C. § 1326(a), which carries a two-year maximum sentence, instead of one count under § 1326(b), which at the time carried a maximum penalty of five or fifteen years, depending on the defendant's criminal history. Other districts have adopted similar charge-based schemes.

First, Andújar asserts that these charge-bargaining programs were not expressly authorized by Congress and, similarly, operate outside the intent of Congress. As a result, Andújar argues the disparities resulting from such programs are unwarranted. Second, Andújar argues that these charge-bargaining programs, as implemented, have produced sentence reductions greater than those intended by Congress and the Attorney General. Such disparities, therefore, have not been approved by Congress and remain "unwarranted" under § 3553(a)(6). We address these arguments in turn.

1. Facial Challenge

In pursuing his facial challenge, Andújar relies on the text of the PROTECT Act and the legislative history high-lighting Congressional intent, neither of which, he claims, authorized the use of charge-bargaining methods to obtain the specific, limited departures Congress authorized. Essentially, he argues that the failure of Congress to include such programs in the text of the statute implies that such programs are per se contrary to Congressional intent.

The PROTECT Act did not explicitly address the issue of charge-bargaining. The decision whether to charge a defendant and, if so, what crime(s) to charge, traditionally resides with the Attorney General and the individual prosecutor. See United States v. LaBonte, 520 U.S. 751, 762, 117 S.Ct. 1673, 137 L.Ed.2d 1001 (1997) ("[T]he discretion a prosecutor exercises when he decides what, if any, charges to bring against a criminal suspect . . . is an integral feature of the criminal justice system, and is appropriate, so long as it is not based upon improper factors."); Bordenkircher v. Hayes, 434 U.S. 357, 364, 98 S.Ct. 663, 54 L.Ed.2d 604 (1978) ("[T]he decision whether or not to prosecute, and what charge to file or bring before a grand jury, generally rests entirely in [the prosecutor's] discretion."); see also United States v. Rodriguez, 162 F.3d 135, 151 (1st Cir.1998). As a result, Congress and the courts play a minimal role in regulating such charging decisions, despite their inevitable impact on sentencing. Wayte v. United States, 470 U.S. 598, 607, 105 S.Ct. 1524, 84 L.Ed.2d 547 (1985) ("[The prosecutor's] broad discretion rests largely on the recognition that the decision to prosecute is particularly ill-suited to judicial review."); United States v. Smith, 178 F.3d 22, 26 (1st Cir.1999) ("[W]e note that 'the exercise of prosecutorial discretion, at the very core of the executive function, has long been held presumptively unreviewable.'" (quoting In re Sealed Case, 131 F.3d 208, 214 (D.C. Cir.1997))). The PROTECT Act's silence on this issue, therefore, is not evidence of Congress's intent to preclude such charge-bargaining programs. Cf. Martaínez-Flores, 428 F.3d at 28 (noting that even if the Attorney General had not authorized fast-track programs under the PROTECT ACT, "prosecutors still would have been free to achieve the same outcomes via the bargaining process").

Further, although the PROTECT Act does not expressly address charge-bargaining programs, there is contemporaneous documentation which indicates that both Congress and the Attorney General were aware of several fast-track charge-bargaining programs and approved of them. The House Report on the Feeney Amendment recognized that several districts had already instituted such programs years earlier. See 149 Cong. Rec. at H2421 (daily ed. Mar. 27, 2003). These pre-existing programs relied on charge-bargaining measures. See Schendel Norris, supra, at 750. In his 2003 memo to the United States Attorneys outlining the principles for implementing a fast-track program, Attorney General John Ashcroft specifically referenced "charge-bargaining" fast-track programs, noting that, if established, they should also provide for sentencing reductions commensurate with the downward departure system discussed in the memo and the PROTECT Act itself. Ashcroft Memorandum, supra, at *3. Finally, in 2004, Deputy Attorney General James Comey distributed a memo re-authorizing sixteen existing fast-track programs, including several that used charge-bargaining practices, and confirming the Attorney General's policy that required districts relying on charge-bargaining practices to obtain the approval of the Attorney General prior to implementation. Comey Memorandum, supra, at 1. The inclusion of this confirmatory language by Deputy Comey suggests that the Attorney General believed, notwithstanding the text of the PROTECT Act, that such schemes were within the province of the executive branch and unaffected by the statute.

That many charge-bargaining programs predated the PROTECT Act, and were subsequently authorized by the Attorney General after the act's passage, further supports the proposition that charge-bargaining systems are not at odds with Congress's directives. See Interim Report, supra, at 31 (noting that some fast-track programs involve charge-bargaining to restrict the defendant's maximum statutory penalty).

2. As-Applied Challenge

Finally, Andújar argues that even if charge-bargaining programs are not per se contrary to Congressional intent, many such programs are currently operated outside the bounds of the PROTECT Act and U.S.S.G. § 5K3.1, therefore creating un-warranted disparities for purposes of § 3553(a)(6). Specifically, he argues that some programs are at odds with Congressional intent because they provide more than a four-level departure from the Guidelines. To support this claim, he relies on two items. First, Andújar cites the decision in a single district court case, United States v. Medrano-Duran, which examined how charge-bargaining fast-track systems currently operate in many districts. 386 F.Supp.2d at 946 ("The fast track districts that rely on charge-bargaining use methodologies that permit far greater sentence reductions tha[n] contemplated by Congress' directive in the PROTECT Act and the Sentencing Commission's policy statement in § 5K3.1."). Next, he offers the Interim Staff Report on Immigration Reform and the Federal Sentencing Guidelines, which notes that some fast-track programs charge an illegal reentry defendant with two counts of 8 U.S.C. § 1325 rather than a single count under 8 U.S.C. § 1326, which carries a greater penalty. Interim Report, supra, at 31. A term of 30 months' imprisonment under § 1325 represents an eight-level departure for a defendant who otherwise shares the same criminal history and sentencing characteristics as Andújar. See U.S.S.G. ch. 5, pt. A (sentencing table). Andújar thus argues that disparities resulting from these unauthorized charge-bargaining schemes are clearly "unwarranted."

In the face of Andújar's evidence on this point, the district court declined to find that charge-bargaining programs in fast-track districts had produced sentences outside the parameters set by Congress. We agree with the district court's decision. When a party contests a sentence based on the failure of the court to consider a specific factor, that party must provide the factual basis for its argument. Jiménez-Beltre, 440 F.3d at 519. This obligation is most significant when the party has raised an as-applied challenge, which necessarily requires a concrete and developed factual record for the court to consider. See Sam Ali, Inc. v. Ohio Dept. of Liquor Control, 158 F.3d 397, 399-400 (6th Cir.1998) (declining consideration of an as-applied, statutory challenge because the plaintiffs had failed to develop a sufficient factual record); McGuire v. Reilly, 260 F.3d 36, 47-48 (1st Cir.2001) (noting, in the context of the First Amendment, that plaintiffs remain free to challenge the Act, as applied, in a concrete factual setting) (emphasis added); Carey v. Wolnitzek, No. 3:06-36-KKC, slip op. at 8, 2007 WL 2726121 (E.D.Ky. Sept. 17, 2007) (rejecting an as applied challenge because the factual basis of plaintiffs claim remained speculative).

As the government notes, Andújar's sole evidence that these programs operate outside the bounds of the PROTECT Act was a government submission in another case in another district, stating that, by their terms, charge-bargaining fast-track programs can lead to a sentence reduction greater than the four levels specifically authorized by Congress in the PROTECT Act. Andújar provided no evidence that charge-bargaining has, in fact, resulted in any sentence reduction equivalent to a five level departure (or more) in any particular case. Moreover, the government notes that the Ashcroft Memorandum specifically provides that charge-bargaining fast-track programs "should provide for sentencing reductions that are commensurate with" the four-level reduction authorized under U.S.S.G. § 5K3.1. Ashcroft Memorandum, supra, at *3. Without evidence that prosecutors in individual districts have deviated from this requirement, Andújar's argument that charge-bargaining fast-track programs result in unwarranted sentence disparities must fail, and we need not address whether such outcomes would constitute unwarranted disparities under § 3553(a)(6).

Moreover, even if we agreed with the premise of Andújar's argument — that charge-bargaining fast-track programs fall outside the parameters of the PROTECT Act — we fail to see how his argument justifies the remedy he seeks. That is, we do not believe that the appropriate response to unauthorized prosecutorial action in some fast-track districts is to require judges in other districts to consider ad hoc downward variances that would equalize sentences at the level resulting from the unauthorized actions. See Perez-Pena, 453 F.3d at 243, n. 3 ("[A]llowing sentencing courts to determine whether they should sentence non-fast-track defendants as if they had been fast-tracked would produce 'unwarranted sentence disparities' between similarly situated non-fast-track defendants, some of whom would benefit from the existence of others' fast-track deals and some of whom would not."). To find otherwise would be to allow judges to ignore the decisions of United States Attorneys and the Attorney General to not have such a program in a specific district and to disregard Congressional intent on the proper scope of such programs. See Albert Llosas Barrueco, Fast-Tracking United States v. Booker: Why Judges Should Not Fix Fast Track Disparities, 6 Conn. Pub. Int. L.J. 65, 105 (2006) ("[J]udges are ill-advised to impose Fast Track programs by judicial mandate. Such decisions have the inescapable result of invading the province of the prosecutor by delivering the discretion to charge and establish enforcement policies into the hands of judges."). If Congress agrees with Andújar's criticism of the fast-track regime as applied, it is in the best position to reshape the system.

E. Equal Protection

Andújar also argues that fast-track sentence disparities violate his constitutional right to equal protection under the Fifth and Fourteenth Amendments. While our review of constitutional claims is generally plenary, United States v. Proctor, 166 F.3d 396, 401 (1st Cir.1999), we review only for plain error because Andújar raises this argument for the first time on appeal. See United States v. Olano, 507 U.S. 725, 732-33, 113 S.Ct. 1770, 123 L.Ed.2d 508 (1993).

Andújar cites no case law in support of his equal protection claim; any error, therefore, could not qualify as "plain." See Id. at 734, 113 S.Ct. 1770. Moreover, he has failed to show any constitutional deficiency. Because his challenge does not implicate a suspect class, Andújar must show that there is no rational basis for the legislated distinction — in this case between the treatment of illegal reentry defendants in fast-track districts and of similar defendants in Massachusetts. Naeem v. Gonzales, 469 F.3d 33, 38 (1st Cir.2006) (citing Heller v. Doe, 509 U.S. 312, 320, 113 S.Ct. 2637, 125 L.Ed.2d 257 (1993)). Where, as here, Congress has made its purpose explicit, Andújar must show the statutory purpose has been implemented in a manner "so attenuated as to render the distinction arbitrary or irrational." City of Cleburne v. Cleburne Living Ctr., 473 U.S. 432, 446, 105 S.Ct. 3249, 87 L.Ed.2d 313 (1985). Andújar has failed to meet these burdens.

Indeed, we have already denied a claim presenting a facial equal protection attack on fast-track programs. Melendez-Torres, 420 F.3d at 52-53 (concluding that the incidence of fast-track programs in some states and not others could rationally be related to different incidences of crime across districts, different prosecutorial needs, or a determination that the absence of such a program increases deterrence). We believe that these justifications apply with equal strength in this case. Further, for the same reasons we rejected Andújar's as-applied challenge to fast-track disparities, we find that such programs have not been implemented in a manner that is "so attenuated as to render the [Congressionally established] distinction arbitrary or irrational." Cleburne, 473 U.S. at 446, 105 S.Ct. 3249. We conclude, therefore, that fast-track programs, as applied, are rationally related to the Congressional goal of protecting prosecutorial resources.

III.

We may quickly dispatch Andújar's claim that Almendarez-Torres is no longer persuasive precedent. We rejected a similar argument in United States v. Jiménez-Beltre, 440 at 520 ("Whatever the continuing validity of Almendarez-Torres, we have previously held that we are bound to follow it until it is expressly overruled.") (citation omitted), and have since repeated that conclusion many times. See, e.g., United States v. Shoup, 476 F.3d 38, 46 n. 5 (1st Cir.2007); United States v. Coplin, 463 F.3d 96, 105 (1st Cir.2006); United States v. Peralta, 457 F.3d 169, 172 (1st Cir.2006). We find nothing new in Andújar's argument that requires us to reopen this issue.

Affirmed.