concurring.
I join the court’s opinion but add one thought. The gymnastics performed in Part II.B.l to show that a two-level increase in Vivit’s offense level is compatible with the ex post facto clause are unnecessary, because the sentencing guidelines are not “laws” within the scope of that clause. See United States v. Seacott, 15 F.3d 1380, 1391-93 (7th Cir.1994) (concurring opinion); cf. Prater v. U.S. Parole Commis*925sion, 802 F.2d 948, 951-52 (7th Cir.1986) (en banc) (parole release guidelines are not “laws” for ex post facto purposes). Many cases say, and a few hold, that changes in the guidelines must be treated like changes in statutory punishments for purposes of the ex post facto clause, and the parties to this case accept that view, but these decisions are unconvincing. The only “law” at issue is the Sentencing Reform Act of 1984, enacted long before Vivit’s crimes. Nothing that has occurred since Vivit committed his acts changed the definition of the offense, its maximum punishment, or the evidence that may be used to support conviction. See Carmell v. Texas, — U.S. -, 120 S.Ct. 1620, 146 L.Ed.2d 577 (2000). When open-ended discretion prevailed before the guidelines, no one would have doubted that Presidents could appoint hard-nosed judges who handed out steep penalties, provided they did not exceed the statutory maximum at the time of the defendant’s deeds. Large swings in effective punishment occurred because of changes in the composition of the bench and prevailing views about the seriousness of particular offenses.
What judges used to do without offending the ex post facto clause, the Sentencing Commission may do. The Sentencing Reform Act moves discretion from the individual judge to the Commission. Because the ex post facto clause does not apply to the judicial branch, see Marks v. United States, 430 U.S. 188, 191, 97 S.Ct. 990, 51 L.Ed.2d 260 (1977), and the Commission is in the judicial branch, see Mistretta v. United States, 488 U.S. 361, 384-97, 109 S.Ct. 647, 102 L.Ed.2d 714 (1989), the effective constraint is the due process clause, which requires judges to refrain from adopting startling interpretations of existing rules. E.g., Bouie v. Columbia, 378 U.S. 347, 84 S.Ct. 1697, 12 L.Ed.2d 894 (1964); Prater, 802 F.2d at 952. Vivit does not contend that the increase in his sentence is so surprising that it violates the due process clause, and given the history of variability in sentencing practices over time (and across judges) such an argument would be untenable. “Changing the guidelines after the commission of a crime does not deprive the criminal of notice of the elements of the offense or the statutory limits of punishment. It may upset the expectations of the few would-be wrongdoers who study sentencing practices to determine their risks — though even a small change in the probability of arrest or prosecution will have a much greater effect on the anticipated punishment than does a change in the guidelines, and no one believes that pouring extra resources into the detection and prosecution of crime violates the ex post facto or due process clause.” Seacott, 15 F.3d at 1392-93. So although my colleagues faithfully implement the complex rules that have sprouted up to limit the damage caused by applying the ex post facto clause to a subject outside its proper domain, I would prefer a shorter path to affirmance. Congress has told courts to use the guidelines in force at the time of sentencing. 18 U.S.C. § 3553(a)(4). That command is constitutional, and I would follow it notwithstanding the United States Attorney’s failure to defend (or even cite) the governing statute.