Emerson Molt has asked us to reconsider our refusal to grant him bail under Fed.R.App.P. 9(b) pending his appeal to this court from his conviction and eight-year prison sentence for drug crimes (the district court having refused to grant him bail pending appeal). Although we adhere to our decision, we have thought it best to publish our reasons for doing so because the main issue he raises--the constitutionality of applying the standard for bail in the recently enacted federal criminal code to persons convicted of crimes committed before October 12, 1984, when the new standard took effect (United States v. Angiulo, 755 F.2d 969 (1st Cir.1985); see United States v. Gavrilovic, 551 F.2d 1099, 1103 (8th Cir.1977))--is bound to recur.

Until October 12, the rule was that bail pending appeal could be denied (in the absence of any danger of flight or to public safety) only "if it appears that an appeal is frivolous or taken for delay." 18 U.S.C. Sec. 3148 (repealed). Under the new code, bail can be granted only if the court finds "that the appeal is not for purpose of delay and raises a substantial question of law or fact likely to result in reversal or an order for a new trial." Bail Reform Act of 1984, Sec. 203, 98 Stat. 1981-82, to be codified at 18 U.S.C. Sec. 3143(b)(2). This is a more stringent standard, since an appeal could be at once nonfrivolous yet insubstantial in the sense of quite unlikely to succeed; and the parties are agreed that Molt's appeal is not frivolous, so that if the old standard rather than the new were applicable he clearly would be entitled to bail.

Before we take up the constitutional issue we shall consider Molt's other, nonconstitutional ground for reconsideration, which is that the district court misapplied the new standard. At Molt's trial the judge had allowed narcotics users to testify to conversations between Molt and his alleged coconspirators without making a determination that those coconspirators were unavailable to be cross-examined. Molt argues that this denied his Sixth Amendment right to confront the witnesses against him, and hence that the appeal "raises a substantial question of law ... likely to result in reversal or an order for a new trial." But in this circuit "challenges to co-conspirators' statements should be based on the requirements of Rule 801(d)(2)(E) [of the Federal Rules of Evidence], not on the Sixth Amendment." United States v. Williams, 737 F.2d 594, 610 (7th Cir.1984). Although some other circuits take a different view, see, e.g., United States v. Inadi, 748 F.2d 812, 818-19 (3d Cir.1984), we are unlikely to reexamine our own position on the question in Molt's appeal, especially after having expressly declined to do so in United States v. Xheka, 704 F.2d 974, 987 n. 7 (7th Cir.1983), before Williams was decided. Rule 801(d)(2)(E) makes "a statement by a coconspirator of a party during the course and in furtherance of the conspiracy" admissible against that party--period. There is no requirement of showing that the out-of-court declarant is unavailable, or that there is some special reason to think the evidence reliable. Compare Rule 801(d) with Rule 804(b)(5).

Since the only question Molt raises that has any possible merit cannot be viewed as a substantial--an open--question in this court, we need not decide what the words "likely to result in reversal or an order for a new trial" in the new statute add to "substantial question" or whether the standard means the same thing when applied by the district court and when applied by the court of appeals. Maybe, as the Third, Ninth, and Eleventh Circuits recently have held, the issue for the district court ought to be just whether the defendant has raised a substantial question that will result in a reversal if the defendant's position is adopted. See United States v. Miller, 753 F.2d 19, 23-24 (3d Cir.1985); United States v. Handy, 753 F.2d 1487 (9th Cir.1985) (per curiam); United States v. Giancola, 754 F.2d 898 (11th Cir.1985) (per curiam). The reason is to avoid putting the district judge in the anomalous position of being unable to admit a defendant to bail pending appeal unless the judge thinks he is likely to be reversed, which would mean, thinks he probably ruled erroneously--in which event he should have granted a new trial without putting the defendant to the bother of an appeal. For us, though, faced with an application to us for bail under Fed.R.App.P. 9(b), the issue--as we held in United States v. Oliver, 683 F.2d 224, 235 (7th Cir.1982), under the old standard for bail--is not what the district judge should have done but what we should do. In other words, are we likely to reverse? We are not, as we said. But whether this is the right standard for us, or whether we should apply the same standard as the district judge--and if so whether that is the standard which the Third, Ninth, and Eleventh Circuits have now agreed on--the result must be the same. A substantial question "is a 'close' question or one that very well could be decided the other way." United States v. Giancola, supra, at 901. This one is not--at least in this circuit.

We come then to the constitutional issue. Molt argues that the new standard for bail pending appeal cannot be applied to him without violating Article I, section 9 of the Constitution ("No ... ex post facto Law shall be passed [by Congress]"). The argument has persuaded one district judge in this circuit, see United States v. Cirrincione, 600 F. Supp. 1436, 1443-45 (N.D.Ill.1985), but has been rejected by others (see, besides the opinion of the district judge in this case, United States v. Giangrosso, 605 F. Supp. 697, (N.D.Ill. 1985); cf. United States v. DiVarco, 602 F. Supp. 1029, 1036-1038 (N.D.Ill.1985)), and by the only other court of appeals to consider it in a reported decision, see United States v. Miller, supra, 753 F.2d at 21; see also United States v. Crabtree, 754 F.2d 1200 (5th Cir.1985) (opinion of one judge, in chambers). State courts have split on the cognate question whether retroactive application of statutes denying a right to bail pending appeal violates the prohibition in Article I, section 10 against the states' enacting ex post facto laws. Compare Greene v. State, 238 So. 2d 296, 300-01 (Fla.1970), with State ex rel. Dorton v. Circuit Court, 274 Ind. 373, 412 N.E.2d 72, 73-75 (1980).

We think the change in the standard for bail pending appeal is not an ex post facto law. "Even though it may work to the disadvantage of a defendant, a procedural change is not ex post facto." Dobbert v. Florida, 432 U.S. 282, 293, 97 S. Ct. 2290, 2298, 53 L. Ed. 2d 344 (1977); see, e.g., United States ex rel. Massarella v. Elrod, 682 F.2d 688, 689 (7th Cir.1982) (extending the statute of limitations); Dunn v. Maggio, 712 F.2d 998, 1001-02 (5th Cir.1983) (per curiam) (repealing pardon statute). It would be odd to think that by committing a crime, a person acquired an indefeasible right to be tried for it in a particular way, or that in deciding whether to commit a crime the prospective criminal will have regard to the particulars of the procedures for the trial and appeal of criminal cases. And it would be both odd and confusing to make courts use two sets of procedures at the same time in criminal cases, depending on the date when the crime was committed. Maybe this is why the courts are much more critical of slight increases in formal punishment than of substantial shifts in the balance of procedural advantage from defendant to prosecutor--why, for example, applying retroactively a law requiring solitary confinement of prisoners on death row violates the ex post facto clause, but not a law giving the state for the first time the right to appeal from a judgment in favor of a criminal defendant. Compare Ex parte Medley, 134 U.S. 160, 10 S. Ct. 384, 33 L. Ed. 835 (1890), with Mallett v. North Carolina, 181 U.S. 589, 593-97, 21 S. Ct. 730, 732-33, 45 L. Ed. 1015 (1901).

It is true that the practical effect of denying bail on appeal is to accelerate the defendant's punishment; and for any defendant who prefers present to future liberty, the effect is to make the punishment more severe. But by this token, the Speedy Trial Act, 18 U.S.C. Secs. 3161 et seq., violated the prohibition against ex post facto punishments by accelerating the date at which a convicted defendant begins to serve his sentence (it accelerates the date at which the innocent defendant is acquitted, too, but the vast majority of federal criminal trials end in conviction, as shown in Table D4 in the annual reports of the Director of the Administrative Office of the United States Courts)--a proposition no one asserts, or believes. It is very doubtful that criminal defendants have settled expectations regarding the precise time when they will begin to serve a sentence if convicted for crime--though of course the same thing could be said with regard to many of the changes that have been held to be impermissible when applied ex post facto.

If Molt is right, any law that decreased the amenities of prison life, as by reducing the appropriations for the Bureau of Prisons, would, by making punishment more onerous, violate the prohibition against ex post facto laws; and that is not the law either. If Molt is right, Elrod is wrong. A law that extends the statute of limitations increases the expected cost of punishment to the criminal, by making it more likely that he will be caught and charged before the statute of limitations has run, just as bringing the punishment closer increases that cost.

Although the prohibitions against ex post facto laws cannot be evaded just by calling a change in law procedural, only the "alteration of a substantial right" is forbidden, Weaver v. Graham, 450 U.S. 24, 29 n. 12, 101 S. Ct. 960, 964 n. 12, 67 L. Ed. 2d 17 (1981); and merely accelerating the effective date of a convicted defendant's sentence does not alter a substantial right. It is all a matter of degree, of course, which may be why, as Professor Tribe has argued, see American Constitutional Law 483-84 (1978), the cases do not compose a harmonious pattern. It is hard to reconcile Mallett v. North Carolina, supra, with a case decided a few years earlier which held that substituting an 8-man for a 12-man jury violated the ex post facto clause: Thompson v. Utah, 170 U.S. 343, 352-55, 18 S. Ct. 620, 623-24, 42 L. Ed. 1061 (1898). Still, the presumption is against construing a procedural change as an ex post facto law, and must carry the day in the absence of a stronger showing than made in this case that the change works an increase in punishment. For though we have been speaking of the Bail Reform Act as if it had abolished the right to bail pending appeal, it did no such thing; it merely made it harder to get bail pending appeal--and maybe not much harder, if the Third, Ninth, and Eleventh Circuits are correct. The change in the balance of advantages against the defendant is too slight to bring the change within the scope of the ex post facto clause.

The petition for reconsideration of our denial of the motion to admit the defendant to bail pending appeal is therefore

DENIED.