MOORE, J., delivered the opinion of the court, in which ENGEL, J., joined. MERRITT, J. (pp. 423-26), delivered a separate dissenting opinion.
MOORE, Circuit Judge.
These four consolidated cases present the question of whether the administrative exhaustion requirement of the Prison Litigation Reform Act of 1996,1 Title VIII of the Omnibus Consolidated Rescissions and Appropriations Act of 1996, Pub.L. No. 104-134, 110 Stat. 1321, applies to prisoner civil rights cases that were pending before this court when the Act took effect. For the reasons discussed below, we hold that the administrative exhaustion requirement does not apply to appeals already pending on the enactment date.
I. FACTS AND PROCEDURAL HISTORY
We address the merits of plaintiffs’ claims in separate opinions; it is therefore sufficient for the purposes of this opinion to [417]*417note that each of the four plaintiffs is an inmate in one of the four state prison systems in this circuit2 who filed a pro se suit in the appropriate United States district court under 42 U.S.C. § 1983 challenging the conditions of his confinement. None of the inmates contends that he has exhausted all of the administrative remedies which were available at the time of the alleged violations.3 The cases were filed between 1993 and 1995 and were dismissed by the district courts in 1995. The plaintiffs filed timely appeals.
While these appeals were pending in this court, Congress passed the Prison Litigation Reform Act of 1996 [hereinafter “PLRA” or “Act”], which requires inter alia that inmates exhaust “such administrative remedies as are available” before filing suit challenging prison conditions under § 1983. PLRA § 803(d) (amending 42 U.S.C. § 1997e(a)). The new law was signed by the President on April 26, 1996, and went into effect that same day. See Lyons v. Ohio Adult Parole Auth., 105 F.3d 1063, 1066 (6th Cir.1997) (statutes become effective when enacted absent indication to contrary). The clerk of this court chose one case from each state within this circuit and asked attorneys who were involved with prisoner civil rights litigation and attorneys for the four states to submit briefs on the issue of whether the new administrative exhaustion provision applies to these pending cases.4
II. JURISDICTION AND STANDARD OF REVIEW
We have jurisdiction over these § 1983 cases under 28 U.S.C. §§ 1331, 1291. We examine de novo the purely legal question of whether a new statute applies to pending cases.5 Lyons, 105 F.3d at 1065-66.
III. DISCUSSION
Before this year, prisoners challenging the conditions of their confinement under 42 U.S.C. § 1983 were not, as a rule, required to exhaust administrative remedies before filing suit. Although 42 U.S.C. § 1997e allowed district courts to “continue such case[s] for a period of not to exceed 180 days in order to require exhaustion of such plain, speedy, and effective administrative remedies as are available,” exhaustion was only to be required “if the court believe[d] that such a requirement would be appropriate and in the interests of justice.” 42 U.S.C. § 1997e(a)(1) (1995). See Patsy v. Board of Regents, 457 U.S. 496, 502-07, 102 S.Ct. 2557, 2560-63, 73 L.Ed.2d 172 (1982) (discussing § 1983); id. at 508-12, 102 S.Ct. at 2563-66 (discussing § 1997e). The PLRA, however, amended § 1997e to require that prisoners seeking to bring such claims first exhaust any available administrative remedies. 42 U.S.C. § 1997e(a). The question before us is whether this new requirement should be applied to eases which were filed, dismissed by the district courts, and appealed to this court before the PLRA was signed into law.
[418]*418The Supreme Court has determined that, in deciding whether a new statute should be applied to pending cases,
the court’s first task is to determine whether Congress has expressly prescribed the statute’s proper reach. If Congress has done so, of course, there is no need to resort to judicial default rules. When, however, the statute contains no such express command, the court must determine whether the new statute would have retroactive effect, i.e., whether it would impair rights a party possessed when he acted, increase a party’s liability for past conduct, or impose new duties with respect to transactions already completed. If the statute would operate retroactively, ... it does not govern absent clear congressional intent favoring such a result.
Landgraf v. USI Film Prods., 511 U.S. 244, 280, 114 S.Ct. 1483, 1505, 128 L.Ed.2d 229 (1994). Accordingly, we “first look to the statute’s text for an expression that the [Act] should, or should not, apply to pending cases.” Lyons, 105 F.3d at 1065.
A. Textual Analysis
The PLRA amended 42 U.S.C. § 1997e to read, “[n]o action shall be brought with respect to prison conditions under section 1983 of this title, or any other Federal law, by a prisoner confined in any jail, prison, or other correctional facility until such administrative remedies as are available are exhausted.” 42 U.S.C. § 1997e(a) (emphasis added). We believe that Congress, by its use of the highlighted language, “has expressly prescribed the statute’s proper reach.” Landgraf, 511 U.S. at 280, 114 S.Ct. at 1505. The statute expressly governs the bringing of new actions, not the disposition of pending cases. Actions brought before the statute was enacted are -not affected by the new administrative exhaustion requirement.
The Seventh Circuit has used similar reasoning to find that another provision of the PLRA does not apply to pending cases. In Abdul-Wadood v. Nathan, 91 F.3d 1023 (7th Cir.1996), that court had to decide whether a case filed before the PLRA’s enactment could be dismissed under the new 28 U.S.C. § 1915(g), which states that “[i]n no event shall a prisoner bring a civil action or appeal a judgment in a civil action or proceeding under this section if’ he has brought three previous frivolous actions or appeals. 28 U.S.C. § 1915(g) (1996). Judge Easterbrook held for the court that the case should not be dismissed because “[s]ection 1915(g) governs bringing new actions or filing new appeals ... rather than the disposition of existing cases.” 91 F.3d at 1025. Similarly, because § 1997e(a) governs only the bringing of actions, it does not affect pending cases.
Section 802 of the PLRA, which delineates the courts’ authority to order prospective relief with respect to prison conditions, lends some additional support to this textual analysis. This section specifically states that the amendment in that section to 18 U.S.C. § 3626 “shall apply with respect to all prospective relief whether such relief was originally granted or approved before, on, or after the date of the enactment of this title.” PLRA § 802(b)(1). Plaintiffs argue expres-sio unius est exclusio alterius: that because Congress specified that one part of the PLRA applies to pending actions, it must have intended that the rest of the Act does not. Such an argument would usually not be dispositive by itself. See Landgraf, 511 U.S. at 259, 114 S.Ct. at 1493-94; Lyons, 105 F.3d at 1067. Here, however, the text of the new requirement plainly states that “[n]o action shall be brought” without exhaustion of administrative remedies. Thus, it is likely that had Congress intended the new requirement to pertain to pending eases it would have employed the same language as it used in § 802(b)(1) to make that intent clear. This strengthens our conclusion that the text of the PLRA indicates that the new administrative exhaustion requirement applies only to eases filed after the Act’s passage.
B. Impermissible Retroactive Effect
Even if the language of the statute did not mandate that administrative exhaustion be required only in actions brought after the effective date of the Act, the Supreme Court’s decision in Landgraf would do so. Footnote 29 of the Court’s opinion is controlling:
[419]*419A new rule concerning the filing of complaints would not govern an action in which the complaint had already been properly filed under the old regime, and the promulgation of a new rule of evidence would not require an appellate remand for a new trial.
Landgraf, 511 U.S. at 275 n. 29, 114 S.Ct. at 1502 n. 29. Although the states suggest that this language could refer only to changes to the technical requirements of filing a complaint, and that the footnote is thus distinguishable from the eases at bar, we will not give the opinion such a narrow reading. The new exhaustion requirement concerns the bringing of actions and, under Landgraf does not affect cases brought before the Act’s passage.6
The states argue that because footnote 29 is dictum it is not binding on this court. Cf. Covino v. Reopel, 89 F.3d 105, 108 (2d Cir.1996) (dismissing footnote 29 as distinguishable dictum). We believe this footnote is instructive of the Supreme Court’s views and cannot be dismissed out of hand. If anything, this particular “dictum,” which eight Justices endorsed, carries more weight than much of the Landgraf analysis, which commanded a bare majority. See Landgraf, 511 U.S. at 290-92, 114 S.Ct. at 1524-25 (Scalia, J., concurring in judgment). Where there is no clear precedent to the contrary, we will not simply ignore the Court’s dicta. See Jordon v. Gilligan, 500 F.2d 701, 707 (6th Cir.1974) (“Even the [Supreme] Court’s dicta is of persuasive precedential value.”).
The states argue that there is, in fact, contrary Supreme Court precedent: Cort v. Ash, 422 U.S. 66, 95 S.Ct. 2080, 45 L.Ed.2d 26 (1975), in which the plaintiff sought to enjoin a corporation from including partisan political literature with its dividend checks, and Weinberger v. Salfi, 422 U.S. 749, 95 S.Ct. 2457, 45 L.Ed.2d 522 (1975), which involved a challenge to Social Security eligibility requirements. Neither case is apposite. Cort involved two of the situations, inapplicable here, in which, under Landgraf, a court should “apply the law in effect at the time it renders its decision.” Landgraf, 511 U.S. at 273, 114 S.Ct. at 1501 (quoting Bradley v. School Bd. of Richmond, 416 U.S. 696, 711, 94 S.Ct. 2006, 2016, 40 L.Ed.2d 476 (1974)). First, Cort differs from the eases at bar in that Ash was seeking to enjoin the defendants from future misconduct. Cort, 422 U.S. at 71, 74, 95 S.Ct. at 2084-85, 2086. “In that circumstance, a statute enacted after the decision of the Court of Appeals requires reversal of the holding of the Court of Appeals.” Id. at 74, 95 S.Ct. at 2086. (citation omitted) (emphasis added). See id. at 77, 95 S.Ct. at 2087 (“[N]or is there any possible ‘manifest injustice’ in requiring respondent to pursue with respect to alleged violations which have yet to occur the statutory remedy for injunctive relief created by the [intervening] Amendments.”) (emphasis added). Under Landgraf, applying a new statute to pending cases is proper where plaintiff seeks only future injunctive relief. Landgraf, 511 U.S. at 273-74, 114 S.Ct. at 1501-02.7
Second, the intervening statute8 in Cort “expressly vest[ed] the [Election] [420]*420Commission with ‘primary jurisdiction’ over any claimed violation of [the relevant election law] within its purview.” Cort, 422 U.S. at 75-76, 95 S.Ct. at 2086. Under Landgraf, “intervening statutes conferring or ousting jurisdiction” will regularly be applied to pending cases because “jurisdictional statutes ‘speak to the power of the court rather than to the rights or obligations of the parties,’ ” Landgraf, 511 U.S. at 274, 114 S.Ct. at 1502 (quoting Republic Nat. Bank of Miami v. United States, 506 U.S. 80, 100, 113 S.Ct. 554, 565-66, 121 L.Ed.2d 474 (1992) (Thomas, J., concurring)). See Lyons, 105 F.3d at 1075. The statute in Cort contained a specific grant of jurisdiction to the Commission which, the Court held, served at the same time to divest the federal courts of jurisdiction over such cases. Cort, 422 U.S. at 75-76 & n. 9, 95 S.Ct. at 2086-87 & n. 9. Section 1997e(a), in contrast, addresses a party’s right to bring suit in court and does not speak in terms of jurisdiction or the power of the court: it is merely an exhaustion requirement. See Castille v. Peoples, 489 U.S. 346, 349, 109 S.Ct. 1056, 1059, 103 L.Ed.2d 380 (1989) (“Codified since 1948 in 28 U.S.C. § 2254,9 the [habeas corpus] exhaustion rule, while not a jurisdictional requirement, creates a ‘strong presumption in favor of requiring the prisoner to pursue his available state remedies.’ ”) (citing 28 U.S.C. § 2254 and quoting Granberry v. Greer, 481 U.S. 129, 131, 107 S.Ct. 1671, 1673-74, 95 L.Ed.2d 119 (1987)) (emphasis added); McCarthy v. Madigan, 503 U.S. 140, 144, 112 S.Ct. 1081, 1085-86, 117 L.Ed.2d 291 (1992) (“The doctrine of exhaustion of administrative remedies is one among related doctrines — including abstention, finality, and ripeness — that govern the timing of federal court decisionmaking.”) (emphasis added).10
Cort, then, does not suggest that the exhaustion requirement should apply to these pending cases. Landgraf specifically cites Cort as involving “application of [an] intervening statute transferring to [an] administrative agency jurisdiction over [a] claim for injunctive relief." Landgraf, 511 U.S. at 279 n. 33, 114 S.Ct. at 1504 n. 33. In the cases before us, where plaintiffs seek monetary damages and the intervening statute does not purport to oust district court jurisdiction, Cort is inapposite.
The states next argue that the new administrative exhaustion requirement is jurisdictional under Weinberger v. Salfi, 422 U.S. 749, 95 S.Ct. 2457, 45 L.Ed.2d 522 (1975), and should thus govern pending cases.11 The essence of their argument is that Salfi held that a statute that involved agency exhaustion and contained seemingly prospective language was jurisdictional, and therefore that § 1997e(a) must also be jurisdictional. The Salfi opinion makes plain, however, the difference between the statute involved there and the one now before us. In Salfi, the Court held that it was clear that the relevant statute12 was “more than a codified requirement of administrative exhaustion” on the grounds that the statute’s “own language” was “sweeping and direct and [ ] statefd] that [421]*421no action shall be brought under [28 U.S.C.] § 1331, not merely that only those actions shall be brought in which administrative remedies have been exhausted.” Salfi, 422 U.S. at 757, 95 S.Ct. at 2463. See id. at 756-57, 95 S.Ct. at 2462-63 (“On its face, this provision bars district court federal-question jurisdiction over suits, such as this one, which seek to recover Social Security benefits. Yet it was § 1331 jurisdiction which appellees successfully invoked in the District Court.”). Section 1997e(a), in contrast, contains neither the sweeping and direct language of § 405(h) nor that statute’s explicit bar to district court jurisdiction. It, rather, indicates “merely that only those actions shall be brought in which administrative remedies have been exhausted.” Salfi, 422 U.S. at 757, 95 S.Ct. at 2463.
Because neither Cort nor Salfi requires a contrary result, we believe that even if the prospective statutory language were not dis-positive in this case, footnote 29 of Landgraf would require that the new administrative exhaustion provision not apply to pending cases.13
C. Policy Considerations
The states and the plaintiffs argue at great length that applying, or refusing to apply, respectively, this new provision to pending cases would be good policy or would better serve the purposes of the new law. Such considerations do not figure in the Landgmf analysis. See Landgraf, 511 U.S. at 285-86, 114 S.Ct. at 1507-08 (“It will frequently be true ... that retroactive application of a new statute would vindicate its purpose more fully. That consideration, however, is not sufficient to rebut the presumption against retro-activity---- A legislator who supported, a prospective statute might reasonably oppose retroactive application of the same statute.”). Moreover, it is not at all clear in this case which side has the better of the policy argument.
The Supreme Court has called “protecting administrative agency authority and promoting judicial efficiency” the “twin purposes” of exhaustion requirements. McCarthy v. Madigan, 503 U.S. 140, 145, 112 S.Ct. 1081, 1086-87, 117 L.Ed.2d 291 (1992). Sending these plaintiffs back to exhaust their administrative remedies would, of course, lighten this court’s docket, at least temporarily.14 See id. (judicial efficiency includes avoiding unnecessary judicial controversy and piecemeal appeals). However, “policy considerations alone cannot justify judicially imposed exhaustion unless exhaustion is consistent with congressional intent.” Patsy v. Board of Regents, 457 U.S. 496, 513, 102 S.Ct. 2557, 2566, 73 L.Ed.2d 172 (1982). Because Congress did not intend that all prisoners who filed suit before the Act took effect be required to exhaust administrative remedies, we could not impose such a requirement even if we thought it wise policy.
In addition, although administrative exhaustion in the usual case “may produce a useful record for subsequent judicial consideration,” McCarthy, 503 U.S. at 145, 112 S.Ct. at 1087, the facts in these cases are already stale. Administrative findings might still be useful to a reviewing court, but administrative fact finding in these old cases would not have the same promptness advantage over judicial fact finding that would exist where the administrative procedure occurred soon after the alleged violation.
Other policy considerations counsel against applying the new exhaustion requirement to [422]*422pending cases. The sheer volume of prisoner civil rights cases makes it difficult for the federal courts to take the time that is often necessary to distinguish between the frivolous and the meritorious suits. That inmates are forced to file nearly all of these cases pro se, and can rarely conduct any sort of factual investigation, only compounds this problem. To the extent, then, that applying the new administrative exhaustion requirement to pending cases would allow the courts more time to examine the merits of each case, or would provide a way to separate the meritorious from the meritless eases, to do so would be good policy. Conversely, we would hesitate to do this if we thought it would aggravate these problems.
The majority of the prisoner’s rights cases pending before this court were dismissed below; if past experience is any indication, to dismiss the cases and allow the prisoners to refile after they háve exhausted their administrative remedies, or to retain jurisdiction over the cases while the inmates exhaust those remedies, will not reduce the district courts’ dockets or workloads. To the contrary, the district courts would then have to reexamine these same cases, supplemented with new records from the administrative proceedings, sometime in the future. Although a district court that had erroneously previously dismissed a particular case might conceivably make a different determination based on the new administrative record, if an inmate had thought that going through the administrative process would help his case, he could have filed an administrative complaint in the first place under the old law. And if the district court had thought that administrative consideration would have been “appropriate and in the interests of justice,” and that the available grievance procedure was “fair and effective,” the district judge could have continued the case for up to 180 days and ordered administrative exhaustion in the first place. 42 U.S.C. § 1997e(a) (1995).15 Furthermore, there is no indication that any plaintiffs with pending cases want to be required, or even allowed, to go back and exhaust state administrative procedures. In this regard, then, we believe it is better policy to apply the old law to these pending cases and address them now, rather than recycling them through the system and perhaps perpetuating the problems that the old law engendered. We emphasize that we do not believe that it is our job to weigh policy — that is Congress’s job.- We address these policy arguments simply because the parties spent so much time doing so.
In the end, even if it were proper for us to use policy arguments to supersede congressional intent and apply the new law to these pending cases, it is unclear which outcome is the better policy.
Sending these cases back for administrative exhaustion would also cause unnecessary jurisprudential questions in the future. The events giving rise to these cases occurred several years ago; in some, the statute of limitations has expired. Although all four of the states’ representatives stated at oral argument that they would waive any applicable time limits on filing administrative claims, none of the assistant attorneys general present could assure us that they had the authority to extend the applicable statutes of limitations. Were we to apply the new law to these cases, it is not completely certain that we could toll the statutes of limitations. Compare Board of Regents v. Tomanio, 446 U.S. 478, 490-91, 100 S.Ct. 1790, 1798, 64 L.Ed.2d 440 (1980) (“Unless [a federal remedy] is structured to require previous resort to state proceedings, so that the claim may not even be maintained in federal court unless such resort be had, it cannot be assumed that Congress wishes to hold open the independent federal remedy during any period of time necessary to pursue alternative state-court remedies.”) (citation omitted), with Patsy, 457 U.S. at 514 n. 17, 102 S.Ct. at 2566-67 n. 17 (“Unless the doctrine that statutes of limitations are not tolled pending exhaustion were overruled, see [Tomanio ], a [423]*423judicially imposed exhaustion requirement might result in the effective repeal of § 1983”). Finally, the states have not pointed to anything to distinguish the cases before us, where the inmates lost at the district court level, from cases pending in this court in which the plaintiffs were victorious below. Requiring successful plaintiffs to relitigate their grievances would have impermissible retroactive effect under Landgraf. See Landgraf, 511 U.S. at 280, 114 S.Ct. at 1505 (statute will have retroactive effect if it “impose[s] new duties with respect to transactions already completed”). It may well be that the courts could develop doctrines and remedies to solve these potential problems. However, we should not invite unnecessary problems upon the courts and simply hope that we might be able to remedy them in the future.16
These considerations lend further support to our conclusion: where the language of the statute and Landgraf both require that we apply this new requirement only with respect to eases filed after the PLRA’s passage, it would be improper to hold, on policy grounds, that the provision applies to the cases before us. Cf. Patsy, 457 U.S. at 514, 102 S.Ct. at 2567 (“These and similar questions might be answered swiftly and surely by legislation, but would create costly, remedy-delaying, and court-burdening litigation if answered incrementally by the judiciary in the context of diverse constitutional claims relating to thousands of different state agencies.”).
IV. CONCLUSION
Because the language of 42 U.S.C. § 1997e(a) is explicitly prospective and there is no reason to think that Congress intended a retroactive effect, we will not apply the new administrative exhaustion requirement to these cases where appeals were pending in this court on April 26, 1996, the day the PLRA was enacted. These four cases are properly before this court, and can be decided without undertaking administrative exhaustion. We address the merits of these four eases in separate opinions.