EASTERBROOK, Circuit Judge.

A small claims court in Marion County, Indiana, entered a judgment against Kevin Harold for a little more than $1,000. He did not pay, even though he had agreed to the judgment's entry. Almost two decades later Christopher Steel, claiming to represent the judgment creditor, asked the court to garnish Harold's wages. It entered the requested order, which Harold moved to vacate, contending that Steel had misrepresented the judgment creditor's identity (transactions after the judgment's entry may or may not have transferred that asset to a new owner) and did not represent the only entity authorized to enforce the judgment. But he did not contend that the request was untimely. After a hearing, a state judge sided with Steel and maintained the garnishment order in force. Instead of seeking review within Indiana's judiciary, Harold filed this federal suit under the Fair Debt Collection Practices Act, contending that Steel and his law firm (Peters & Steel, LLC, which we do not mention again) had violated 15 U.S.C. § 1692e by making false statements. But the district court dismissed the suit for want of subject-matter jurisdiction, ruling that it is barred by the Rooker-Feldman doctrine because it contests the state court's decision. 2014 U.S. Dist. LEXIS 43154 (S.D.Ind. Mar. 31, 2014).

More than a decade ago, this court held in Epps v. Creditnet, Inc., 320 F.3d 756 (7th Cir.2003), that the Rooker-Feldman doctrine bars federal suits seeking to recover on a theory that a debt collector made false statements during state litigation. The facts of Epps are similar to those of this case, right down to the location of the state suit: a small-claims court in Marion County, Indiana. The only differences are that the supposed misrepresentations in Epps concerned the amount of damages rather than the creditor's identity, and that Epps dropped his § 1692e claim on appeal and relied on a state-law theory under the supplemental jurisdiction. The abandonment of the § 1692e claim in Epps raises the possibility that this strategic choice affected federal jurisdiction. Harold maintains that it does, relying on Exxon Mobil Corp. v. Saudi Basic Industries Corp., 544 U.S. 280, 125 S.Ct. 1517, 161 L.Ed.2d 454 (2005). But three years after Exxon Mobil this court applied the approach of Epps to a claim under the Fair Debt Collection Practices Act. See Kelley v. Med-1 Solutions, LLC, 548 F.3d 600 (7th Cir.2008). Harold wants us to overrule Kelley and Epps.

Rooker v. Fidelity Trust Co., 263 U.S. 413, 44 S.Ct. 149, 68 L.Ed. 362 (1923), and District of Columbia Court of Appeals v. Feldman, 460 U.S. 462, 103 S.Ct. 1303, 75 L.Ed.2d 206 (1983), hold that the Supreme Court of the United States is the only federal court that may review judgments entered by state courts in civil litigation. The Rooker-Feldman doctrine applies when the state court's judgment is the source of the injury of which plaintiffs complain in federal court. See Exxon Mobil, 544 U.S. at 293, 125 S.Ct. 1517; GASH Associates v. Rosemont, 995 F.2d 726, 729 (7th Cir.1993). Harold insists, however, that the doctrine does not apply to interlocutory decisions by state tribunals; he maintains that these may be reviewed by federal district courts.

Why not? Nothing in the Supreme Court's decisions suggests that state-court decisions too provisional to deserve review within the state's own system can be reviewed by federal district and appellate courts. The principle that only the Supreme Court can review the decisions by the state judiciary in civil litigation is as applicable to interlocutory as to final state-court decisions. A truly interlocutory decision should not be subject to review in any court; review is deferred until the decision is final.

We recognize that the courts of appeals disagree about the issue. Compare Pieper v. American Arbitration Association, Inc., 336 F.3d 458, 461-62 (6th Cir.2003) (applying Rooker-Feldman to a federal suit challenging an interlocutory order); Kenmen Engineering v. Union, 314 F.3d 468, 474 (10th Cir.2002) (same); Brown & Root, Inc. v. Breckenridge, 211 F.3d 194, 199 (4th Cir.2000) (same); and Port Authority Police Benevolent Association, Inc. v. Port Authority of New York, 973 F.2d 169, 178-79 (3d Cir.1992) (same); with Cruz v. Melecio, 204 F.3d 14, 21 n. 5 (1st Cir.2000) (Rooker-Feldman doctrine is limited to final judgments); Green v. Mattingly, 585 F.3d 97, 102 (2d Cir.2009) (same, but in dictum). Our decision in Mehta v. Attorney Registration and Disciplinary Commission, 681 F.3d 885, 887 (7th Cir.2012), does not choose sides; instead we observed that the state decision in question was final. Nor need we resolve the question in this case, again because the decision is final. United States v. Kollintzas, 501 F.3d 796, 801-02 (7th Cir.2007), and United States v. Sloan, 505 F.3d 685, 687 (7th Cir.2007), are among many opinions holding that garnishment orders enforcing a judgment are final and appealable. Indiana follows the same approach. Tipton v. Flack, 149 Ind.App. 129, 134, 271 N.E.2d 185 (1971).

Harold maintains that his claim is independent of the state court's decision and thus outside the scope of the Rooker-Feldman doctrine under Exxon Mobil, which holds that the doctrine applies only when the state court has caused the injury of which the federal suit complains. If the state court just failed to remedy an injury that predated the litigation (or is independent of it), the Court held, the federal district judge should apply principles of issue and claim preclusion under 28 U.S.C. § 1738 rather than dismiss for want of jurisdiction — and under Indiana law decisions of small claims courts do not have issue-preclusive effect. See Geico Insurance Co. v. Graham, 14 N.E.3d 854, 860 (Ind.App.2014). Harold insists that the false statements, rather than the state court's decision, inflicted the injury of which he complains.

It is easy to imagine situations in which a violation of federal law during the conduct of state litigation could cause a loss independent of the suit's outcome. Suesz v. Med-1 Solutions, LLC, 757 F.3d 636 (7th Cir.2014) (en banc), illustrates. The Fair Debt Collection Practices Act limits debt collectors to suits in the "judicial district or similar legal entity" where the contract was signed or the debtor resides. 15 U.S.C. § 1692i. If a debt collector violates that statute, it inflicts an injury measured by the costs of travelling or sending a lawyer to the remote court and moving for a change of venue, no matter how the suit comes out.

Harold was not injured in that way, however. He complains about representations that concern the merits. If Steel's client did not own the judgment, then Harold was entitled to a decision in his favor. No injury occurred until the state judge ruled against Harold. The need to litigate was not a loss independent of the state court's decision; costs of litigation were inevitable whether or not Steel was telling the truth about his client's rights — and it should be cheaper to defeat a false claim than to defeat a true one.

As Harold sees things, the Rooker-Feldman doctrine does not apply to the procedures that state courts use to reach decisions or the evidence that state judges consider. This line of argument is embarrassed by the fact that Rooker itself arose from a contention that the state court (at the adverse litigant's instigation) had used constitutionally forbidden procedures to reach its judgment. Unless Rooker were to be overruled, there could not be a "procedural exception" to the Rooker-Feldman doctrine.

Federal review of the procedures or evidence used in state court would collapse the distinction between civil and criminal cases. Collateral review of state criminal judgments under 28 U.S.C. §§ 2241 and 2254 is a search for improper procedures; most substantive decisions are governed by state law and cannot be reviewed in federal court. See, e.g., Bradshaw v. Richey, 546 U.S. 74, 126 S.Ct. 602, 163 L.Ed.2d 407 (2005); Gilmore v. Taylor, 508 U.S. 333, 113 S.Ct. 2112, 124 L.Ed.2d 306 (1993); Estelle v. McGuire, 502 U.S. 62, 112 S.Ct. 475, 116 L.Ed.2d 385 (1991). If Harold were to prevail in this suit, however, federal courts could award damages every time a litigant in state court used an improper procedure or considered evidence that a federal judge does not think trustworthy. That duplication would greatly increase the already high cost of civil litigation.

The Rooker-Feldman doctrine is a matter of statutory interpretation, not of constitutional command. Congress is free to authorize federal collateral review of state civil judgments — though there may be limits to how far national law can specify procedures that state courts must use, as Judge Sykes's concurring opinion in Suesz explains, 757 F.3d at 650-55 — but 15 U.S.C. § 1692e does not approach the limits of federal power. Section 1692e forbids debt collectors to tell lies but does not suggest that federal courts are to review state-court decisions about whether lies have been told. Section 1692e does not even hint that federal courts have been authorized to monitor how debt-collection litigation is handled in state courts. Section 1692i (the subject of Suesz) authorizes federal courts to address one specific aspect of state debt-collection litigation; § 1692e lacks a parallel reference to the conduct of litigation in state courts, so the norm from the Rooker-Feldman doctrine controls.

Harold might have used § 1692e to file a counterclaim in Indiana and could have appealed within the state system. He did neither. His federal suit was properly dismissed.

AFFIRMED.