Cathleen M. Combs, Daniel A. Edelman, Edelman Combs Latturner & Goodwin, Chicago, IL, for Plaintiff–Appellant. Peter A. Velde, Mark D. Gerth, Nicholas W. Levi, Kightlinger & Gray LLP, Indianapolis, IN, for Defendant–Appellee.
Before WOOD, Chief Judge, and POSNER, FLAUM, EASTERBROOK, KANNE, ROVNER, WILLIAMS, SYKES, TINDER, and HAMILTON, Circuit Judges.

HAMILTON and POSNER, Circuit Judges.

The federal Fair Debt Collection Practices Act (“FDCPA”) requires a collector of consumer debts to file its debt-collection suit in the “judicial district or similar legal entity” where the contract was signed or where the debtor resides. 15 U.S.C. § 1692i. This appeal requires us to apply this statutory language to the nine small claims courts in Marion County, Indiana, which together hear some 70,000 civil cases each year. This interpretive issue has significant consequences not only for consumer debtors and debt collectors in Marion County but also for parties to debt-collection suits in other court systems that, depending on the answer to the interpretive question, may be vulnerable to abusive forum-shopping by debt collectors.

Defendant Med–1 Solutions, LLC filed suit in the Pike Township of Marion County Small Claims Court to collect a consumer debt from plaintiff Mark Suesz. The plaintiff alleges that the defendant violated § 1692i by filing in that court because the contract was not signed in Pike Township and the plaintiff does not live there.

In Newsom v. Friedman, 76 F.3d 813 (7th Cir.1996), a panel of this court held that the intra-county districts used to delineate the venue of small claims cases in Illinois's Cook County Circuit Court were not separate judicial districts for purposes of § 1692i. In this case, the district court and a split panel of this court followed the reasoning of Newsom to hold that the township small claims courts in Marion County are likewise not separate judicial districts; rather, the entire county is the relevant district, giving the debt collector a wide choice of venue. Suesz v. Med–1 Solutions, LLC, 734 F.3d 684 (7th Cir.2013). We granted the plaintiff's petition for rehearing en banc.

We conclude that the correct interpretation of “judicial district or similar legal entity” in § 1692i is the smallest geographic area that is relevant for determining venue in the court system in which the case is filed. See Hess v. Cohen & Slamowitz LLP, 637 F.3d 117, 123–24 (2d Cir.2011). For the small claims courts in Marion County, that smallest area is a township. We therefore reverse the judgment of the district court. We also overrule Newsom, which adopted a test based on details of court administration rather than on the applicable venue rules.

I. The Factual Allegations

Because the district court dismissed this action under Federal Rule of Civil Procedure 12(b)(6) for failure to state a claim, we review its decision de novo and treat as true the factual allegations of the complaint, giving the plaintiff the benefit of favorable inferences from those allegations. E.g., Marshall–Mosby v. Corporate Receivables, Inc., 205 F.3d 323, 326 (7th Cir.2000) (reversing dismissal of FDCPA complaint).

Mark Suesz, who resides in Hancock County, which is immediately east of Marion County, owed money to Community North Hospital. The hospital, which is located in Lawrence Township in the northeast corner of Marion County, turned the debt over to Med–1 Solutions for collection. Med–1 sued Suesz in the Pike Township Small Claims Court. Pike Township is in the northwest corner of Marion County. The court issued a judgment against Suesz for $1,280. The validity of that judgment is not questioned in this federal lawsuit.

Suesz then filed this action under the Fair Debt Collection Practices Act, asserting that Med–1 has a practice of filing collection lawsuits in Marion County in small claims courts located in townships where the debtor defendants neither live nor signed the contracts on which they are being sued. Suesz moved to certify a plaintiff class, but the district court dismissed the case without acting on the motion. On the basis of our decision in Newsom, the district court concluded that the townships of Marion County are not “judicial districts” for purposes of § 1692i and so dismissed Suesz's suit. Suesz v. Med–1 Solutions, LLC, No. 1:12–cv–1517–WTL–MJD, 2013 WL 1183292 (S.D.Ind. March 21, 2013).

II. The Fair Debt Collection Practices Act

The Fair Debt Collection Practices Act seeks “to eliminate abusive debt collection practices by debt collectors.” 15 U.S.C. § 1692(e); see Muha v. Encore Receivable Mgmt., Inc., 558 F.3d 623, 629 (7th Cir.2009); Jacobson v. Healthcare Financial Services, Inc., 516 F.3d 85, 89 (2d Cir.2008). Consumer debts covered by the Act are usually too small to justify a lawsuit unless the suit is promptly defaulted, thereby enabling the debt collector to obtain—without incurring significant litigationcost—a judgment that it can use to garnish the debtor's wages. Given “the costs of litigation and the difficulties establishing the debt, when a debt collector cannot get payment through phone calls and letters and it has to go to court, the debt collector will often rely on default judgments as the last resort.” O'Rourke v. Palisades Acquisition XVI, LLC, 635 F.3d 938, 940 (7th Cir.2011).

As this case illustrates, one common tactic for debt collectors is to sue in a court that is not convenient to the debtor, as this makes default more likely; or in a court perceived to be friendly to such claims; or, ideally, in a court having both of these characteristics. In short, debt collectors shop for the most advantageous forum. By imposing an inconvenient forum on a debtor who may be impecunious, unfamiliar with law and legal processes, and in no position to retain a lawyer (and even if he can afford one, the lawyer's fee is bound to exceed the debt itself), the debt collector may be able to obtain through default a remedy for a debt that the defendant doesn't actually owe.

The FDCPA is designed to protect consumer debtors against unscrupulous methods of consumer debt collection. Thus in Phillips v. Asset Acceptance, LLC, 736 F.3d 1076 (7th Cir.2013), we held that the Act was violated by the disreputable tactic of suing a debtor after the statute of limitations has expired; the debt collector hopes that the debtor will be unaware that he has a complete defense to the suit and so will default, which will enable the debt collector to garnish the debtor's wages. Abusive forum-shopping is another improper method of collecting consumer debts. Accordingly, the Act provides that unless the debt sued on is secured by real estate, a debt collector can sue to collect it “only in the judicial district or similar legal entity—(A) in which such consumer signed the contract sued upon; or (B) in which such consumer resides at the commencement of the action.” 15 U.S.C. § 1692i(a)(2). (If real estate is security for the loan, the suit must be brought where the property is located, § 1692i(a)(1); that will usually be an advantageous venue from the debtor's standpoint.) A violation makes the debt collector liable to the debtor for statutory and actual damages, as well as attorney fees. § 1692k.

Unfortunately the key statutory term—“judicial district”—is vague. The FDCPA does not define it, and as we explain below the phrase has no general definition or meaning that can resolve this dispute. In Indiana, Illinois, and most other states, state trial courts usually are organized by county for purposes of both court administration and venue. When that is so, it may seem natural to interpret the statutory term as referring to the county in which the debtor lives or the contract giving rise to the debt was signed. But terms that seem plain and easy to apply to some situations can become ambiguous in other situations. This statutory term was drafted broadly—indicated by the phrase “or similar legal entity”—presumably so that it could be applied with appropriate flexibility to court systems around the country, which vary in structure and nomenclature. We believe that the term describes the township small claims courts in Marion County, which we examine next with a particular eye to the relevance of venue rules for interpreting and applying the statutory term “judicial district or similar legal entity.”

The alternative approach, favored by the panel majority, would be for the court in an FDCPA case to defer to each state's definition of “judicial district.” One problem with that approach is that “judicial district” is not a defined term in state law. A deeper problem is that, had Congress been content to adopt the states' rules governing jurisdiction and venue, there would have been no reason to impose venue limitations on debt collectors, as the federal Act does; the debt collector could have sued wherever state law permitted him to sue. The presence of the venue provision in the Act shows congressional dissatisfaction with allowing state law to determine where suits to collect consumer debts can be filed.

This is not to suggest that the federal law alters state rules governing jurisdiction and venue. Federal law merely imposes a limit on which state courts having jurisdiction and venue over a debt collector's claim the debt collector may sue in, consistent with the policy of the federal law. There is nothing unusual or untoward about requiring compliance with both state and federal rules. Next we explain why the township small claims courts in Marion County must be regarded as occupying separate judicial districts in order to enforce the policy of the federal law.

III. The Township Small Claims Courts in Marion County

Most trial courts in Indiana are county-wide circuit or superior courts of general jurisdiction. The small claims courts serving the nine townships in Marion County (which is coterminous with Indianapolis) are important, high-volume exceptions. Pursuant to the system established by the Northwest Ordinance of 1787, the nearly square county is divided into a grid of nine nearly square townships.

State law makes the small claims courts of Marion Country nine separate courts, each designated the “_____ Township of Marion County Small Claims Court,” Ind.Code § 33–34–1–2, the blank being the name of the township in question. Each court has its own judge elected by the voters of the township. § 33–34–2–1. Each court has jurisdiction over civil cases founded on contract or tort in which the debt or damage claimed does not exceed $6,000, exclusive of interest and attorney fees. § 33–34–3–2. The township small claims courts are housed, funded, and staffed by the respective township governments rather than the state or county governments. § 33–34–6–1 et seq. “In essence, the Marion County small claims courts are township-level judicial entities.” In re Mandate of Funds for Center Township of Marion County Small Claims Court, 989 N.E.2d 1237, 1239 (Ind.2013) (resolving disputes between one township court and the township's governing bodies over court's funding, location, and administration).

These smaller judicial districts were established for the convenience of litigants and the avoidance of docket congestion, but the panel's decision gave debt collectors suing in Marion County the choice of which of the nine courts to sue in, just as if the nine were one. The debt collector thus could choose the township court that was most inconvenient for the defendant, friendliest to creditors, or both. To adopt this interpretation would undermine the venue provision of the Fair Debt Collection Practices Act. It would amount to saying that Congress had created the provision with one hand and simultaneously nullified it with the other.

Indiana's current statute determining venue, Indiana Code § 33–34–3–1, makes venue depend on townships in a small claims suit in Marion County.

When defendant Med–1 Solutions sued plaintiff Suesz in the Pike Township small-claims court, the applicable court rule that had been promulgated by the Indiana Supreme Court provided for broader venue than the current state statute. Small Claims Rule 12 then read in relevant part:

(A) Proper venue. Proper venue for a case filed in the small claims docket of a Circuit, Superior, or County Court shall be in the county where the transaction or occurrence actually took place or where the obligation was incurred or is to be performed, or where one of the defendants resides or has his or her place of employment at the time the complaint is filed.

Proper venue of any claim between landlord and tenant ... filed in county small claims courts created pursuant to IC 33–34–1–2 [i.e., township courts in Marion County] shall be in the county and township division of the Small Claims Court where the real estate is located.

The differences between the venue court rule and the venue statute reflect significant changes in the small claims courts that were made in 1999 to settle a Voting Rights Act lawsuit, but to which the court rules were not conformed until much later.

Before 1999, the small claims courts in the nine townships had been organized as divisions of the Marion Superior Court, and both the venue statute and the venue rule specified venue by township rather than by county only for landlord-tenant claims. For all other cases venue was countywide. The problem under the Voting Rights Act was that the nine judges of those small claims divisions of the Marion Superior Court were each elected by only one township, but all exercised authority over disputes arising throughout the county. The nine townships differed dramatically in terms of overall population and racial composition. The combination of election-by-township and countywide authority substantially diluted the voting power of African–Americans.

After a federal district court denied the defendants' motion for summary judgment in Anderson v. Mallamad, No. IP94–1447–C H/G, 1997 WL 35024766 (S.D.Ind. March 28, 1997), the parties to the voting rights suit settled by agreeing to the enactment of state legislation that would retain the election of judges by township but make each township court much more independent and—critically for the present case—provide for venue by township. See Ind. Pub.L. 95–1999. Tying together the election district for the judge and the venue of the court provided an arguable justification for what would otherwise have been unacceptable disparities in voting populations in judicial elections. See Houston Lawyers' Ass'n v. Attorney General of Texas, 501 U.S. 419, 426–27, 111 S.Ct. 2376, 115 L.Ed.2d 379 (1991) (state's interest in maintaining link between a state district judge's jurisdiction and the area of residency of his or her voters was a legitimate factor as part of “totality of circumstances” in deciding whether arrangement violated 42 U.S.C. § 1973).

But even after the 1999 statutory changes a defendant sued in the wrong township court had to appear in that court (within just ten days) in order to move for a change of venue. If he missed the deadline, he was stuck in that court, however inconvenient for him. Moreover, as the panel opinion in this case pointed out, 734 F.3d at 690–91, the 1999 legislation did not make the township small claims courts entirely independent of one another or of the Marion Circuit Court. The circuit judge was required to “extend aid and assistance to the judges in the conduct of the township small claims courts,” Ind.Code § 33–34–1–5, and to “make and adopt uniform rules for conducting the business of the small claims court,” § 33–34–3–6, and was empowered to transfer cases from one township small claims court to another, § 33–34–5–1, and to arrange for the judges of the township small claims courts to substitute for each other with the consent of the respective judges. § 33–34–5–3.

An investigation of the township courts by a task force of two Indiana Court of Appeals judges identified serious venue problems in those courts. Small Claims Task Force, Report on the Marion County Small Claims Courts, pp. 13–14 www. in. gov/ judiciary/ files/ pubs- smclaims- rept- 2012. pdf (visited July 2, 2014). Many defendants are unaware of their right to ask the courts to transfer a case to the townships where they live. Id. at 13. And paradoxically, although township courts were intended to be more convenient for parties, they could be less convenient than if the venue were countywide. The combination of the size of the county, the nine court locations, limited public transportation other than to and from the center of the county, and the debt collectors' ability to file in any township made it harder for many small claims defendants in Marion County to get to court than it was for defendants in counties in which the courts were centrally located. Id. at 14.

The task force also acknowledged concerns that “large-volume filers appear to file their cases in township courts that appear to provide outcomes favorable to them or provide less oversight for settlement negotiations and settlement agreements,” and that townships have an incentive to pressure judges to “favor large-volume filers in order to generate revenue for the township from filing fees.” Id. Without specifically endorsing those concerns, the task force found that judges who “have made efforts to review settlement terms, as opposed to judges who allegedly rubber-stamp settlement agreements, have seen dramatic declines in new filings in their township courts,” as shown by state judicial statistics. Id. IV. “ Judicial District or Similar Legal Entity ”

The key statutory term that we must interpret—“judicial district”—is not a term of art. It has no statutory definition, and it is inherently flexible, enabling it to be adapted to a variety of state court structures.

The few cases dealing with unusual court structures such as found in Marion County or in Cook County, Illinois, take three different approaches to deciding what is a judicial district or similar legal entity. First, in Newsom we relied on what was said to be the plain language of the statute, 76 F.3d at 816–17, though the language is not plain at all when applied to the Marion County township courts. Going beyond the statutory language, a second approach also found in Newsom and in the panel opinion in the present case focuses on a variety of details of internal judicial administration that affect judges and court personnel.

The third approach, which we adopt, focuses on the state court venue rules faced by parties and lawyers, and the relevant geographic unit for applying those rules. Under this approach the relevant judicial district or similar legal entity is the smallest geographic area relevant to venue in the court system in which the case is filed. This interpretation of the statutory term discourages abusive forum-shopping by debt collectors rather than enabling it. In addition to better serving the debtor-protective policy of the FDCPA than the alternative approaches, this venue-based approach should be more predictable and easier to apply than Newsom's multi-factor test, which requires consideration of numerous details of court administration. A. Plain Language?

Newsom held that a “municipal department district” of the Cook County Circuit Court in Illinois is not a “judicial district or similar legal entity” under § 1692i. A general order of that court had established six geographically distinct “municipal department districts” that would hear civil actions seeking damages not to exceed $30,000, including many consumer debt-collection actions. 76 F.3d at 818. Another court order directed that civil actions be filed in the municipal department district where a defendant resided or the transaction occurred. Id.

Newsom asserted that “judicial district” in the FDCPA has a plain meaning that prevents classifying a “municipal department district” in Cook County as a judicial district. The opinion relied heavily on the edition of a legal dictionary dating from when the FDCPA was enacted, which defined “judicial district” as

one of the circuits or precincts into which a state is commonly divided for judicial purposes; a court of general original jurisdiction being usually provided in each of such districts, and the boundaries of the district marking the territorial limits of its authority; or the district may include two or more counties, having separate and independent county courts, but in that case they are presided over by the same judge.
76 F.3d at 817, quoting Black's Law Dictionary 848 (6th ed.1990), and noting that the definition in the 4th edition, which was the current edition in 1977 when the FDCPA was enacted, was identical.

For two reasons this dictionary definition does not provide a “plain language” resolution of the issue we face. First, its content is too vague to provide meaningful guidance. Dictionaries can be useful in interpreting statutes, see, e.g., Octane Fitness, LLC v. ICON Health & Fitness, Inc., –––U.S. ––––, 134 S.Ct. 1749, 1756, 188 L.Ed.2d 816 (2014), but judges and lawyers must take care not to “overread” what dictionaries tell us.

The most striking features of the dictionary definition in question are its looseness and the inconclusiveness of the component parts of the definition—“circuits or precincts,” “commonly divided,” “usually provided,” and “may include.” Add in the FDCPA's safety valve—“or similar legal entity”—and the dictionary definition, instead of mandating the decision in Newsom, allows ample room for classifying Cook County's divisional courts and Marion County's small-claims courts as “judicial districts.”

Second, even if the definition were narrow and specific enough to support Newsom and the defendant's position in the present case, the critical question would be what weight to give it in interpreting the venue provision of the FDCPA. The dictionary definition has no solid basis in law that helps us with our question. Nor is there any link to the enactment of the FDCPA.

A law dictionary differs from an ordinary dictionary, one would suppose, in basing its definitions on legal rather than ordinary usage. So where did Black's get its definition of “judicial district”? That's not easy to say. The second edition of Black's cited at the end of the definition three cases, presumably believed to be the sources of the definition or to explain or illustrate it: Ex parte Gardner, 22 Nev. 280, 39 Pac. 570 (1895); Lindsley v. Board of Supervisors of Coahoma County, 69 Miss. 815, 11 So. 336 (1892); and Commonwealth v. Hoar, 121 Mass. 375 (1876). The third edition added a fourth case, Consolidated Flour Mills Co. v. Muegge, 127 Okla. 295, 260 Pac. 745 (1927), reversed on other grounds, 278 U.S. 559, 49 S.Ct. 17, 73 L.Ed. 505 (1928) (per curiam), while retaining the previous three. The fourth edition retained all four cases. The fifth edition (1979) retained the definition of “judicial district” (unchanged since the second edition), but dropped the citations without explanation.

The later editor may have realized that the four cases cited in the earlier editions did not support the definition. The definition, interpreted as favorably as it could be to the decision in Newsom, might be thought to imply that a judicial district is created by the legislature and encompasses one or more counties. But the dictionary's case citations undermine that implication. The Lindsley and Hoar cases enforced the division of counties into smaller judicial districts. Consolidated Flour Mills (the only twentieth-century case of the four) rejected a foolish challenge to a court's power to act after the forum county had been reassigned from one multi-county district to another. The court in that case defined “judicial district” very broadly (semantically, rather than geographically): “The term ‘judicial district’ is but a political and convenient arrangement for electing judges of the district court and prescribing primarily the territorial jurisdiction of the district judge, or where he may lawfully preside without special authority from the Supreme Court.” 260 Pac. at 752. That flexible definition encompasses the divisional courts in Newsom and the township small claims courts in the present case.

Gardner is delphic: “A judicial district is simply a political division, provided for by the constitution, but arranged by the legislature, for the purpose of economizing in the number of judges. In fact, the inclusion of any two counties in the same district may almost be said to be accidental.”39 Pac. at 570. The court did go on to hold (and this part of the holding is echoed in the definition of “judicial district” in Black's ) that even though the two counties' courts were in the same “judicial district,” “the courts of those counties are still separate and distinct,” and neither could exercise jurisdiction, even with the parties' consent, over a case brought in the other court. Id. Realistically, then, the two counties in Gardner were separate judicial districts.

The loose definition of “judicial district” in Black's Law Dictionary thus had no real basis in law, as implied by the editor's deletion first of the case citations (in the 1979 edition) and finally of the definition itself (in the 1999 edition).

There is also no indication that the drafters of the Fair Debt Collection Practices Act were aware of the dictionary definition, let alone that they viewed it as helpful, particularly in light of the catch-all extension of § 1692i to a “similar legal entity.” The report of the Senate Committee urging adoption of the proposed Act did not mention the definition but instead expressed concern with

the problem of “forum abuse,” an unfair practice in which debt collectors file suit against consumers in courts which are so distant or inconvenient that consumers are unable to appear. As a result, the debt collector obtains a default judgment and the consumer is denied his day in court. In response to this practice, the bill adopts the “fair venue standards” developed by the Federal Trade Commission. A debt collector who files suit must do so either where the consumer resides or where the underlying contract was signed.... The Commission reports that this standard is effective in curtailing forum abuse without unreasonably restricting debt collectors.
S. Rep. 95–382, at 5, 1977 U.S.C.C.A.N. 1695, 1699. The Senate report is entitled to more weight than a vague dictionary entry. Although the report offers no guidance for our case more specific than a reminder of the statutory goal of preventing abusive forum-shopping, that goal deserves great weight in interpreting an uncertain statutory term.
B. The Judicial Administration Approach in Newsom

Newsom did not rely on the statutory language and the law dictionary alone. It also focused on court administration. The opinion recognized that some orders of the Cook County Circuit Court seemed to reflect a venue requirement that small claims cases be filed in particular subdistricts of the county. But relying on another order of that court, the opinion held that the entire county was the relevant judicial district for purposes of venue under the FDCPA. 76 F.3d at 818–19. That second order provided that any case could be assigned to any judge of the court and heard in any courtroom in the county, regardless of where the case had been filed. The order further provided that no action would be dismissed or judgment vacated because the action had been filed or decided in the wrong department, division, or district. Id. at 819. Newsom concluded that what seemed to be venue rules were merely matters of administrative convenience for the unitary Cook County Circuit Court, which had just one chief judge and one administration, and that the boundaries between municipal department districts did not set any territorial limits to the legal authority of the courts sitting in particular districts. Id.

The panel majority in the present case was understandably reluctant, for reasons of stare decisis, to depart from the reasoning of Newsom. The panel opinion thus reasoned that even though the township small claims courts in Marion County had limited venue, the power of each township court to hear a case from elsewhere in the county and the circuit judge's power to transfer cases and judges among townships made the case similar enough to Newsom to dictate the same outcome.

But Newsom's focus on an array of details of judicial administration lost sight of the purpose of § 1692i: to prevent abusive forum-shopping. By treating an entire county as a “judicial district” even though the county has been subdivided into smaller districts for purposes that included delineating the venue of small claims courts, Newsom turned a protection for consumer debtors into a weapon for debt collectors. Where the county courts are in one central location there is at least a reasonably level playing field for both sides in terms of venue. Newsom gave debt collectors new opportunities for forum-shopping. They could choose from among several courts to find one inconvenient for the debtor, friendly to the collector, or both. Cf. Report on the Marion County Small Claims Courts, supra, at 13–14.

The Newsom approach, which is indifferent to distance and inconvenience even where the state courts use smaller units to decide venue, has even graver implications for counties larger than Cook (land area 1635 square miles) or Marion (403 square miles). There are many much larger counties. The area of the nation's 100th largest county, Eureka County in Nevada, is 4180 square miles—twice the combined area of Cook and Marion Counties. The largest county in the United States is San Bernardino County in southern California. With more than 20,000 square miles, it's more than twelve times larger than Cook County and almost fifty times larger than Marion County.

Like Cook County, San Bernardino County has divisions, each with its own courts, and like Marion County it has small claims courts. But all the courts are part of the Superior Court of San Bernardino County, which corresponds to the Cook County Circuit Court. See Small Claims, Superior Court of California, County of San Bernardino, www. sbcourt. org/ Divisions/ Small Claims. aspx (visited July 2, 2014). The logic of Newsom would allow a debt collector to sue a debtor in any court in San Bernardino County, regardless of distance. And distances in San Bernardino County can indeed be long. The county is more than 200 miles from east to west at its widest point, and 150 miles from north to south.

There are differences between the Cook County Circuit Court in Newsom and the township small claims courts in Marion County in terms of judicial administration, as plaintiff Suesz argues. The township small claims courts have greater independence than Cook County's municipal department district courts, all nine courts having been as we've pointed out established as separate courts with separate election districts, administration, staffing, and funding, and even separate seals, and their separate status and the accompanying venue rules having been created in order to remedy a problem of constitutional dimensions under the Voting Rights Act. Our dissenting colleagues do not find these differences persuasive. These special circumstances might enable us to distinguish Newsom's treatment of the municipal department districts of Cook County. We do not follow that path, however, because the differences in judicial administration between Cook County and Marion County have nothing to do with the purpose of § 1692i in particular or the FDCPA in general. C. The Venue Approach

Our approach is similar to that of the Second Circuit in Hess v. Cohen & Slamowitz LLP, 637 F.3d 117 (2d Cir.2011). A consumer debt collector had sued a debtor in the Syracuse City Court in New York. The city court lacked power to hear the case because the debtor did not reside in Syracuse or a town contiguous to it, though he did live within the county containing the city court. The city court dismissed the debt-collection suit. The debtor then sued the debt collector in federal court under the FDCPA. The district court dismissed the case on the theory that the county was the relevant “judicial district,” and the debt collection suit had been filed in the county in which the debtor lived. The Second Circuit reversed. Its decision was based on the applicable venue rules of the court in which the collection case had been filed. Here are the key passages in its opinion:

Because the court system of which [the collector] availed itself is governed by laws that limit the territorial extent of those courts based on, inter alia, a defendant's contacts with the forum, we hold that those laws delimit the ‘judicial district’ by which compliance with the FDCPA's venue provisions must be measured.

* * * * * *

Where, as here, a state law outlines the required nexus between the residence or activities of the consumer and the location of the court, we hold that such a law sets forth the appropriate ‘judicial district’ for purposes of the FDCPA with respect to debt collection actions brought in that court, regardless of whether that provision is styled as jurisdictional or otherwise.
637 F.3d at 123, 124.

The court made clear that it was focusing on geographic divisions for purposes of determining venue rather than jurisdiction. The geographic limits of the Syracuse City Court's power were not jurisdictional, for they could be waived by the parties. See 637 F.3d at 122–24 & n. 4. They were limits on venue, much like the waivable limits on venue in the Cook County municipal department districts in Newsom and the Marion County township small claims courts in this case.

Focusing on the geographic unit that is relevant for venue in the court system where the case was filed adapts the FDCPA venue provision in § 1692i to the varied court systems among and within the states. That approach worked in Hess for the city courts in New York and it will work for the township small claims courts in Marion County. It should also work in the huge San Bernardino County, which also uses internal districts for small claims cases. See generally Fox v. Citicorp Credit Services, Inc., 15 F.3d 1507, 1515 (9th Cir.1994) (two Arizona counties were separate judicial districts under FDCPA, even though state had one unitary superior court, where the state provided a formal transfer mechanism between counties); Nichols v. Byrd, 435 F.Supp.2d 1101, 1108 (D.Nev.2006) (finding FDCPA venue violation where suit was filed in a township court in which the debtor did not live and the lease giving rise to the alleged debt had not been signed).

What we are calling the “venue approach” should avoid the confusion that would be likely to result if we distinguished between small claims courts that are divisions of a larger court and those that are independent, or if we relied on an ill-defined mixture of administrative details such as whether cases can be transferred for the convenience of the parties among smaller districts within a county or whether judges can substitute for each other, let alone how the courts are managed, funded, and staffed. Such details are not easy for either debt collectors or debtors to learn, and there is no predictable formula for applying them. But more fundamentally, they have nothing to do with preventing abusive forum-shopping to collect consumer debts. D. The Relationship Between Federal and State Law

The Fair Debt Collection Practices Act does not tell states how to organize or operate their court systems. Nor does it directly control court procedures such as venue rules. A debt collector for consumer debts may comply with state law, obtain a perfectly valid state court judgment, and simultaneously violate the FDCPA by suing in the wrong venue.

Suppose, for example, that state law allowed venue in the township where the plaintiff does business, and suppose a debt collector filed all its consumer debt collection suits in that township regardless of where the defendants lived or where the contracts giving rise to the alleged debts had been signed. Those suits would comply with state law but would violate the FDCPA. Such violations would not undermine the validity of state court judgments in favor of a debt collector, but they would provide the basis for federal remedies against the debt collector.

In essence, then, the FDCPA takes state courts as states choose to structure and operate them. Section 1692i then provides federal remedies for violations of the new federal requirements for venue in consumer debt-collection cases covered by the federal law. The remedies are available whether or not the filing of the case complies with state law.

At the same time, and for the same reasons, we do not think it matters for purposes of § 1692i whether the state venue rules are established by state statute or court rule (as here), by standing court order (as in Newsom ), or by any other mechanism. The relevant judicial district or similar legal entity is the smallest geographic unit relevant for venue purposes in the court system in which the case was filed, regardless of the source of the venue rules.

One implication of our analysis is that a consumer debt collector filing suit in Marion County still retains a limited choice of venue, at least in theory and as a matter of federal law. The jurisdiction of the township small claims courts over small claims cases is concurrent with the jurisdiction of the county's circuit and superior courts. Ind.Code § 33–34–3–2. In the circuit and superior courts in Marion County, as in the rest of Indiana, venue is county-wide. If therefore a debt collector chooses to file suit in a circuit or superior court, he could file it in a courthouse in the center of the county. But if the debt collector chooses to file suit in a township small claims court, venue is determined at the township level, thus requiring the debt collector to select a township consistent with the FDCPA's limitations on abusive forum-shopping.

In practice, though, the potential for confusion or misuse should be minimal. The judges of the circuit and superior courts can transfer any small claims case to an appropriate township small claims court, Ind.Code § 33–34–5–2, and this is done routinely. (In fact, despite the concurrent jurisdiction, the clerk of those county courts informs litigants that a case seeking a judgment equal to or less than $6,000 must be filed in a small claims court and that the county clerk's office does not even support the small claims courts. See Civil Filings: Case Type, City of Indianapolis and Marion County, www. indy. gov/ e Gov/ County/ Clerk/ civil/ Pages/ Case- Types. aspx (visited July 2, 2014).) Also, the civil filing fee in the circuit and superior courts is nearly twice the filing fee for a small claims case ($141 versus $82), and the township small claims courts generally move their civil dockets faster than the circuit and superior courts, which makes the small claims courts more attractive to debt collectors. Again, however, the FDCPA takes the state courts as it finds them.

V. Retroactivity

Our interpretation of § 1692i requires us to reverse the judgment of the district court and to remand for further proceedings on class certification and the merits of plaintiff's claim. But Med–1 Solutions, perhaps seeing the handwriting on the wall, asks that if we overrule Newsom, as we do today, we should do so only on a prospective basis. It argues that debt collectors have relied on Newsom to allow them to choose venue anywhere in the appropriate county.

As a general matter, adopting a new rule while refusing to apply it to the parties before us would raise serious constitutional concerns. We exercise judicial authority rather than the prospective authority of a legislature. See Harper v. Virginia Dep't of Taxation, 509 U.S. 86, 106, 113 S.Ct. 2510, 125 L.Ed.2d 74 (1993) (Scalia, J., concurring) (“The true traditional view is that prospective decisionmaking is quite incompatible with the judicial power, and that courts have no authority to engage in the practice.”) (emphasis in original); James B. Beam Distilling Co. v. Georgia, 501 U.S. 529, 547, 111 S.Ct. 2439, 115 L.Ed.2d 481 (1991) (Blackmun, J., concurring in judgment) (“the nature of judicial review constrains us to consider the case that is actually before us, and, if it requires us to announce a new rule, to do so in the context of the case and apply it to the parties who brought us the case to decide. To do otherwise is to warp the role that we, as judges, play in a Government of limited powers”); Firestone Tire & Rubber Co. v. Risjord, 449 U.S. 368, 379, 101 S.Ct. 669, 66 L.Ed.2d 571 (1981) (no prospective application of jurisdictional rulings).

The Supreme Court has left itself some room to give its rulings in civil cases only prospective effect, at least “to avoid injustice or hardship to civil litigants who have justifiably relied on prior law.” See Harper v. Virginia Dep't of Taxation, supra, 509 U.S. at 110–13, 113 S.Ct. 2510 (Kennedy, J., concurring in part and concurring in the judgment), quoting American Trucking Ass'ns v. Smith, 496 U.S. 167, 199, 110 S.Ct. 2323, 110 L.Ed.2d 148 (1990) (plurality opinion); Felzen v. Andreas, 134 F.3d 873, 877 (7th Cir.1998).

The Supreme Court's reservation of such a power does not persuade us to make the present decision effective only prospectively; and this for two reasons: First, reliance on prior law is insufficient in itself to justify making a new judicial ruling prospective. See Reynoldsville Casket Co. v. Hyde, 514 U.S. 749, 753–54, 115 S.Ct. 1745, 131 L.Ed.2d 820 (1995) (reversing state court's decision to give new U.S. Supreme Court ruling only prospective effect). Second, a prior decision of one intermediate appellate court does not create the degree of certainty concerning an issue of federal law that would justify reliance so complete as to justify applying a decision only prospectively in order to protect settled expectations. See Anderson–Bey v. Zavaras, 641 F.3d 445, 454–55 (10th Cir.2011) (declining to apply new decision only prospectively despite party's reliance on prior circuit decision); Hawknet, Ltd. v. Overseas Shipping Agencies, 590 F.3d 87, 91 & n. 7 (2d Cir.2009) (same); but see Nunez–Reyes v. Holder, 646 F.3d 684, 690 (9th Cir.2011) (en banc) (giving only prospective effect to new decision conforming circuit law to decisions of other circuits and of the Board of Immigration Appeals). Prospective overruling on reliance grounds is impermissible unless the law had been so well settled before the overruling that it had been unquestionably prudent for the community to rely on the previous legal understanding.

So suppose we affirmed the dismissal of this case and the Supreme Court then granted certiorari and reversed. Neither our prior decision in Newsom nor the panel's decision in this case would have justified giving the ruling only prospective effect on the basis of justified reliance by (in this case) debt collectors in Marion County.

The judgment of the district court dismissing this action is REVERSED and the case is REMANDED to the district court for further proceedings consistent with this opinion.

SYKES, Circuit Judge, concurring.

This case concerns the proper interpretation of the so-called “venue provision” in the Fair Debt Collection Practices Act (“FDCPA”). Here's the full text of the statute, with the key language italicized:

§ 1692i. Legal actions by debt collectors

(a) Venue

Any debt collector who brings any legal action on a debt against any consumer shall

(1) in the case of an action to enforce an interest in real property securing the consumer's obligation, bring such action only in a judicial district or similar legal entity in which such real property is located; or

(2) in the case of an action not described in paragraph (1), bring such action only in the judicial district or similar legal entity

(A) in which such consumer signed the contract sued upon; or

(B) in which such consumer resides at the commencement of the action.
15 U.S.C. § 1692i (emphases added).

By its terms, § 1692i requires any debt collector who sues on a consumer debt to file suit in the “judicial district or similar legal entity” in which the consumer resides or contracted the debt. But the statute does not define “judicial district,” and the term is difficult to interpret and apply to debt-collection actions filed in state courts, which vary widely in their administrative structures and rules for jurisdiction and venue.

In Newsom v. Friedman, 76 F.3d 813 (7th Cir.1996), we resolved the interpretive difficulty by extrapolating from the Black's Law Dictionary definition of “judicial district” and also by looking to the jurisdictional, procedural, and administrative rules that govern this kind of litigation in the relevant state court system. The panel majority followed that approach here. We reheard the case en banc to consider whether to overrule Newsom and adopt a different interpretation of the phrase “judicial district or similar legal entity.”

I agree with much of what Judges Hamilton and Posner have written for the en banc court, as well as the court's decision to overrule Newsom. But I also share some of the concerns expressed by Judge Flaum in dissent, which I understand to be rooted at least in part in principles of federalism. In the end, I arrive at the same conclusion as the majority, though by a somewhat different route.

Section 1692i is an unusual federal statute. As written, it establishes a venue rule for “any legal action” to collect a consumer debt, including debt-collection actions filed in state court. Most suits to collect consumer debts are state-law claims for breach of contract, and most of these actions are filed in state court because they don't meet the $75,000 amount-in-controversy threshold for invoking the federal court's diversity jurisdiction. SeeFed. Trade Comm'n, Repairing a Broken System: Protecting Consumers in Debt Collection Litigation and Arbitration 6 (2010) (“Debt collection lawsuits almost invariably are filed in state courts....”); Fed. Reserve Bank of N.Y., Quarterly Report on Household Debt and Credit 15 (Nov.2012) (reporting that the average collection amount was about $1,500 in the first quarter of 2012); see also28 U.S.C. § 1332. But it's doubtful that Congress has the power to prescribe procedural rules for state-law claims in state courts. See generally Anthony J. Bellia Jr., Federal Regulation of State Court Procedures, 110 Yale L.J. 947 (2001); Wendy E. Parmet, Stealth Preemption: The Proposed Federalization of State Court Procedures, 44 Vill. L.Rev.. 1 (1999).

Of course, when Congress creates a cause of action over which the state courts have concurrent jurisdiction, the state courts are bound by the Supremacy Clause to adjudicate the claim, see Testa v. Katt, 330 U.S. 386, 393–94, 67 S.Ct. 810, 91 L.Ed. 967 (1947), and sometimes this includes the obligation to follow federal procedural rules that are specifically tied to the federal claim. For example, the Supreme Court has held that some federal procedural rules may apply in state-court litigation if the rules are “part and parcel” of the federal cause of action being adjudicated in state court. Dice v. Akron, Canton & Youngstown R.R., 342 U.S. 359, 363, 72 S.Ct. 312, 96 L.Ed. 398 (1952) (holding that the statutory right to a jury trial in actions under the Federal Employers' Liability Act (“FELA”) applies in Ohio state court despite a state procedural rule requiring that certain factual questions be decided by the court); see also Cent. Vt. Ry. Co. v. White, 238 U.S. 507, 512, 35 S.Ct. 865, 59 L.Ed. 1433 (1915) (holding that a FELA provision allocating the burden of proving contributory negligence to the defendant applies in state court despite a contrary state procedural rule).

The Court has also held that state procedural rules may be displaced when they conflict with or unnecessarily burden the substance of a federal cause of action being litigated in state court. See, e.g., Felder v. Casey, 487 U.S. 131, 150, 108 S.Ct. 2302, 101 L.Ed.2d 123 (1988) (holding that a state law imposing a 120–day notice-of-injury prerequisite for claims against governmental defendants is preempted in actions under 42 U.S.C. § 1983); Brown v. W. Ry. of Ala., 338 U.S. 294, 298–99, 70 S.Ct. 105, 94 L.Ed. 100 (1949) (barring application of a state-court pleading rule that interfered with the plaintiff's substantive federal rights).

But these cases involved federal claims being adjudicated in state court. It's an open question whether Congress has the power to prescribe procedural rules for state-law claims in state court. The Supreme Court has twice noted the issue but declined to decide it. In Jinks v. Richland County, 538 U.S. 456, 123 S.Ct. 1667, 155 L.Ed.2d 631 (2003), the Court addressed a constitutional challenge to a provision in the supplemental jurisdiction statute that tolls the statute of limitations on a supplemental state-law claim “while the claim is pending and for a period of 30 days after it is dismissed unless State law provides for a longer tolling period.” 28 U.S.C. § 1367(d). The Court held that the tolling provision is “necessary and proper for carrying into execution Congress's power ‘[t]o constitute Tribunals inferior to the supreme Court,’ U.S. Const., Art. I, § 8, cl. 9, and to assure that those tribunals may fairly and efficiently exercise ‘[t]he judicial Power of the United States,’ Art. III, § 1.” Jinks, 538 U.S. at 462, 123 S.Ct. 1667. The challenger had also argued that § 1367(d) violates the principles of state sovereignty articulated in Printz v. United States, 521 U.S. 898, 918–22, 117 S.Ct. 2365, 138 L.Ed.2d 914 (1997), see also New York v. United States, 505 U.S. 144, 112 S.Ct. 2408, 120 L.Ed.2d 120 (1992), and that Congress lacks the authority to “prescribe procedural rules for state courts' adjudication of purely state-law claims.” Jinks, 538 U.S. at 464, 123 S.Ct. 1667. The Court construed the tolling provision as a substantive rule, not a procedural one, and on that basis declined to address the state-sovereignty question. Id. at 465, 123 S.Ct. 1667 (“[T]he tolling of limitations periods falls on the ‘substantive’ side of the line. To sustain § 1367(d) in this case, we need not (and do not) hold that Congress has unlimited power to regulate practice and procedure in state courts.”).

In Pierce County v. Guillen, 537 U.S. 129, 132–33, 123 S.Ct. 720, 154 L.Ed.2d 610 (2003), the Court rejected a constitutional challenge to 23 U.S.C. § 409, which protects traffic-hazard data “compiled or collected” by the States pursuant to certain federal highway safety programs from being used as evidence in federal and state court proceedings. The Court held that § 409 was a proper exercise of Congress's power under the Commerce Clause to regulate the channels of interstate commerce—there, the nation's highways. Id. at 146–47, 123 S.Ct. 720. In a footnote the Court noted that the challengers had also argued that § 409 “violates the principles of dual sovereignty embodied in the Tenth Amendment because it prohibits a State from exercising its sovereign powers to establish discovery and admissibility rules to be used in state court for a state cause of action.” Id. at 148 n. 10, 123 S.Ct. 720. The Court declined to address the question because the lower court had not yet done so. Id.

No constitutional argument is raised here; this case presents only a question of statutory interpretation. Still, our statutory interpretation inquiry should be informed by important background principles of federalism. See Bond v. United States, ––– U.S. ––––, 134 S.Ct. 2077, 2090, 189 L.Ed.2d 1 (2014) (“[I]t is appropriate to refer to basic principles of federalism embodied in the Constitution to resolve ambiguity in a federal statute.”); see also Gregory v. Ashcroft, 501 U.S. 452, 460–61, 111 S.Ct. 2395, 115 L.Ed.2d 410 (1991).

My colleagues seem to acknowledge this point, if only implicitly. For example, they say that § 1692i “does not tell states how to organize or operate their court systems” or “directly control court procedures such as venue rules,” but instead “takes the state courts as it finds them.” Majority op. at 647–48, 649. I agree, but not because that limitation is clear from the text of the statute, which on its face establishes a uniform federal venue rule applicable to all actions to collect consumer debts, even state-law actions filed in state court. Section 1692i can function as written—as a venue rule—for debt-collection actions in federal court. But it can't function that way for debt-collection actions in state court, or at least it probably can't. Operating a judicial system is a core function of state government, and because the States have the sovereign authority to structure their court systems and establish their own jurisdictional and procedural rules, we must give § 1692i an interpretive gloss—by saying that it “takes the state courts as it finds them”—in order to avoid serious constitutional difficulty.

So as applied to debt-collection actions in state court, § 1692i must be understood not as a venue rule but as a penalty on debt collectors who use state venue rules in a way that Congress considers unfair or abusive. See15 U.S.C. § 1692k(a) (establishing a civil remedy for damages against any debt collector who “fails to comply with any provision of this subchapter”); see also15 U.S.C. § 1692 (congressional findings and declaration of purpose). Even on this understanding, however, the statute is not free from constitutional doubt. If Congress lacks the authority to prescribe venue rules for the state courts, then it may also lack the authority to impose a federal penalty—in the form of a damages remedy—on litigants who do not file their state-court lawsuits where Congress says they must. In the final constitutional analysis, there may not be much difference between a federal law establishing a venue rule for state-law litigation in state court (probably unconstitutional) and a federal law establishing a damages remedy for not following a federally prescribed rule about venue in state court.

My colleagues note that “[t]he presence of the venue provision in the Act shows congressional dissatisfaction with allowing state law to determine where suits to collect consumer debts can be filed.” Majority op. at 640. No doubt that's true, but this just highlights the constitutional question about the limits on Congress's authority to regulate venue in the state courts. They say that § 1692i doesn't “alter[ ] state rules governing jurisdiction and venue” but “merely imposes a limit on which state courts having jurisdiction and venue over a debt collector's claim the debt collector may sue in, consistent with the policy of the federal law.” Id. In the next sentence, we are reassured that “[t]here is nothing unusual or untoward about requiring compliance with both state and federal rules.” Id. I'm not so sure that's true in this context. The States have the sovereign authority to establish the jurisdictional and procedural rules that apply in their courts. Section 1692i overrides state venue rules, albeit indirectly, by imposing a federal damages remedy against state-court litigants who do not comply with the federal requirement.

Sometimes Congress can do indirectly what it lacks the power to do directly. See, e.g., South Dakota v. Dole, 483 U.S. 203, 210–11, 107 S.Ct. 2793, 97 L.Ed.2d 171 (1987) (holding that Congress may condition States' receipt of a portion of their federal highway funds on adoption of a drinking age of 21). But it's not clear to me that indirect regulation is permissible here. When construed as a penalty against debt collectors for litigating in state court in a way that violates federal policy, § 1692i may or may not be a valid exercise of Congress's power under the Commerce Clause. But even an affirmative answer to that question “does not conclusively resolve the constitutionality of the [statute]”; the federalism question under the Tenth Amendment remains. Reno v. Condon, 528 U.S. 141, 149, 120 S.Ct. 666, 145 L.Ed.2d 587 (2000) (holding that a law permissible under the Commerce Clause may still be unconstitutional under the Tenth Amendment). And we know that private parties, not just the States, may bring Tenth Amendment challenges. See Bond v. United States, ––– U.S. ––––, 131 S.Ct. 2355, 2363–64, 180 L.Ed.2d 269 (2011).

At a minimum § 1692i must be interpreted with the limits on Congress's power in mind. This requires, as Judge Flaum suggests, sensitivity to the structure of each State's court system and the particular rules that govern its judicial subdivisions.

Returning to the interpretive problem at hand, to avoid liability under the FDCPA for violation of § 1692i, a debt collector must file suit in the “judicial district” in which the consumer resides or contracted the debt. As applied to debt-collection actions in state courts, the phrase “judicial district” is ambiguous because the States structure their court systems in a variety of ways. They may not use the term “district” to describe their judicial subdivisions, or they may use the term in a way that is not relevant to determining compliance with § 1692i.

I agree with Judges Hamilton and Posner that the Black's definition of “judicial district”—an interpretive tool prominently featured in Newsom—doesn't help resolve the ambiguity. When the FDCPA was adopted, Black's defined the term by reference to the territorial boundaries of the court's authority and the presence of a court of general subject-matter jurisdiction. Judges Hamilton and Posner explain at length why this definition is not useful. I concur and have nothing to add to this discussion.

I agree as well that state venue rules play an important role in determining the appropriate “judicial district” under § 1692i in debt-collection actions filed in state court. So too does the structure of the State's court system. Deciding how far to drill down into the State's judicial hierarchy to find the relevant “judicial district” for purposes of § 1692i will depend in large part on how the State structures its court system for jurisdiction and venue over this kind of litigation.

Section 1692i is obviously aimed at promoting a venue convenient to the consumer debtor. In light of this manifest purpose, and in deference to the prerogatives of the States to set their own procedural rules, I agree that the phrase “judicial district or similar legal entity” is best understood to mean the judicial subdivision, defined by state law, that is relevant to determining venue in the court system in which the case is filed.

See Majority op. at 637, 642–43, 647–48. Judges Hamilton and Posner amply explain why this is so, and again I have nothing to add to their analysis.

This construction of § 1692i resolves the statutory ambiguity but does not remove the constitutional cloud. The constitutional question is for another day.

One final point before I conclude. Looking to state venue rules at the “system” level is important. In deference to the State's right to set its own procedural rules for litigation in its courts, § 1692i probably can't be understood to impose liability on debt collectors for using a court of record instead of a local court (i.e., a limited jurisdiction, nonrecord municipal court) if state law makes the former available. As Judges Hamilton and Posner explain, under Indiana law the small-claims jurisdiction of the Marion County township courts is concurrent with the county's circuit and superior courts. Majority op. at 648; see alsoInd.Code § 33–34–3–2. The township courts, however, are not courts of record. SeeInd.Code § 33–34–1–3.

There may be practical and administrative impediments to filing a small-claims action in the Marion County circuit or superior courts, see Majority op. at 648–49, and debt collectors apparently prefer to use the township courts, see id. at 642–43, 649. But sensitivity to federalism suggeststhat § 1692i should not be interpreted to penalize debt collectors for suing in a court of record if state law gives them that option. In other words, the statute should not be interpreted to force debt collectors to use a local, nonrecord court system if state law permits them to sue in a court of record. That's why compliance with § 1692i is determined by reference to the judicial subdivisions that are relevant for venue purposes in the court system in which the debt collector has chosen to file.

Accordingly, with these limitations and additional observations, I agree with the court's interpretation of § 1692i and the decision to overrule Newsom, and I join the judgment reversing and remanding for further proceedings.

FLAUM, Circuit Judge, with whom KANNE, Circuit Judge, joins, dissenting.

The court's new rule—defining “judicial district” in § 1692i of the Fair Debt Collection Practices Act as the smallest geographic area used for determining venue wherever the debt collector files the case—may be a laudable one, as a matter of policy. However, our task is not to fine-tune statutes, but to reason through Congress's language as we find it. And while the majority's definition is arguably consistent with the general purpose behind the FDCPA, it is clearly not compelled—nor even suggested—by the statutory text.

When Congress does not define a statutory term or phrase, courts normally use its ordinary meaning. See, e.g., Sebelius v. Cloer, ––– U.S. ––––, 133 S.Ct. 1886, 1893, 185 L.Ed.2d 1003 (2013). Congress's omission of a definition for the phrase “judicial district” suggests that it meant to refer to judicial districts as they are defined by the government that established the relevant courts. For debt-collection suits, those courts are usually state courts. In this case, the state is Indiana, so our inquiry should be whether Indiana created separate judicial districts when it established the Marion County township courts. In my view, the township courts are subdivisions of a single judicial district, which is Marion County as a whole. Cf.Ind.Code § 33–33–49–2 (“Marion County constitutes the nineteenth judicial circuit.”).

But instead of deferring to the state's definition of its districts, the majority replaces congressional silence in § 1692i with a purposive definition of judicial district that is of the majority's own design. In doing so, the court federalizes the term “judicial district” for the purposes of the FDCPA. I decline to join this decision because I believe the court's rule seizes upon a general congressional purpose behind the FDCPA—protecting debtors from abusive collection practices—to craft a rule more exacting than Congress intended. A high-level statutory purpose is simply an insufficient justification for this stringent new rule.

To take a step back: § 1692i makes it a violation of federal law for a debt collector to file a collection action any place other than the “judicial district or similar legal entity [where the] consumer signed the contract sued upon [or where the] consumer resides.” 15 U.S.C. § 1692i. This prohibition restricts the locations where a collection action can lawfully be filed, but only to the level of the “judicial district” (or similar legal entity); if a judicial district is subdivided into smaller component parts, a debt collector may still have some leeway in deciding where to file.

The majority holds that the judicial district referenced in § 1692i should be defined so as to advance the federal goals of combating unfair and abusive collection practices. But as I read the statute, that section directs us to incorporate judicial districts as they are defined by the states— § 1692i is not meant to modify the concept of judicial district. The majority believes that this approach renders § 1692i meaningless, because if Congress merely wished to adopt state jurisdictional rules, there would have been no need for a federal prohibition in the FDCPA. This understanding is incorrect for two reasons.

First, under my interpretation of § 1692i, the FDCPA still imposes a harsher penalty on collectors who fail to honor existing state jurisdictional and venue rules. Congress meant to combat default-judgment-seeking debt collectors who file suit in courts “so distant or inconvenient that consumers are unable to appear.” S.Rep. No. 95–382, at 5 (1977). Without the FDCPA's federal overlay, this tactic may be a low-stakes gambit—the only downside to filing in the wrong court is that the case will get moved if perchance the debtor shows up. Section 1692i thus gives state-law jurisdiction and venue rules teeth.

Second, my reading of § 1692i still places meaningful restrictions on state law, by restricting filings to the judicial district where the debtor lives or signs a contract. Even if state venue rules permitted a creditor to file in the district where the debtor works, for example, this would not be permissible under the FDCPA.

Where I part with the majority is its sole emphasis on venue rules to define the scope of a judicial district or similar legal entity. Judicial districts are products of positive law—they exist because a governmental entity established courts and then divided them into units. But states need not set the boundaries of their judicial districts by their venue rules alone. It is perfectly appropriate for a state to choose to consider other factors, like geography, administrability, convenience, or subject-matter specialization—as I believe Indiana has done in Marion County. Indeed, the evidence of congressional intent regarding § 1692i suggests that Congress intended to preserve state law: a House Committee report indicates that the FDCPA's venue provision does “not change State or Federal law relating to venue [or] service of process.” H.R.Rep. No. 95–131, at 6 (1977). But the majority's approach could force a de facto change on the states.

We simply have no basis to conclude that Congress would have thought it unfair or abusive if a debt collector files suit within the state-defined judicial district, but in a venue that is not closest to the debtor. If Congress had wanted to impose a new federal concept of a judicial district—one more exacting than the existing units created by state law—Congress would have had to do so far more clearly. Cf. BFP v. Resolution Trust Corp., 511 U.S. 531, 543–45, 114 S.Ct. 1757, 128 L.Ed.2d 556 (1994) (defining a disputed term in the Bankruptcy Code in light of relevant state law and noting that Congress must be explicit if it wishes to adjust the balance of state and federal authority). Further, it makes sense that Congress would reference the judicial districts already created by the states, as the FDCPA must be applied to all fifty of them despite their individual structural idiosyncrasies. The Federal Trade Commission's fair venue standards, cited approvingly in a Senate Committee report on the FDCPA, likewise do not define judicial district, but instead take the state's structure as a given. S.Rep. No. 95–382, at 5. In short, I see no indication that Congress would have thought “judicial district or similar legal entity” meant only “the smallest geographic area that is relevant for determining venue.” Op. at 638. (Indeed, as Judge Kanne's dissent demonstrates, the peculiarity that would result from applying the majority's rule in federal debt-collection actions strongly suggests that this rule is not what Congress had in mind.)

Our decision in Newsom v. Friedman, 76 F.3d 813 (7th Cir.1996), properly recognized that states are responsible for defining their own judicial districts. Newsom refers to the definition of “judicial district” found in the Black's Law Dictionary in effect when the FDCPA was enacted: “One of the circuits or precincts into which a state is commonly divided for judicial purposes; a court of general original jurisdiction being usually provided in each of such districts, and the boundaries of the district marking the territorial limits of its authority.” Newsom, 76 F.3d at 817.

I continue to believe that Newsom's virtue lies in the way it balances the protection of debtors with the realities of varied and unique state-court systems. Newsom has us look to state law—and more specifically to court rules and administration—to discover how the state defines its judicial districts.

Judge Kanne thoroughly explains why the county, and not the township, is the proper judicial district when we focus on Indiana specifically. But I hasten to add that in many cases, the analysis under either Newsom or the court's new rule will be the same. For example, if an Indiana debt collector files in the superior court—a county-level trial court of general jurisdiction that is found throughout the state—the analysis is straightforward and the debt collector need only file in the proper county. The new rule is satisfied because the county is the smallest area for determining venue in the superior court. And the county is the judicial district under Newsom because the superior court is the court of general jurisdiction and the common division statewide. 76 F.3d at 818. Things get trickier in courts of limited jurisdiction like the township courts in this case. The difficulty arises in deciding whether these limited-jurisdiction courts are subdivisions of a larger district or freestanding judicial districts themselves. Limited-jurisdiction courts can be judicial districts, but need not be in all cases. Compare Newsom, 76 F.3d at 819 (limited-jurisdiction municipal-department districts were not separate judicial districts), with Hess v. Cohen & Slamowitz, 637 F.3d 117, 119 (2d Cir.2011) (Syracuse city courts were separate judicial districts). For these limited-jurisdiction courts, Newsom asks how the courts function and considers territorial restrictions on the filing of suits. 76 F.3d at 817–19. This inquiry requires us to consider whether an action has been filed in the proper judicial district as defined by the state—which is quite likely what Congress had in mind.

The majority opinion deconstructs the historical bona fides behind the Black's Law definition of judicial district and concludes that there is no common-law definition of the phrase. But I do not find this analysis particularly illuminating. In Newsom, we turned to Black's in part because its definition took a commonsense tack to defining a judicial district. What is important about Newsom's definition is that it gives courts valuable guidance while also providing leeway to decide what states consider to be their own judicial districts.

I close by noting that I believe Newsom is consistent with the Second Circuit's decision in Hess, 637 F.3d 117—more so, in fact, than the court's rule. Like Newsom, Hess directs courts to look at the specifics of state law to define judicial districts. The Syracuse city courts at issue in that case had no authority to hear a case if the resident lived outside of Syracuse or the adjoining towns. If a defendant from outside this zone showed up to court and objected, the court did not transfer a case, but dismissed it, because the filing was improper under state law. Id. at 121–23. Therefore, the Hess court concluded that the city court was a freestanding judicial district under New York law. Newsom similarly looked to state law. But in contrast to the city court in Hess, the courts at issue in Newsom were not judicial districts, in part because the filing of the debt collection suits was not improper under state law.

The court's new rule, contrary to Hess and Newsom, does not give appropriate deference to the way a state chooses to structure its own court system. For this reason, I would uphold Newsom, and I respectfully dissent from the majority's decision to overrule it.

KANNE, Circuit Judge, dissenting.

This is a simple statutory interpretation case. The parties ask us to determine the meaning of the words “judicial district or similar legal entity” as they appear in the venue provision of the Fair Debt Collection Practices Act, and the majority is happy to oblige. But there is a problem: This court has already answered that question. In Newsom v. Friedman, 76 F.3d 813 (7th Cir.1996), this court found that the operative phrase was unambiguous and susceptible of a plain-meaning interpretation. We relied on the edition of Black's Law Dictionary current when the FDCPA was passed to define a judicial district or similar legal entity as:

One of the circuits or precincts into which a state is commonly divided for judicial purposes; a court of general original jurisdiction being usually provided in each of such districts, and the boundaries of the district marking the territorial limits of its authority; or the district may include two or more counties, having separate and independent county courts, but in that case they are presided over by the same judge.
Newsom, 76 F.3d at 817 (quoting Black's Law Dictionary 848 (6th ed.1990) (and referencing earlier editions)). Since then, the only other circuit court to directly consider the issue adopted the exact same definition. Hess v. Cohen & Slamowitz LLP, 637 F.3d 117, 121 (2d Cir.2011).

We have said, too many times to count, that we require a “compelling” reason to overturn circuit precedent, as the principle of stare decisis requires that we give considerable weight to prior decisions of this court. United States v. Lara–Unzueta, 735 F.3d 954, 961 (7th Cir.2013). The presumption in favor of existing law is overcome only in the most exceptional of circumstances, such as when our decision is overruled or undermined by the decisions of a higher court or by legislative action. Id. We might also consider whether the existing rule is simply unworkable as a practical matter, whether surrounding principles of law have developed so far as to leave an old rule behind, or whether facts have changed so much as to rob the rule of its general applicability or justification. Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833, 854–55, 112 S.Ct. 2791, 120 L.Ed.2d 674 (1992). What we will certainly not do is overturn a previous decision simply because a majority of the members of this court believe it was incorrect:

[I]f the fact that a court considers one of its previous decisions to be incorrect is a sufficient ground for overruling it, then stare decisis is out the window, because no doctrine of deference to precedent is needed to induce a court to follow the precedents that it agrees with; a court has no incentive to overrule them even if it is completely free to do so. The doctrine of stare decisis imparts authority to a decision, depending on the court that rendered it, merely by virtue of the authority of the rendering court and independently of the quality of its reasoning. The essence of stare decisis is that the mere existence of certain decisions becomes a reason for adhering to their holdings in subsequent cases.
Tate v. Showboat Marina Casino P'ship, 431 F.3d 580, 582–83 (7th Cir.2005) (citation and internal quotation marks omitted).

Or so I thought. Despite the fact that the principle of stare decisis carries “special force” in the statutory interpretation context, John R. Sand & Gravel Co. v. United States, 552 U.S. 130, 139, 128 S.Ct. 750, 169 L.Ed.2d 591 (2008)—so much force, in fact, that asking us to overrule a previous statutory interpretation decision can result in the summary disposition of your appeal, see, e.g., United States v. Howell, 557 Fed.Appx. 579 (7th Cir.2014)—I cannot find any remotely valid justification for overruling Newsom in the entirety of the majority's opinion. It makes no difference that Newsom was a panel decision. Nearly all of the decisions that issue from this court are panel decisions, and the principles of stare decisis still apply. The fact that we are now sitting en Banc means that we have the authority to disregard a prior decision, but it does not mean that we can or should do so without a valid reason.

What reason is given? The majority opinion complains that the Newsom/Black's definition is too vague, and that the sources cited by Black's are underwhelming. Those are not sufficient reasons to abandon existing law; they are just another way of saying the majority believes that Newsom was decided incorrectly. After casting that decision aside, the majority then moves immediately to the policy objectives underlying the FDCPA, and summarily redefines the phrase “judicial district or similar legal entity” to mean “the smallest geographic area that is relevant for determining venue in the court system in which the case is filed.” Even if I leave the stare decisis issue behind, that definition is wrong because the process that led to it disregarded the established canons of statutory construction, and it is wrong because it cannot be usefully or consistently applied to the court systems in our jurisdiction. I write separately to address each of those points, and I ultimately join the dissent of Judge Flaum.

I.

I will begin by noting the most obvious flaws in the majority's process. As I have mentioned, the core of this case is a simple question of statutory construction. The first canon of statutory construction is that we begin with the text itself. Connecticut Nat'l Bank v. Germain, 503 U.S. 249, 253–54, 112 S.Ct. 1146, 117 L.Ed.2d 391 (1992). Absent ambiguity, the first canon is also the last: “judicial inquiry is complete.” Id. (quoting Rubin v. United States, 449 U.S. 424, 430, 101 S.Ct. 698, 66 L.Ed.2d 633 (1981)). The majority opinion summarily concludes that the words “judicial district or similar legal entity,” at least in this context, are ambiguous, and so turns to legislative intent to discern the meaning of the statute. Like Judge Flaum, I believe the majority is mistaken. The operative phrase is unambiguous and susceptible of a common understanding, and there is therefore no need to turn to policy concerns or legislative history to decide this case. By shortchanging the first step in the process, the majority has engaged in disordered statutory interpretation.

First, not only was the original panel's reference to Black's Law Dictionary consistent with existing law as set out in Newsom, it was consistent with usual interpretive practice. “Without a statutory definition, we construe [a] term ‘in accordance with its ordinary or natural meaning,’ a meaning which may be supplied by a dictionary.” Carmichael v. The Payment Ctr., Inc., 336 F.3d 636, 640 (7th Cir.2003) (quoting FDIC v. Meyer, 510 U.S. 471, 476, 114 S.Ct. 996, 127 L.Ed.2d 308 (1994)). I agree with the majority that there are dangers to blind reliance on dictionary definitions, but as Judge Flaum rightly notes, this particular dictionary provided a definition that comports with common sense. I do not grasp the value of the majority's search for a definitive source underlying the Black's definition. Dictionaries—even legal ones—are descriptive, not prescriptive; their goal is to define words according to their common usage. That is why the meanings of words often change from one edition to another, and why some new words enter a dictionary and some old words leave. Does the majority mean to suggest that Black's was wrong about the common usage of the term “judicial district” at the time the statute was enacted? If so, on what basis? If not, what is the point?

Second, the Black's definition is not “vague” at all, at least not in the context of this case. Black's tells us to look for “one of the circuits or precincts into which a state is commonly divided for judicial purposes.” In Indiana, that is easy to do. The “circuit or precinct” into which Indiana is “commonly divided for judicial purposes” is the “judicial circuit.”

See Ind. Const. Article 7, Sections 1, 7–8 (empowering the General Assembly to divide the state into judicial circuits); Ind.Code Title 33, Article 28 (establishing the powers and duties of the circuit courts, generally); see also7 Ind. Law Encyc. Courts § 17. Indiana has 92 counties, and all but two of them (the smallest two, Ohio and Dearborn) comprise their own judicial circuit and have their own circuit court. See Ind.Code Title 33, Article 33, Chapters 1–92 (establishing and numbering a judicial circuit and circuit court for each Indiana county). For most counties, the Indiana legislature has created “superior courts” to share the load with the circuit court, or to supplant it as the primary trial court within that judicial circuit. See, e.g.,Ind.Code § 33–33–49–6 (Marion County). With respect to at least one county, the legislature has taken a different approach, and created additional circuit judgeships. See, e.g., Ind.Code 33–33–53–1–8 (Monroe County). Some counties have been legislatively authorized to operate separate small claims mechanisms within the circuit court. SeeInd.Code § 33–28–3–1–10. One county, Saint Joseph, has a separate probate court created by the legislature. Ind.Code § 33–31–1–1. In short, county-by-county structures vary a great deal, but all of those structures exist within a judicial circuit. The judicial circuit is the level at which uniformity can be found; the common division for judicial purposes.

It is just as easy to see that the Marion County Small Claims Courts do not qualify. The Marion County Small Claims Courts are unique in the State of Indiana in terms of their structure and function. And Marion County, obviously, falls within the Indiana Judicial System. The Marion County Small Claims Courts are therefore not a common division in the state court system in which they exist. They are the exact opposite.

There is nothing vague about any of this. And yet the majority tosses the Newsom/Black's definition aside, claiming it is too vague and unfounded to be useful. Next, it immediately turns to the policy underlying the FDCPA to craft an entirely new definition with no source in any statute, case law, usage guide, dictionary, or any other conceivable source of authority on the meaning of the words, rather than the intent behind the words, in this statute. That is not how statutory construction works.

For one thing, casting aside the Newsom/Black's definition—if I were willing to go that far with the majority—does not mean there can be no “plain meaning.” It is not comprehensible to me to suggest that we have no idea what Congress means when it says the words “judicial district.” We know exactly what a “judicial district,” in the common parlance of the law, is. Our job is to hear appeals from seven of them. If we read the words “judicial district” to refer to judicial districts as they exist in the federal court system and as they are so frequently referenced throughout the United States Code,

the commonsense approach would be to understand the words that follow, “or similar legal entity,” to refer to the analogous level of division in a given state court system, however titled.

But the majority opinion never stops to consider the possibility that the words “judicial district” may just not be that great a puzzle. Why? Because “terms that seem plain and easy to apply to some situations can become ambiguous in other situations.” Op. at 639. That is an inadequate justification for a finding of ambiguity. “The plainness or ambiguity of statutory language is determined by reference to the language itself, the specific context in which that language is used, and the broader context of the statute as a whole.” Vulcan Const. Materials, LP v. Fed. Mine Safety and Health Review Comm'n, 700 F.3d 297 (7th Cir.2012) (citation omitted). Noticeably absent from that list is reference to the unusual facts of any particular case. There are good reasons for that. A law that is subject to change whenever hard facts come along is no law at all. Hard facts mean hard work is before us, not that the law is wrong.

Moreover, even if the statute were ambiguous, I do not see how the majority's chosen definition falls within the range of that ambiguity. Statutory ambiguity presents us with an opportunity to choose from among equally legally justified options. It is not a creative license. I do not see the majority presenting competing common understandings of what a “judicial district or similar legal entity” could be. I see the majority replacing the words “judicial district or similar legal entity” with an entirely new phrase, “smallest geographic area that is relevant for determining venue in the court system in which the case is filed,” because those new words produce the result most consistent with the purpose of the statute in this particular case. That may be an equitable outcome, but it is legally improper: “Ambiguity sometimes justifies resort to legislative history, but it is used to decipher the ambiguous language, not to replace it.” Peterson v. Somers Dublin Ltd., 729 F.3d 741, 749 (7th Cir.2013). The majority's reliance on legislative intent is inappropriate here, and it leads to some confounding results.

II.

I have made my concern with the majority's analytical process clear: I believe the majority too casually tosses aside principles of stare decisis and the canons of statutory construction to arrive at a result consistent with its understanding of the policy aims undergirding the FDCPA. Now I will briefly address why I believe the majority's result is just plain wrong. I believe it is wrong because it leads to bizarre and inconsistent results irreconcilable with the statutory text and the intent behind it, and because the outcome reached by the majority defeats its own stated purpose.

First, the majority's definition means that many federal judicial districts are not “judicial districts,” but divisions of those federal judicial districts are. Many federal judicial districts are broken up into smaller geographical units: divisions. A quick internet search confirms that at least some of those judicial districts' local rules require civil actions to be filed in the division of the district court in which proper venue lies. See, e.g., Northern and Southern Districts of Iowa Local Rule 3(b); District of Montana Local Rule 3.2(b); Western District of Virginia Local Rule 2(b). It cannot be meaningfully argued that the divisions are not “geographic areas” in the first place; their boundaries are tied to county lines.

In these judicial districts, the divisions are therefore the “smallest geographic area that is relevant for determining venue in the court system in which the case is filed.” The only possible way the majority could expand its definition of a “judicial district” to include the districts themselves is to resort to “details of court administration”—same judges, same administrative staff, same court seal, etc. But even where that works (and it will not work everywhere), it is exactly the approach the majority ostensibly eschews in abandoning Newsom. Op. at 638. That is a problem.

Second, the majority's definition leads to inconsistent results within the same judicial systems. Putting aside the problem that some federal districts are still “districts” while others are not, one need only look to Indiana to see how. Marion County is the only county in Indiana for which the small claims courts will constitute the relevant judicial district. The majority's definition of the operative phrase therefore means that a “judicial district or similar legal entity” in Marion County is something different than a “judicial district or similar legal entity” in Shelby County, Saint Joseph County, Allen County, or Jasper County. I can conceive of no possible justification for this outcome. How can we define a “judicial district”—something which by the nature of the words themselves must be a division of some larger entity—in a way that is not consistent with respect to that larger entity?

I understand that the majority outcome is a boon to Marion County residents who have trouble getting from one township to another to defend against a case. Perhaps this is especially true for the typical small claims defendant, who is likely a person of limited means. But at least public transportation exists in Marion County in a meaningful way. There are poor people in rural Indiana, too, and their county seats are often just as far away from them, if not farther, than the distance between the two Marion County townships at issue in this case

The majority's new definition does nothing for them, and the result has an air of arbitrariness about it. Even if I ride along with the majority to the point of considering legislative intent, am I to believe Congress intended the FDCPA to provide greater protection to debtors in Marion County than in the rest of the State of Indiana? I cannot accept as plausible any reading of the words “judicial district” that varies so wildly from one place within the same court system to another.

That leads me to my greatest concern. The majority opinion is sprinkled throughout with language suggesting that its goal is to establish a rule honoring Congress's intent to impose consistent federal limitations on debt collectors' choice of venue when filing in state courts. See, e.g., Op. at 640 (“The presence of the venue provision in the Act shows congressional dissatisfaction with allowing state law to determine where suits to collect consumer debts can be filed.”).

These remarks are often presented as a foil to Judge Flaum's preferred approach: defining a “judicial district,” with respect to a state court system, as that unit into which the state consistently divides itself. But if enacting a federal rule above the reach of meddling, incompetent, or insufficiently protectionist state governments was the majority's intent, it has failed. The majority rule depends entirely on state venue rules, which the state courts or legislatures are free to change on a whim. Ultimately, therefore, the majority reaches a result that frustrates its own purpose; this rule is no “rule” at all.

Finally, none of this is meant to suggest that forum-shopping by debt collectors in Marion County is not a problem, or that nothing should be done to fix it. It is a problem, and Indiana is already taking steps to fix it. But the majority disregarded the principle of stare decisis without even paying lip service to any of the established justifications for doing so, then proceeded to shortchange the first step of the 664statutory interpretation process in its effort to achieve a result consistent with its understanding of the policy aims of the FDCPA. In so doing, the majority reached a result which defeats its own purpose and falls apart with any attempt at general application to the various court systems within our territorial jurisdiction. I cannot sign on to that. While I am of the opinion that the judicial circuits themselves, and not any specific court within them, are the relevant “judicial districts” in Indiana, I find Judge Flaum's reasoned approach and his adherence to our Newsom decision much more persuasive than that put forward by the majority. I therefore join in his dissent.