with whom BAUER, FLAUM, and KANNE, Circuit Judges, join, dissenting.
Today the court announces an unprecedented new rule—one that punishes debt collectors for doing exactly what the controlling law explicitly authorizes them to do at the time they do it. The court’s inverted new standard effectively eradicates the FDCPA’s bona fide error defense. Worse still, by penalizing strict compliance with controlling precedent, the court gravely undermines the rule of law in this circuit and exposes law-abiding citizens to unforeseeable and arbitrary liability in civil proceedings. I respectfully dissent.
I.
In December 2013, debt collector Blatt Hasenmiller Leibsker & Moore, LLC, filed a collection lawsuit against Ronald Oliva. The suit was filed at the Richard J. Daley Center in downtown Chicago, in the first municipal district of the Circuit Court of Cook County. Oliva resided in Orland Park, Illinois, which is in the fifth municipal district of the Circuit Court of Cook County.
In deciding where to file suit, Blatt relied on our then-controlling precedent of Newsom v. Friedman, 76 F.3d 813 (7th Cir. 1996), In Newsom we concluded that the Circuit Court of Cook County is a single “judicial district” for purposes of the FDCPA’s venue provision, 15 U.S.C. § 1692i, which requires collection suits to be filed in the “judicial district or similar legal entity” where the contract was signed or where the debtor resides. The Newsom court unequivocally held that, under § 1692i(a)(2), debt collectors were allowed to file suit in any of the Circuit Court of Cook County’s various municipal districts so long as the debtor resided in Cook County or signed the underlying contract there. Oliva resided in Cook County, so Blatt’s decision to file suit in Cook County’s first municipal district was lawful under § 1692i as definitively interpreted by Newsom. In other words, Blatt’s selection of venue was lawful when it occurred.
No one disputes this. And who could? Newsom was the settled law of this circuit for nearly eighteen years at the time, and it explicitly authorized Blatt to file suit exactly where it did. Even the court admits that “[ujnder our decision in Newsom, Blatt Hasenmiller’s choice of venue ... was permissible.” Nor is there any doubt that Newsom was valid while in effect: as we recently recognized, Newsom was “good law in this circuit” before it was overruled. Jackson v. Blitt & Gaines, P.C., 833 F.3d 860, 865 (7th Cir. 2016).
In July 2014, while Blatt’s lawsuit was still pending, a divided en banc panel of this court overruled Newsom in Suesz v. Med-1 Solutions, LLC, 757 F.3d 636 (7th Cir. 2014). The Suesz court held that a “judicial district or similar legal entity” under § 1692i is “the smallest geographic area that is relevant for determining venue in the court system in which the case is filed.” Id. at 638. Under Suesz, collection lawsuits governed by § 16921(a)(2) may no longer be filed in any of the six municipal districts of the Circuit Court of Cook County, but must instead be filed in the particular municipal district where the debtor resides or where the underlying contract was signed. The court in Suesz further held that its decision applied retroactively.
*502Although Blatt’s selection of venue complied with the then-governing law of New-som, it did not conform to the new rule retroactively applied by Suesz. To comply with the new rule, Blatt dismissed its action against Oliva shortly after Suesz was decided. Oliva then brought an FDCPA claim against Blatt in federal court, alleging that Blatt was retroactively liable for having filed suit in the wrong venue under Suesz. The district court granted summary judgment for Blatt on the ground that Blatt was protected from liability under the FDCPA’s bona fide error defense. 15 U.S.C. § 1692k(c). That conclusion was correct, and the district court’s judgment should be affirmed.
II.
The FDCPA is not a strict-liability statute. While most infractions result in liability, the Act creates an important exemption for violations resulting from a debt collector’s good-faith mistake. The exemption is mandatory, not optional:
A debt collector may not be held'liable in any action brought under this sub-chapter if the debt collector shows by a preponderance of evidence that the violation was not intentional and resulted from a bona fide error notwithstanding the maintenance of procedures reasonably adapted to avoid any such error.
15 U.S.C. § 1692k(c) (emphasis added).
Blatt easily meets each of these elements. It’s undisputed that Blatt’s retroactively imposed violation of Suesz was unintentional and that Blatt maintained reasonable procedures to avoid any error. It’s also undisputed that the retroactive violation resulted from Blatt’s good-faith mistake: the mistake of complying with the controlling law of Newsom rather than the then-nonexistent rule of Suesz that would later be retroactively applied. Indeed, the court concedes that if ever there was a good-faith mistake under the FDCPA, “it was in cases like this.” Maj. Op. at 499-500. Yet the court refuses to draw the only logical conclusion: Blatt is statutorily exempt from liability because its retroactively imposed violation of the rule announced in Suesz was the result of a bona fide error under § 1692k(c).
The court’s refusal to give effect to the statute is not justified by its retroactivity ruling in Suesz. Suesz had nothing to do with the bona fide error defense and made no mention of good-faith mistakes or § 1692k(c). As the original panel in this case stated, Suesz may have created a cause of action for retroactive violations, but it did not “retroactively proscribe the application of the bona fide error defense.” Oliva v. Blatt, Hasenmiller, Leibsker & Moore, LLC, 825 F.3d 788, 791 (7th Cir. 2016).
Nor is today’s decision supported by the Supreme Court’s holding in Jerman v. Carlisle, 559 U.S. 573, 130 S.Ct. 1605, 176 L.Ed.2d 519 (2010). The Jerman Court held that “the bona fide error defense in § 1692k(c) does not apply to a violation of the FDCPA resulting from a debt collector’s incorrect interpretation of the requirements of that statute.” Id. at 604-05, 130 S.Ct. 1605. Blatt did not incorrectly interpret the FDCPA’s legal requirements in failing to follow the then-nonexistent rule of Suesz. Quite the contrary. While it was still the law, Newsom definitively determined the legal requirements of the FDCPA’s venue provision for debt collectors filing suit in Cook County—and Blatt followed Newsom to the letter. Blatt therefore correctly interpreted the statute’s legal requirements in accordance with the controlling law in effect at the time of its conduct.
Blatt’s interpretation was no less correct just because this court later found that *503Newsom was wrongly decided. A court’s controlling but mistaken interpretation of a statute is not the same thing as a party’s decision to comply with the court’s controlling interpretation. The party may, indeed must, adhere to the court’s controlling interpretation even if the party itself correctly interprets the statute and believes that the court’s interpretation is wrong. Thus, assuming that the Newsom court mistakenly interpreted the statute, its mistaken interpretation cannot be imputed to Blatt. See Oliva, 825 F.3d at 792 (noting that Blatt’s decision to follow the controlling rule of Newsom did “not amount to an independent (and entirely futile) ‘interpretation’ of that which Newsom had already definitively interpreted and handed down as the binding law of this Circuit”) (footnote omitted). Jerman excludes the bona fide error defense only when the debt collector’s violation results from the “debt collector’s mistaken interpretation” of the legal requirements of the Act. Jerman, 559 U.S. at 577, 130 S.Ct. 1605 (emphasis added). Because Blatt did not mistakenly interpret the statute’s legal requirements merely by following the controlling interpretation of Newsom, Jerman’& exclusion does not apply.
Put simply, Blatt’s decision to follow the controlling law as it then existed was not a mistake of law, nor was its failure to comply with a rule that did not yet exist but that would one day emerge with retroactive effect. No subsequent change of law— not even a retroactive one—can change that fact. It is not a mistake of law to follow controlling law, even when that law is later overruled.
In sum, Blatt’s retroactively imposed violation was the unfortunate result of its good-faith decision to comply with controlling law while ignorant of a future ruling that would one day reach back in time to declare its lawful conduct unlawful by the legal fiction of retroactivity. In hindsight, that decision was mistaken. But it wasn’t a mistake of law. Blatt’s failure to foresee the retroactive change of law heralded by Suesz was a bona fide error entirely outside Blatt’s control. Blatt is therefore entitled to exemption from liability under § 1692k(c).
III.
In reaching the opposite conclusion, the court makes a number of serious mistakes. It repeatedly misrepresents the original panel’s decision; misinterprets both Jer-man and Suesz; disregards the legal effect of its own binding precedent; runs afoul of the separation of powers by effectively nullifying a lawful statutory defense legislated by Congress; and creates an unprecedented new rule—one that punishes people for following controlling law—that tramples the most cherished principles of due process and strikes at the very heart of American liberty. The most pervasive of these errors is the misrepresentation of the original panel’s decision, so I’ll start there.
A. The Court Misrepresents the Panel’s Decision
The court paints a very different—and very inaccurate—picture of the original panel’s decision. The court begins by asserting that the panel “declined to apply the Suesz holding on retroactivity.” Maj. Op. at 497; see also id. at 495. Not so. The panel explicitly assumed that Suesz’s ret-roactivity did apply in this case—applied it—and then concluded that the bona fide error defense excused Blatt from liability for its retroactive violation:
Although Suesz did not specify the scope of its retroactivity, we assume without deciding that Suesz’s holding applies retroactively to Blatt, and that Blatt’s decision to file suit in the first municipal district of the Circuit Court of Cook *504County was a violation of § 1692i as interpreted by Suesz.... We [ ] hold that Blatt’s violation of § 1692i as interpreted by Srnsz was the result of a bona fide error that precludes liability under the Act.
Oliva, 825 F.3d at 790-92.
The court then launches a litany of mis-characterizations pertaining to the original panel’s interpretation of Jerman, According to the court, the panel concluded that Jerman does “not extend to mistakes of law based on controlling circuit precedent.” Maj. Op. at 495. What the original panel actually said is that Jerman precludes the bona fide error defense based on a debt collector’s mistake of law, but that a debt collector does not make a mistake of law when it correctly interprets controlling circuit precedent. See Oliva, 825 F.3d at 792 (reciting the holding of Jerman verbatim and holding that Blatt’s retroactive violation of Suesz was “wot‘the result of Blatt’s mistaken interpretation of the FDCPA”; Blatt’s “interpretation was undisputedly correct, since [Blatt] relied on Newsom to file suit exactly where Newsom allowed”) (emphasis added).
The court further suggests that the panel read Jerman “narrowly ... to allow § 1692k(c) to protect some mistakes of law about the Act but not others,” Maj, Op. at 498. But that reading isn’t just narrow, it’s dead wrong. And the original panel never said anything of the sort. See Oliva, 825 F.3d at 792. The court.then poses a.line-drawing problem to critique the “narrow” reading that is of its own making: “nor do we see a workable line between protected and unprotected mistakes of law.” Nor do I. Though I do see a line—-a bright one in fact—between those violations that result from a debt collector’s mistake of law, and those that do not. That’s the line drawn by Jerman, and it’s the line the original panel applied in this case.
According to the court, the panel also concluded that Jerman does not bar the bona fide error defense when a debt'collector violates the Act in reliance on “a precedent that was later overruled as mistaken.” Maj. Op. at 498 (emphasis added). In other words, the court continues, the panel found that the defense still applies to mistakes of law so long as the debt collector relies “on a court’s, or at least an appellate court’s, mistaken interpretation of the Act.” Id, (emphasis added). The panel never made those sweeping' statements, which are clearly at odds with Jerman, Relying on “a” precedent or “a” judicial decision may very well be a mistake of law—as when the precedent or decision relied upon conflicts with the controlling law in the relevant jurisdiction. That’s exactly what happened in Jerman\ the debt-collector mistakenly interpreted the controlling law of the Sixth Circuit by relying on non-controlling precedent from another circuit. By contrast, correctly interpreting the controlling law of the relevant jurisdiction is not a mistake of law in that jurisdiction. It’s precisely the opposite. The panel didn’t say that there’s no mistake of law when the debt collector relies on “an appellate court’s’’ conflicting precedent; the panel: said that correctly interpreting the controlling appellate precedent in.the relevant jurisdiction is not a mistake of law.
The court goes on to say, as if the original panel differed, that there are “clear signs in Jerman that the Court was reaching all mistaken interpretations of the Act, regardless of how understandable or. reasonable they might have been.” Maj. Op. at 498, That rather obvious statement might be relevant if the original panel had grossly misinterpreted Jerman to mean that, while good-faith mistaken interpretations aren’t protected, really good-faith *505mistaken interpretations are. There's more than a clear sign that Jerman doesn’t support so flimsy a distinction. And the panel never said it did. The panel didn’t say that Blatt’s mistake of law should be excused because Blatt understandably relied on controlling law. The panel said that because Blatt understandably relied on controlling law, there was no mistake of law to be excused. See Oliva, 825 F.3d at 792 (“Blatt’s interpretation was not mistaken when it was made,”).
■ Notably, the- court does not quote from the original panel’s decision to substantiate any of these mischaracterizations. Nor could it, because the original panel never said the things the court says it said. Today’s reversal knocks down a straw man. The original panel’s real decision concluded that Blatt’s good-faith mistake was not a mistake- of law because Blatt correctly interpreted- (and strictly followed) controlling precedent. That reasoning is not addressed in today’s opinion.
B. The Court Misinterprets Jerman
The court also misreads the Supreme Court’s holding in Jerman. Under Jerman, the court says, Blatt “cannot avoid liability for a violation based on its reliance on circuit precedent....” Maj. Op. at 494. But that’s not what Jerman says. Jerman says that debt collectors cannot avoid liability based on their “mistaken interpretation of the legal requirements of the FDCPA.” Jerman, 559 U.S. at 577, 130 S.Ct. 1605. In Jerman, the debt collector’s reliance on circuit precedent didn’t matter because the precedent was from another circuit and wasn’t controlling. See Daubert v. NRA Grp., LLC, 861 F.3d 382, 394-95 (3d Cir. 2017) (holding that Jerman applies “[wjhere an issue'of law under the FDCPA is unsettled by the Supreme Court or a precedential decision of the relevant court of appeals”) (emphasis added). But what happens when the circuit precedent being relied on is also the controlling determination of the FDCPA’s legal requirements in the relevant jurisdiction, and the debt collector’s interpretation of those requirements is correct? That’s what happened here. When reliance on the law equals compliance with the law, there is no mistake of law. Jerman doesn’t say otherwise.
The court then notes that the Jerman Court rejected a rule that would make an exception for mistakes of law based on a debt collector’s good-faith reliance on “substantial legal authority.” Maj. Op. at 498-99. From there, the court contrasts “substantial” legal authority with “controlling” legal authority, and indicates that it doesn’t see a manageable' way to distinguish reliance on one from reliance on the other. Id. at 498-99. Whatever “substantial” authority means in the court’s analysis, it can’t mean controlling authority, for the court contrasts the two. And if it’s not controlling, then it’s at best persuasive. So what the court is actually saying is that it sees no way to differentiate controlling legal authority from persuasive legal authority.
That’s an astonishing proposition, and certainly not one recognized by the court in Jerman. If controlling authority—the only definitive authority that binds with the force of law—can’t be readily distinguished from its opposite, then the rule of law is a. sham. Here’s the difference between the two: controlling authority is the governing, binding law in a given jurisdiction; persuasive authority is not, And here’s the upshot: one can correctly interpret persuasive authority and still be mistaken about the relevant controlling authority; but one-can’t be mistaken about the relevant controlling authority when he correctly interprets the relevant controlling authority. That’s what Blatt did here.
*506The court also says that Jerman puts the “risk of legal uncertainty on debt collectors.” Maj. Op. at 499. That’s true. But it begs an important question: why is the court punishing Blatt for doing what was certainly lawful at the time and place it did it? It seems the court believes that Jerman’s idea of risk allocation is to arbitrarily penalize law-abiding debt collectors by saddling them with the consequences of a risk they never assumed.1
In essence, the court today reads Jer-man as excluding the bona fide error defense not only when the debt collector’s violation results from its mistaken legal interpretation, but also when it does not. That’s not just an expansive reading of Jerman. It’s a new rule altogether.
C. The Court Misinterprets Suesz
Suesz said nothing—not one word— about the bona fide error defense. Not in the main opinion, not in the concurrence, not in the dissents, not even in the footnotes. It is truly remarkable, then, that the court asserts that the question presented in this appeal—whether the bona fide error defense applies to Blatt—was already “decided” in Suesz. Maj. Op. at 494. Suesz makes no reference to § 1692k(c) or the Supreme Court’s related decision in Jer-man.2 Even a broad search for more generalized terms that might bear some remote connection to today’s issue—terms like mistake, error, good faith, bona fide, or unintentional—comes up dry. The Suesz court redetermined the meaning of “judicial district” under § 1692i and applied its ruling retroactively. That’s all. Suesz presents no obstacle to the mandatory application of the bona fide error defense in today’s case.
So then why does the court insist that Suesz controls? The confusion stems from the court’s conflation of two distinct questions: First, does Blatt’s challenged conduct constitute a retroactive violation? And second, does that retroactive violation result in liability? Suesz resolved only the first of these questions: Blatt’s challenged conduct did not conform to Suesz’s new rule, so the conduct is deemed a retroactive violation under Suesz. The court said nothing about the second question, the resolution of which depends on whether a statutory exemption—such as the bona fide error defense—applies.
According to the court, however, Suesz’s retroactivity also means that Blatt must be held liable for its retroactive violation. After answering (in the negative) the question whether Blatt “can avoid liability” un*507der the bona fide error defense, the court states that it already “decided this question in Suesz when [it] overruled [Newsom] and declined ... to make that ruling effective only prospectively.” Maj. Op. at 494.3 In the court’s view, then, the mere fact of Blatt’s retroactive violation under Suesz precludes the application of the bona fide error defense.
But if the bona fide error defense applies only when there is no violation, then what good is it? The whole point of the defense is to excuse liability precisely when there is a violation—not when there’s no violation and no excuse is needed. By holding that the bare fact of a violation precludes the bona fide error defense, the court ensures that the defense will never apply, for the defense -presumes a violation, and cannot apply without one. Thus, far from precluding the bona fide error defense, Suesz’s retroactivity is the essential prerequisite for the defense’s availability.
The court’s misreading of Suesz is also inconsistent with longstanding principles of retroactivity well outside the context of the FDCPA. The mere fact that a law is retroactively applied has never meant that every retroactively assessed violation is guaranteed a remedy. See, e.g., Davis v. United States, 564 U.S. 229, 243, 131 S.Ct. 2419, 180 L.Ed.2d 285 (2011) (noting that retroactivity and remedy are “separate, analytically distinct issue[s],” and that “the Court has never equated its retroactivity principles with remedial principles”) (citations and internal marks omitted). As the original panel explained, Suesz may create a cause of action for retroactive violations, but it doesn’t guarantee a remedy (in the form of liability for damages) by “retroactively proscrib[ing] the application of the bona fide error defense.” Oliva, 825 F.3d at 791.4
To hold otherwise, as the court does today, is to stretch beyond its breaking point the judicial doctrine of retroactivity. This court has the power to make its rulings retroactive, but it simply cannot, by a mere flick of the judicial wand, retroactively alter objective realities of the past. Try as it might, the court can’t change the historical fact that Blatt correctly interpreted the governing law in effect at the time it filed suit.
D. The Court Disregards the Effect of its Own Controlling Precedent
The court says that Blatt “violated the venue provision of the [FDCPA],” even though Blatt’s “choice of venue was permissible” under Newsom’s then-controlling interpretation of that same provision. Maj. Op. at 494-95. In doing so, the court draws an implicit functional distinction between two “different” binding legal precepts: the FDCPA’s venue provision interpreted by Newsom, and Newsom’s controlling interpretation of the FDCPA’s venue provision. This distinction enables the court to draw an odd conclusion: Blatt correctly interpreted (and did not violate) Newsom’s controlling determination of the legal requirements of § 1692i, but incorrectly interpreted (and violated) § 1692i itself. As applied in this case, the distinction doesn’t hold up.
*508While in effect, Newsom did not propose a suggested legal interpretation of questionable binding value—Newsom’s interpretation, erroneous or not, was the controlling law in this circuit. Blatt’s conduct was thus expressly authorized, not just by Newsom as distinct from § 1692i, but also by § 1692i as interpreted by Newsom. This brings us to a fairly obvious but crucial point: before it was overruled, Newsom definitively determined the practical legal requirements of § 1692i as it applied to lawsuits filed in the Circuit Courts of Illinois. Indeed, the court concedes as much when it says that the “practical effect of [Newsom] was to allow debt collectors in Cook County to choose freely among the six different municipal districts, at least as far as the FDCPA was concerned.” Maj. Op. at 496 (emphasis added). Exactly. Under Newsom—and as far as the statute itself was concerned—Blatt’s conduct was lawful when it occurred.
Similarly, the practical legal scope of § 1692i in this circuit is now determined by our controlling interpretation in Suesz. And the judicial determination of that scope is not only proper but necessary. Section 1692i says only that the debt collector must file suit in a particular judicial district. It doesn’t define judicial district, much less determine whether a particular geographic area in the United States fits within the definition. Nowhere does the FDCPA use the controlling language of Suesz to describe a judicial district. The statute does not state that a “judicial district or similar legal entity” is “the smallest geographic area relevant to venue in the court system in which the case is filed.” Yet right now, in this circuit, § 1692i means just that.5 Lawyers might say that the smallest-geographic-area requirement is mandated by the statute, or that it is mandated by Suesz. Practically speaking, either statement would be correct. While Suesz remains in effect, there is no functional difference in this circuit between the requirements of § 1692i by itself and the requirements of § 1692i as definitively interpreted by the controlling authority of Suesz. And the same was true under Newsom. In short, § 1692i sets forth the law, but it doesn’t say what the law means in the Seventh Circuit. That’s our job (and of course that of the Supreme Court). And as we do our job, the law of this circuit is shaped accordingly.
This is not to say that a statute is identical to the controlling decision that interprets it. There are of course important distinctions, the most obvious of which comes into play when the controlling precedent is overruled based on a renewed examination of its relationship to the text of the statute itself. Nor do I say, as the court claims, that judicial decisions, rather than the statutes they interpret, are “the law” in some theoretical, unqualified sense. See Maj. Op. at 499-500. Of course “the statute” is “the law.” But it’s equally obvious that a statute’s controlling legal effect is determined,, to á greater or lesser extent, by the relevant controlling interpretation of the Supreme Court and the federal courts of appeals.6 Thus, while the *509statute itself remains the same, its practical legal requirements can vary—across space in different circuits, or across time in the same circuit. That is precisely what happened here, when this court replaced the old rule of Newsom with the new rule of jSuesz.
Even the estimable Judge Bork, one of history’s strongest proponents of judicial restraint, freely acknowledged that judges determine law in this way. Writing about the federal judiciary, Judge, Bork candidly states:
It is of course true that judges to some extent-must make law every time they decide a case, but. it is a minor, interstitial lawmaking. The ratifíers of the Constitution put in place the walls, roofs, and beams; judges preserve the major architectural features, adding only filigree.”
Robert H. Boric, The Tempting of America: The Political Seduction of the Law, 5 (1990). Of course, the judge’s power to make law in this way is sharply limited by the separation of powers. The court today crosses that limit by effectively rewriting an Act of Congress and nullifying the bona fide error defense in a case that demonstrates the epitome of a good-faith mistake. Ironically, the court pretends not to know the power of its own controlling precedent in Newsom, even as it assumes a power it does not have.
Perhaps federal appellate judges don’t typically talk about their authority to shape and develop the binding practical effect of federal statutory law in their respective circuits. But in light of today’s extraordinary decision, there’s no way around the topic. Besides, I’m not breaking'any new ground here. I’m just saying what everyone already knows (but what the court won’t acknowledge): before Suesz, there was .one, law in this circuit; after Suesz, another. Suesz wasn’t a declaration that Newsom’s determination of the FDCPA’s venue provision was void ab ini-tio-, it was the binding redetermination— right or wrong—of that provision’s legal effect in the Seventh Circuit.
' The court’s argument to the contrary is self-defeating. The court states that the “controlling, law” is strictly “the statute itself,” as opposed to the definitive interpretation of the statute by the Supreme Court or the .relevant federal court of appeals. As applied to our case, this means that only the FDCPA itself—not Suesz’s interpretation of the FDCPA—controls with the force of law. Yet the court today also authoritatively concludes that Blatt violated the FDCPA as interpreted by Suesz—a conclusion that has controlling effect only if Suesz’s interpretation is controlling. (Only controlling law is binding, so if the court’s decision in Suesz is not controlling law, Blatt is not bound by the Suesz court’s interpretation, and may lawfully rely on its own private interpretation to justify its conduct.) Thus, in concluding that Blatt violated controlling law because it failed to follow the statute as interpreted by Suesz, the court necessarily relies on the very premise it denies: that Suesz’s interpretation. controls with the force of law. By denying the controlling effect of its own legal determinations, the court pulls the rug out from under its own feet.7
*510So let’s put to rest the court’s notion that Blatt somehow correctly interpreted Newsom’s definitive determination of the FDCPA’s venue provision (one law) while incorrectly interpreting the FDCPA’s venue provision itself (another law). For purposes of determining whether Blatt correctly interpreted controlling law, the two are treated as one and the same. As the court notes, “as far as the FDCPA was concerned,” the “practical effect” of New-som was to allow Blatt to file suit exactly where it did. Maj. Op. at 496 (emphasis added). Law is a practical reality. And for practical purposes, Newsom’s interpretation of § 1692i was the controlling law when Blatt filed suit. In this circuit, there was no other.
E. The Court Effectively Nullifies a Mandatory Statutory Defense
The court acknowledges that Blatt relied on Newsom’s controlling interpretation of § 1692i in good faith. Indeed, the court concedes that if ever a mistake was made in good faith, “it was in cases like this.” So if the bona fide error defense does not apply even here, where the debt collector’s mistake was the epitome of an unintentional good-faith oversight that was not also a mistake of law, then when does the defense ever apply? The answer is never—or at least that it never has to. Even if the court applies the defense in the future, under today’s precedent, it won’t have to. Then again, under today’s precedent, even the rule announced today is of questionable value, and law-abiding debt collectors would do well to keep in mind that they follow it at their own risk.
F. The Court Creates an Unprecedented New Rule
The court suggests that to apply the bona fide error defense in this case would be to create an unusual new rule that doesn’t belong in the FDCPA. Such a rule, the court says, would be similar to a far-removed exception to the exclusionary rule under the Fourth Amendment as interpreted by the Supreme Court in Davis v. United States, 564 U.S. 229, 131 S.Ct. 2419, 180 L.Ed.2d 285 (2011). Maj. Op. at 500-01. But a comparison with Davis shows just the opposite. It is the rejection, not the application, of the bona fide error defense in this case that results in a most unusual new rule.
In Davis the Court held that the police should not be sanctioned (through the suppression of evidence under the exclusionary rule) when they obtain evidence in objectively reasonable reliance on binding appellate precedent. Perhaps that rule is “unusual,” as this court now says, in the Fourth Amendment context. But if so, it’s only because unlawfully obtained evidence is generally supposed to be suppressed, making the admission of such evidence an exception to the norm. There’s nothing unusual, however, about recognizing a similar rule here: that debt collectors should not be sanctioned (through the imposition of damages) when they rely on binding appellate precedent to collect a debt. Indeed, could any rule be more commonplace (or more fundamental) than the rule that a party in a civil suit should not be subjected to liability for following controlling law? The panel didn’t create that rule. That rule is the norm—or at least it was.8
*511The same cannot be said of the new rule that the court creates today. Under today’s rule, a party in a civil action may be penalized with statutory damages for strictly adhering to controlling law—even when the statute in question explicitly provides a defense for good-faith mistakes. Davis does not support that rule, nor do basic principles of due process permit it. What the court today calls the “equitable points” of Blatt’s reliance on controlling law, Maj. Op. at 500-01, are in fact the most rudimentary requirements of constitutional due process.
“A fundamental principle in our legal system is that laws which regulate persons or entities must give fair notice of conduct that is forbidden or required.” F.C.C. v. Fox Television Stations, Inc., 567 U.S. 239, 253, 132 S.Ct. 2307, 183 L.Ed.2d 234 (2012). “This requirement of clarity in regulation is essential to the protections provided by the Due Process Clause of the Fifth Amendment.” Id. See also Landgraf v. USI Film Prods., 511 U.S. 244, 265, 114 S.Ct. 1483, 128 L.Ed.2d 229 (1994) (“Elementary considerations of fairness dictate that individuals should have an opportunity to know what the law is and to conform their conduct accordingly!)]”). Blatt had no fair notice that its selection of venue in Cook County would later be retroactively forbidden. Blatt’s selection of venue was expressly authorized under the controlling law in effect at the time, and no reasonable debt collector in Blatt’s position could possibly have known that it would one day be held liable ex post facto for strictly following controlling law.
I’m not sure why the court is bent on punishing debt collectors for following the law. Is the intention to put debt collectors out of business? To allow debtors to refuse to pay their debts with impunity? I can’t think of a rule better suited to those ends than the rule the court announces today.
Today’s decision also gravely undermines the rule of law by discouraging debt collectors from following this court’s controlling precedent. Indeed, the court leaves open the possibility that debt collectors may even be subject to liability for engaging in conduct that controlling precedent not only permits, but mandates. The court notes that Newsom allowed, but did not require, Blatt to file suit where it did. Yet nowhere does the court reassure us that Blatt would not be liable if Newsom had ruled the other way round. Intentional or not, here’s the message today’s ruling sends to debt collectors: Think twice before following the controlling law of this circuit. For tomorrow we may change our mind. And you may wish you hadn’t.
Today, in an almost surreal inversion of law and logic, the court punishes Blatt for doing exactly what the controlling law explicitly authorized Blatt to do at the time it did it. It does so through a fantastical expansion of the (previously) confined judicial doctrine of retroactivity, and in spite of a statutorily mandated bona fide error defense. The court tries to soften the blow by mildly suggesting that Blatt’s punishment may be mitigated because it acted in good faith. Small comfort to Blatt. Blatt is being punished for dutifully adhering to controlling law notwithstanding its legal entitlement to a statutory defense. A mere *512reduction in punishment does nothing to right that wrong;
Not long ago, this court' recognized that “[t]he FDCPA was created to prevent abusive debt-collection practices, not'to prevent law-abiding creditors from collecting oh legally enforceable debts.” Jackson, 833 F.3d at 866 (emphasis added). Sadly, after today’s decision, that is no longer true.
XV„
In Franz Kafka’s The Trial, Joseph K. was tried and punished for breaking a law he knew nothing about. At no point during the proceedings, or even at the time of his punishment, did the court ever tell poor Joseph precisely what law he had violated. Today this court does Kafka one better. For today the court punishes the defendant, not for breaking a law that was never given, but for following the very law we gave it. That is not only inconsistent' with the FDCPA’s bona fide error defense; it. is inconsistent with the judicial function and the rule of law.
I dissent.
. Alternatively, if the court believes that law-abiding debt collectors do assume the risk of being penalized for strictly following controlling law, then what of debt collectors who choose not to follow controlling law? Don’t they assume the same risk? After today’s decision, even strict compliance with Suesz may one day result in ex post facto liability: and when it does, the bona fide error defense won't be there to help.
. The court notes that the debt collector in Suesz may have chosen not to rely on the bona fide error defense out of concern that the defense would be barred by Jerman. Maj. Op. at 497 & n.3. Maybe so. But that's because, unlike here, the debt collector’s conduct in Suesz wasn’t authorized by Newsom. As the original panel noted, the debt collector in Suesz filed suit in Marion County, Indiana, whereas the holding of Newsom was plainly “limited to the Circuit Courts of Illinois, and did not extend to the filing of lawsuits in other states.” Oliva, 825 F.3d at 792 n.2. See also Newsom, 76 F.3d at 818-20 (holding only that the Circuit Courts of Illinois, including the Circuit Court of Cook County, are individual judicial districts under § 1692i), Because the scope of Newsom's controlling authority was limited to lawsuits filed in Illinois, the debt collector in Suesz could have relied on Newsom (as persuasive authority) while still mistakenly interpreting the legal requirements of § 1692i as applied to the courts of Marion County.
. Obviously if the ruling were only prospective there would be no backdated violation and hence no need for the bona fide error defense in this case.
. Along the same lines, the bona fide error defense would still apply in this case if the Supreme Court; rather than this court, had overruled Newsom in Suesz. If the Court applied the new rule retroactively, it would establish only a "potential ground for relief,” see Davis, 564 U.S. at 243, 131 S.Ct. 2419, not an absolute decree of liability in spite of an otherwise available statutory defense.
. This would be the case even if Suesz was wrongly decided. A controlling interpretation is no less controlling simply because it’s wrong.
. That’s why we call binding appellate statutory interpretations "good law” that "controls" the outcome of future cases in the relevant jurisdiction. See Gacho v. Butler, 792 F.3d 732, 737 (7th Cir. 2015) (“Moore remains good law and is controlling here[.]”); see also Shalala v. Schaefer, 509 U.S. 292, 300, 113 S.Ct. 2625, 125 L.Ed.2d 239 (1993) (“Hudson remains good law as applied to remands ordered pursuant to sentence six.”). It’s also why, when overruling a previously controlling decision that mistakenly interpreted a statute or the constitution, we say that the prior decision is "no longer” the law—not *509that it never was the law. See, e.g., Mitchell v. Helms, 530 U.S. 793, 808, 120 S.Ct. 2530, 147 L.Ed.2d 660 (2000) ("Meek and Wolman are anomalies in our case law.’ We thérefore conclude that they are no longer good law.”); Pearson v. Helman, 103 F.3d 133 (7th Cir. 1996) ("Although Alvarez is no longer good law with respect to § 4B1.1 enhancements ,,., ”); see also Jackson, 833 F.3d at 865 (Newsom was "good law in this circuit” before it was overruled).
. The court’s position also leads to untenable results. If it’s true, as the court says, that Suesz’s interpretation of the statute is correct *510and that only the "statute itself" controls, then the statute would require noncompliance with a decision from the Supreme Court if the Court overruled Suesz and mistakenly replaced it with a conflicting rule. Opening the door to private interpretation of the FDCPA in this way is particularly worrisome because, as the court recognizes, many interpretations are possible. See Maj. Op. at 499 ("[TJhere is so much room to argue different interpretations of the FDCPA.”).
. Davis is instructive on another point, too. In deciding not to apply the exclusionary rule, the Court reasoned that to sanction the police even though they strictly complied with then-binding appellate precedent (which, as here, was later overruled) would be to convert the exclusionary rule into "a strict-liabilily regime.” Id. at 240, 131 S.Ct. 2419. Today, by punishing debt collectors under those exact same circumstances, this court effectively converts the FDCPA into a strict-liability statute.