Electronics for Imaging, Inc. ("EFI") appeals from the decision of the district court granting a motion to dismiss its lawsuit against Jan R. Coyle and Kolbet Labs under the Declaratory Judgment Act ("the Act"). Elecs. for Imaging, Inc. v. Coyle, No. C 01-4853 (N.D.Cal. Feb. 13, 2003) ("Dismissal Order"). Because the district court erred as a matter of law and hence abused its discretion, we reverse and remand.

EFI is a company that specializes in network printing solutions, particularly print controllers. It sells its products to companies such as Canon, Hewlett-Packard, and Xerox, who in turn incorporate EFI's technology into their own print servers, printers, and copiers. EFI is a Delaware corporation with its principal place of business in Foster City, California. Jan Coyle, a Nevada resident, is a listed inventor on U.S. Patents 6,337,746 and 6,618,157, both directed to printing technology.1 Kolbet Labs is a Nevada sole proprietorship owned by Daniel Kolbet, who is also a Nevada resident and an inventor listed on both patents.

Coyle first contacted EFI in September 1997, offering to license his technology, which was still in development and not yet the subject of an issued U.S. patent. Subsequently, Coyle and EFI met in Nevada under a non-disclosure agreement. EFI left that meeting uninterested in Coyle's work and did not retain any written information from it. It was not until 1999 that the two parties convened again, apparently at the request of EFI. According to EFI, its inquiry to Coyle was made in light of Coyle's statements concerning progress on the technology and a pending patent application; Coyle also mentioned his own history of filing patent infringement lawsuits against such corporations as Atari, Nintendo, Sega, and NEC Technologies. (Decl. of James L. Etheridge ¶ 2.) In April 2000, the two parties met under a new non-disclosure agreement to discuss possible licensing arrangements, but those talks also ended without any agreement.

In June 2001, Coyle discovered certain EFI sales and marketing information that convinced him that EFI was manufacturing products that were within the scope of Coyle's patent application, then still pending. In September 2001, Coyle notified EFI that the United States Patent and Trademark Office would soon issue Coyle's patent, and he asserted that his patent would cover all of EFI's print controllers. EFI claims that during the week of November 26, 2001, Coyle began to pressure EFI on an almost daily basis, threatening to drive EFI out of business. EFI alleges that Coyle repeatedly warned that he would bring multiple lawsuits, stating that "we'll sue all of your customers" and "bad things are going to happen." (Compl. ¶ 18.) Coyle even identified specific attorneys and law firms in support of his litigation threats.2 Id.

EFI's general counsel traveled to Nevada on December 5, 2001 to negotiate terms of an agreement by which EFI could purchase or license Coyle's technology. In the course of the discussions, Coyle stated that "I will sue you and I will fight" and that "all hell w[ill] break loose, I will fight until the end...." Dismissal Order, slip op. at 5. Additionally, Coyle purportedly gave EFI an ultimatum, warning that December 15 was the deadline to pay ("If we don't get a deal, we will pull the trigger and execute the litigation," (Compl. ¶ 19)), but negotiations between the two parties again broke down. On December 11, 2001, EFI sued Coyle and Kolbet Labs in the United States District Court for the Northern District of California, seeking a declaratory judgment that EFI did not breach the two non-disclosure agreements and that EFI did not misappropriate Coyle's trade secrets.

On January 8, 2002, Coyle's '746 patent issued, and EFI amended its complaint that same day to assert noninfringement and invalidity of the patent. Later that month, Coyle filed motions to dismiss the complaint for lack of personal jurisdiction, for improper venue, and for failure to comport with the objectives of the Declaratory Judgment Act. In March 2002, the court granted Coyle's motion to dismiss specifically for lack of personal jurisdiction, but it reserved judgment on Coyle's other motions to dismiss.

In August 2003, a panel of this court reversed the district court's dismissal and remanded the case to the district court. Elecs. for Imaging, Inc. v. Coyle, 340 F.3d 1344 (Fed.Cir.2003). We determined that the district court erred by applying Ninth Circuit law on personal jurisdiction to the patent invalidity claim. Instead, we applied Federal Circuit law, concluding that Coyle had not shown that the case was "one of the `rare' situations in which sufficient minimum contacts exist but where the exercise of jurisdiction would be unreasonable." Id. at 1352.

In February 2004, on remand, the district court considered the other grounds for dismissal previously raised by Coyle. The court determined that EFI did not have any uncertainty about Coyle's intention to sue because Coyle had provided EFI with specific deadlines for reaching an agreement, which deadlines had not yet been met. Dismissal Order, slip op. at 4-5. Additionally, the court found that EFI was not uncertain about the strength of its legal position. Id. The court thus decided that EFI's declaratory suit did not serve the objectives of the Act and was merely anticipatory, designed to preempt Coyle's suit and to secure EFI's own choice of forum instead. Id. at 6-7. Ultimately, the court granted Coyle's motion to dismiss the complaint for failure to comport with the objectives of the Act and did not reach the remaining issues (e.g., motion to dismiss or transfer for improper venue and motion to transfer for convenience). Id. at 1. EFI timely appealed; we have jurisdiction pursuant to 28 U.S.C. § 1295(a)(1).

The sole issue now before us is whether the district court properly dismissed EFI's lawsuit under the Declaratory Judgment Act. The decision to stay or dismiss a declaratory action is reviewed for an abuse of discretion, Wilton v. Seven Falls Co., 515 U.S. 277, 289, 115 S. Ct. 2137, 132 L. Ed. 2d 214 (1995), which occurs when: "(1) the court's decision was clearly unreasonable, arbitrary, or fanciful; (2) the decision was based on an erroneous conclusion of law; (3) the court's findings were clearly erroneous; or (4) the record contains no evidence upon which the court rationally could have based its decision," Minn. Mining & Mfg. Co. v. Norton Co., 929 F.2d 670, 673 (Fed.Cir.1991).

A declaratory action allows a party "who is reasonably at legal risk because of an unresolved dispute, to obtain judicial resolution of that dispute without having to await the commencement of legal action by the other side." BP Chems. Ltd. v. Union Carbide Corp., 4 F.3d 975, 977 (Fed.Cir.1993). The Declaratory Judgment Act provides that:

In a case of actual controversy within its jurisdiction, ... any court of the United States, upon the filing of an appropriate pleading, may declare the rights and other legal relations of any interested party seeking such declaration, whether or not further relief is or could be sought. Any such declaration shall have the force and effect of a final judgment or decree and shall be reviewable as such.

28 U.S.C. § 2201(a) (2000).

The district court is not required to exercise declaratory judgment jurisdiction, but has "unique and substantial discretion" to decline that jurisdiction. Wilton, 515 U.S. at 286, 115 S. Ct. 2137; EMC Corp. v. Norand Corp., 89 F.3d 807, 810 (Fed.Cir.1996). The use of discretion is not plenary, however, for "[t]here must be well-founded reasons for declining to entertain a declaratory judgment action." Capo, Inc. v. Dioptics Med. Prods., 387 F.3d 1352, 1355 (Fed.Cir.2004); see also Genentech v. Eli Lilly & Co., 998 F.2d 931, 937 (Fed.Cir.1993) ("When there is an actual controversy and a declaratory judgment would settle the legal relations in dispute and afford relief from uncertainty or insecurity, in the usual circumstance the declaratory judgment is not subject to dismissal."); cf. BP Chems., 4 F.3d at 981 ("A court may decline to exercise declaratory judgment jurisdiction if it would not afford relief from the uncertainty, insecurity, and controversy giving rise to the proceeding.").

Federal courts must act "in accordance with the purposes of the Declaratory Judgment Act and the principles of sound judicial administration" when declining jurisdiction in declaratory suits. EMC Corp., 89 F.3d at 813-14. The question whether to accept or decline jurisdiction in an action for a declaration of patent rights in view of a later-filed suit for patent infringement impacts this court's mandate to promote national uniformity in patent practice. Because it is an issue that falls within our exclusive subject matter jurisdiction, we do not defer to the procedural rules of the regional circuits nor are we bound by their decisions. Serco Servs. Co., L.P. v. Kelley Co., Inc., 51 F.3d 1037, 1038 (Fed.Cir.1995); Genentech, 998 F.2d at 937.

A. EFI's "Uncertainty"

On appeal, EFI argues that the district court incorrectly determined that EFI's suit against Coyle and Kolbet Labs would not serve the objectives of the Declaratory Judgment Act. Particularly, EFI assigns error to the court's conclusion that because EFI was confident in its legal position, EFI had no "uncertainty" of the type contemplated by the Act. Instead, EFI asserts that the district court should have focused on the uncertainty created by Coyle's persistent and forceful threats of patent infringement against EFI.

Coyle responds by arguing that the district court properly rejected EFI's argument regarding uncertainty in the context of the Act. He claims that because EFI had assured itself that Coyle's claims of patent infringement were, in EFI's view, meritless, EFI had no ultimate uncertainty regarding its liability to Coyle. Coyle avers that because EFI was, in fact, "certain" of its rights following an investigation of Coyle's claims, EFI's suit did not serve the purposes of the Act and EFI thus had no basis upon which it could properly file a declaratory action.

We agree with EFI that the district court erred as a matter of law when it held that EFI suffered no uncertainty of the kind recognized by the Declaratory Judgment Act. We have stated that "the purpose of the Declaratory Judgment Act ... in patent cases is to provide the allegedly infringing party relief from uncertainty and delay regarding its legal rights." Goodyear Tire & Rubber Co. v. Releasomers, Inc., 824 F.2d 953, 956 (Fed.Cir.1987). The very scenario that EFI faced was a motivation for enacting the Act, as we have stated in the past:

[A] patent owner ... attempts extra-judicial patent enforcement with scare-the-customer-and-run tactics that infect the competitive environment of the business community with uncertainty and insecurity.... Before the Act, competitors victimized by that tactic were rendered helpless and immobile so long as the patent owner refused to grasp the nettle and sue. After the Act, those competitors were no longer restricted to an in terrorem choice between the incurrence of a growing potential liability for patent infringement and abandonment of their enterprises; they could clear the air by suing for a judgment that would settle the conflict of interests.

Arrowhead Indus. Water, Inc. v. Ecolochem, Inc., 846 F.2d 731, 735 (Fed.Cir.1988) (citation omitted).

Here, the district court misinterpreted the term "uncertainty" in the context of the Declaratory Judgment Act. The proper inquiry should not have been whether a party is "certain" that its legal position and defense theories are sound, because litigation is rarely "certain," even if one is confident of one's position. When a party is threatened as EFI has been, there are other uncertainties, including whether there will be legal proceedings at all, not just whether one will prevail. There is the uncertainty whether one will have to incur the expense and inconvenience of litigation, and how it will affect the threatened party's customers, suppliers, and shareholders. Reservation of funds for potential damages may be necessary. If "certainty" in one's position were to preclude invocation of the Act, then the Act would fail to achieve its purpose of promoting resolution of disputes, as parties threatened with unjustified litigation, even if they are "certain" that they will prevail, must incur other uncertainties. When threatened, they are therefore entitled to sue to bring an end to the threat, despite their confidence in their position. "Uncertainty" in the context of the Act refers to the reasonable apprehension created by a patentee's threats and the looming specter of litigation that results from those threats. See Minn. Mining, 929 F.2d at 673. Those uncertainties were surely present here, and they were created by Coyle.

In Minnesota Mining, which is instructive on this point, the Norton Company threatened 3M and its customers with allegations of patent infringement. Meanwhile, 3M continued to sell products that it believed did not infringe. As a result, in that case, despite confidence in its noninfringement position, 3M's potential liability grew with each sale. Id. at 673-74. We recognized that 3M's situation was the quandary that the Act was designed to alleviate and stated: "In promulgating the Declaratory Judgment Act, Congress intended to prevent avoidable damages from being incurred by a person uncertain of his rights and threatened with damage by delayed adjudication." Id. at 673.

Likewise, in the present case, EFI may in fact be confident that it would eventually prevail in a patent infringement action. Indeed, Coyle has identified communications between EFI and its counsel that indicated confidence in their invalidity defense against Coyle's "meritless" allegations. Nevertheless, Coyle's forceful threats created a cloud over EFI's business, shareholders, and customers, and EFI's potential liability increased as it continued to sell the allegedly infringing products. EFI is entitled under the Declaratory Judgment Act to seek a timely resolution of Coyle's threats of litigation and remove itself from "the shadow of threatened infringement litigation." Serco Servs., 51 F.3d at 1038. The fact that Coyle had stated a deadline for negotiations to be concluded, and that that deadline had not passed when EFI brought suit, does not deprive EFI of the right to sue. Given that Coyle had a record of threatening suit in such clear and descriptive language without always following through promptly on those threats, EFI was not required to await suit by Coyle. It was entitled to bring the matter to a head and have it resolved in court.

EFI also argues that the district court abused its discretion by finding that EFI's declaratory action was filed in anticipation of Coyle's impending suit, and that such a filing justified dismissal. EFI asserts that the district court inaccurately described the filing of its suit as prompting a "race to the courthouse." Coyle responds that the district court properly concluded that EFI's suit was anticipatory because EFI knew that Coyle's suit was imminent.

We agree with EFI that the district court abused its discretion in its dismissal order by focusing on the anticipatory nature of the suit. We apply the general rule favoring the forum of the first-filed case, "unless considerations of judicial and litigant economy, and the just and effective disposition of disputes, requires otherwise." Genentech, 998 F.2d at 938; see Serco Servs., 51 F.3d. at 1039. "Exceptions ... are not rare," but we have explained that "[t]here must ... be sound reason that would make it unjust or inefficient to continue the first-filed action." Genentech, 998 F.2d at 937-38.

While it is true that a district court may consider whether a party intended to preempt another's infringement suit when ruling on the dismissal of a declaratory action, Serco Servs., 51 F.3d. at 1040, we have endorsed that as merely one factor in the analysis. Other factors include "the convenience and availability of witnesses, or absence of jurisdiction over all necessary or desirable parties, or the possibility of consolidation with related litigation, or considerations relating to the real party in interest." Genentech, 998 F.2d at 938. "The considerations affecting transfer to or dismissal in favor of another forum do not change simply because the first-filed action is a declaratory action." Id. No such "other factors" have been cited here.

In Genentech, we reversed the district court's dismissal of a declaratory action, as it was premised solely on the fact that the suit was designed to anticipate a later-filed complaint in another forum. Id. On the other hand, in Serco Services, the district court dismissed Serco's declaratory action as anticipatory, but did so in view of other factors, including the location of witnesses and documents. Serco Servs., 51 F.3d at 1039-40. We affirmed the dismissal in Serco Services because the district court did not rely solely on the anticipatory nature of Serco's declaratory action. Id. at 1040.

Here, the facts are more similar to those in Genentech and are distinguishable from those in Serco Services. As we have determined that the district court's reliance on EFI's lack of uncertainty in its legal position was erroneous, and no other compelling factors have been cited, the court's decision is left to rest exclusively on the alleged anticipatory nature of EFI's suit. Our precedent, however, favors the first-to-file rule in the absence of circumstances making it "unjust or inefficient" to permit a first-filed action to proceed to judgment, and, as indicated, no such circumstances have been shown here.

The district court's order dismissing EFI's suit as anticipatory and contrary to the purposes of the Declaratory Judgment Act was thus an abuse of discretion.

The district court abused its discretion by dismissing EFI's suit against Coyle, and we therefore reverse the court's dismissal and remand for further proceedings consistent with this opinion.