UNITED STATES OF AMERICA, APPELLEE v. HONEYWELL INTERNATIONAL INC., APPELLANT
No. 21-5179
United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT
Argued March 30, 2022 Decided August 30, 2022
Appeal from the United States District Court for the District of Columbia (No. 1:08-cv-00961)
Craig S. Primis argued the cause for appellant. With him on the briefs was James Y. Xi.
Tara S. Morrissey, Andrew R. Varcoe, Beth S. Brinkmann, and Daniel G. Randolph were on the brief for amicus curiae Chamber of Commerce of the United States of America in support of appellant.
Sean R. Janda, Attorney, U.S. Department of Justice, argued the cause for appellee. With him on the brief were Brian M. Boynton, Acting Assistant Attorney General, and Michael S. Raab and Charles W. Scarborough, Attorneys.
Before: TATEL* and RAO, Circuit Judges, and GINSBURG, Senior Circuit Judge.
Opinion for the Court filed by Circuit Judge RAO.
RAO, Circuit Judge:
I.
A.
The False Claims Act (“FCA“) “impose[s] liability for fraud against the government.” United States ex rel. Cimino v. Int‘l Bus. Machs. Corp., 3 F.4th 412, 415 (D.C. Cir. 2021). See generally Act of Mar. 2, 1863, ch. 67, 12 Stat. 696 (codified as amended at
In FCA cases, when multiple parties cause the same indivisible harm to the government, courts have applied joint and several liability1 without a right to contribution. See, e.g., United States v. TDC Mgmt. Corp., Inc., 288 F.3d 421, 429 (D.C. Cir. 2002) (applying joint and several liability to an FCA claim). In general, when a joint tortfeasor settles, the settlement counts against a non-settling party‘s potential liability. See RESTATEMENT (SECOND) OF TORTS § 886A cmt. m (1979) (assuming settlement offsets and recognizing the lack of consensus about how to assess the proper offset). The parties agree as to these basic principles of law. The question of first impression we decide is the correct rule for calculating the settlement credit for FCA damages.
B.
Because this case comes to us on an interlocutory appeal, we recount the undisputed facts as found by the district court on summary judgment. The litigation arises from allegedly fraudulent claims for bulletproof vests made from “Z Shield.” Z Shield is an “anti-ballistic material” made of “Zylon” fiber that Honeywell purchased from third parties. Honeywell sold more than $15 million worth of Z Shield to Armor Holdings, Inc., which in turn incorporated the material into bulletproof vests. Armor Holdings sold those vests to the federal government and to state and local law enforcement agencies who purchased the vests with federal funding. United States v. Honeywell Int‘l Inc., 502 F. Supp. 3d 427, 435 (D.D.C. 2020). “Honeywell marketed Z Shield ... as the best ballistic product in the market for ballistic resistance.” Id. (cleaned up). But when two police departments tested the vests in high temperatures, they allegedly degraded. Id. at 446, 462–63. Upon further investigation, the federal government concluded that “Zylon [was a] material that appears to create risk of death or serious injury as a result of degraded ballistic performance when used in body armor” and stopped buying the vests. Id. at 447 (cleaned up).
In light of these settlements, Honeywell moved for summary judgment on the question of damages (as well as other issues not before us on this interlocutory appeal). Honeywell claimed it was entitled to a credit in the amount of the settlements already secured by the government as compensation for the allegedly defective Z Shield vests. Honeywell maintained the court should apply a pro tanto2 approach, reducing any common damages Honeywell owed by the amount of the settlements. Applying that approach here would mean that, even if the government‘s allegations were true, Honeywell would pay no damages because the settlements exceeded its alleged damages liability. Opposing this pro tanto approach to calculating settlement offsets, the government argued the district court should apply a proportionate share approach. Under this approach, Honeywell‘s settlement credit would be limited by the other parties’ proportion of fault, meaning Honeywell would still be responsible for its proportionate share of the $35 million, regardless of the amount of the settlements.
The district court adopted the proportionate share approach advanced by the government. The court relied on McDermott, Inc. v. AmClyde, in which the Supreme Court addressed the proper settlement credit rule for admiralty suits and chose the proportionate share rule over the pro tanto rule. 511 U.S. 202, 207, 217 (1994); see also id. at 208–209 (distinguishing the proportionate share and pro tanto approaches). The district court balanced four considerations in choosing the proportionate share rule: “consistency with prior decisions, promotion of settlement, judicial economy, and equity.” Honeywell, 502 F. Supp. 3d at 480 (citing McDermott, 511 U.S. at 208–17). For the purposes of the FCA, the district court found the consistency with precedent factor inconclusive; the proportionate share approach “does not undermine the incentive to settle” and “is no less efficient or workable than” the pro tanto approach; and “the proportionate share approach is more equitable.” Id. at 482. The court emphasized that applying the pro tanto rule “would be wholly inequitable” because it would “permit Honeywell to escape damages liability altogether.” Id. at 485.
The court certified this question for interlocutory review under
II.
By permitting this interlocutory appeal on the question of the appropriate measure of damages offsets under the
Although the de novo standard of review is well established in these circumstances, the government maintains we should review the district court‘s choice of a settlement offset approach for abuse of discretion. It contends the offset rule is not a question of law because the FCA does not resolve the issue and courts have discretion to pick a settlement offset approach on a case-by-case basis.
We reject the government‘s characterization of the issue on appeal. Although we agree the FCA does not provide a settlement offset rule, as explained further below, the choice of a damages rule is not a matter of judicial discretion, but rather requires a common law determination of the proper rule to apply in FCA suits. In other contexts when it was necessary for a court to establish a proper settlement offset rule, the Supreme Court and other courts of appeals have treated the determination as a question of law.3 See, e.g., McDermott, 511 U.S. at 207 (granting certiorari to “fashion the rule” for settlement offsets in admiralty); Franklin v. Kaypro Corp., 884 F.2d 1222, 1228 (9th Cir. 1989) (deciding a single rule to govern settlement offsets for a securities statute); Singer v. Olympia Brewing Co., 878 F.2d 596, 599 (2d Cir. 1989) (same). When federal courts set damages rules, they do so as a matter of common law, not amorphous judicial discretion. Fashioning a common law rule is not the same as exercising discretion: common law courts decide cases one at a time, but their reasoning provides a binding rule of decision for like cases in the future. This comports with the basic function of the judicial branch to say what the law is, and it allows future parties to structure their decisions around predictable legal rules. Cf. Antonin Scalia, The Rule of Law as a Law of Rules, 56 U. CHI. L. REV. 1175, 1185 (1989) (“[T]he establishment of broadly applicable general principles is an essential component of the judicial process.“).
The government also suggests we should apply an abuse of discretion standard because the district court‘s reliance on “equity” shows it was exercising discretion. While the district court referred to equitable considerations, its orders follow McDermott to set a legal rule for calculating settlement offsets. And the court certified the settlement offset question under Section 1292(b) because it was a “quintessential abstract legal issue.” Honeywell, 2021 WL 2493382, at *2, *5 (cleaned up).
This appeal requires us to decide the proper rule for the calculation of settlement offsets under the FCA, a legal question we decide de novo.
III.
The correct rule for settlement offsets is a component of FCA damages liability, and therefore the threshold question is whether the FCA provides a settlement offset rule. Because “[a]ll legislative Powers” are vested in Congress,
The FCA effectively created a tort cause of action for the government. But the FCA makes no mention of either settlement credits or joint liability. It states that a person who makes a false claim to the government “is liable” to the United States for “3 times the amount of damages which the Government sustains because of the act of that person.”
That literal interpretation, however, would conflict with United States v. Bornstein, in which the Supreme Court applied joint and several liability. 423 U.S. 303, 314 (1976). A literal reading would also conflict with the principle, long reflected in the common law, that settlement with, or successful litigation against, one party reduces the damages owed by other parties who are jointly liable. When courts consider how to handle settlements between a plaintiff and one of several joint tortfeasors, they assume tort regimes must have some way of adjusting damages when one has paid in part. See, e.g., McDermott, 511 U.S. at 208–11; Zenith Radio Corp. v. Hazeltine Rsch., Inc., 401 U.S. 321, 343–46 (1971) (explaining the various approaches to crediting partial settlement of statutory tort claims); see also RESTATEMENT (SECOND) OF TORTS § 886A cmt. m. We cannot conclude from the simple use of the term “liable” in the FCA that Congress abrogated the background assumption that at least some settlement credit is appropriate. See Allen v. District of Columbia, 969 F.3d 397, 402 (D.C. Cir. 2020) (explaining that courts assume statutes’ continuity with “well established” common law principles unless the statute “speaks[s] directly to the question addressed by the common law“) (cleaned up).
In any event, the parties do not question whether joint and several liability is appropriate under the FCA, and they assume that common damages must be subject to some type of settlement offset. A rule for allocating settlement credits is necessary to resolve disputes between joint violators of the FCA, and the text of the statute provides none.
Honeywell posits that even if the FCA‘s text does not provide an answer, its historical context does. Honeywell maintains the FCA incorporates the pro tanto rule because “[w]hen Congress enacted the False Claims Act in 1863, the pro tanto approach was firmly established” as a background principle of law. Because Congress enacts statutes against a common law baseline, it is sometimes appropriate to interpret statutory terms in light of common law principles and presume Congress employs common law terms with their common law meaning, absent a contrary indication in the statute. See Morissette v. United States, 342 U.S. 246, 263 (1952) (“[W]here Congress borrows terms of art in which are accumulated the legal tradition and meaning of centuries of practice, it presumably knows and adopts the cluster of
With respect to the appropriate settlement rule for the FCA, however, the historical context does not settle the interpretive question. First, the pro tanto rule is not part of the definition of any common law term in the FCA. “The first precondition of any term-of-art reading is that the term be present in the disputed statute.” Borden v. United States, 141 S. Ct. 1817, 1828 (2021) (plurality opinion). Honeywell relies exclusively on the statutory term “liable,” but that is merely a general term meaning “[r]esponsible or answerable in law.” Liable, BLACK‘S LAW DICTIONARY (11th ed. 2019). By providing liability for damages, the FCA does not use a term of art, common law or otherwise, that necessitates a particular settlement offset rule. Nor is a settlement offset rule an element implicit in the common law of fraud, aspects of which are adopted by the FCA. Cf. Univ. of Tex. Sw. Med. Ctr. v. Nassar, 570 U.S. 338, 347 (2013) (applying the traditional but-for causation standard to Title VII retaliation claims by applying “textbook tort law“). The text of the FCA includes no tort term of art codifying the common law of settlement offsets from 1863, and therefore we have no basis to assume that Congress incorporated the pro tanto rule through its use of the word “liable.”
Moreover, examining the history the parties have presented to this court, we note that what appears to have been the prevailing common law rule in 1863 was much harsher than the modern pro tanto rule. It is unclear whether a consensus on settlement offsets existed in 1863. But to the extent we can glean a consensus, it was a strict version of the “one satisfaction rule,” according to which settlement of any amount with one joint tortfeasor would bar all collection from the others. See Lovejoy v. Murray, 70 U.S. (3 Wall.) 1, 17 (1865) (holding future litigation is barred if the plaintiff “has received full satisfaction, or that which the law must consider as such“); Sheldon v. Kibbe, 3 Conn. 214, 220 (1819) (explaining a plaintiff‘s acceptance of satisfaction is deemed full satisfaction and bars future litigation). Neither party advocates this harsh rule, which has been thoroughly rejected by American courts. RESTATEMENT (SECOND) OF TORTS § 885 cmt. b; see also Zenith Radio Corp., 401 U.S. at 344 (noting the one satisfaction rule has never been applied to a federal statutory cause of action). In this context, there is simply no venerable common law principle of settlement offsets that we can reasonably presume is implied in the FCA. Cf. Allen, 969 F.3d at 403 (applying an implicit “well-established common-law principle” that “is as old as the Republic“).
Nor have prior decisions of the Supreme Court, this court, or other courts of appeals explicitly determined which settlement offset rule should be applied to FCA damages. Precedent establishes that liability under the FCA is joint and several with no right to contribution among joint violators. See, e.g., Mortgages, Inc. v. U.S. Dist. Ct. for Dist. of Nev., 934 F.2d 209, 212 (9th Cir. 1991) (per curiam) (“Where one or more persons have committed a fraud upon the government in violation of the FCA, each is joint and severally liable for the treble damages and statutory penalty.“); id. at 213–14 (holding that there is no
IV.
Because the question of what settlement offset rule to apply is not answered by the text of the FCA, the common law background in 1863, or existing case law, we must establish the correct rule in order to decide this case. A federal common law rule is necessary in this context, and we conclude the pro tanto rule is the appropriate measure of settlement offsets when calculating damages under the FCA.
A.
The Article III courts have limited jurisdiction, and it is long established that “[t]here is no federal general common law” because federal courts cannot arrogate the power of the states or Congress to make substantive rules of law. Erie R.R. Co. v. Tompkins, 304 U.S. 64, 78 (1938). Nonetheless, since Clearfield Trust Co. v. United States, 318 U.S. at 366–67, in cases involving “the government‘s proprietary interests” it has been clear that federal courts have the “competence” to create a federal common law rule. R. FALLON, J. MANNING, D. MELTZER, & D. SHAPIRO, HART AND WECHSLER‘S THE FEDERAL COURTS AND THE FEDERAL SYSTEM 661 (7th ed. 2015). Federal courts create rules of decision when necessary to protect “uniquely federal interest[s],” such as when “the application of state law would frustrate specific objectives of federal legislation.” Boyle v. United Techs. Corp., 487 U.S. 500, 507 (1988) (cleaned up).
As the Supreme Court has explained, a “precondition” for applying federal law is a “significant conflict between some federal … interest and the use of state law” which must be “specifically shown.” Atherton v. FDIC, 519 U.S. 213, 218 (1997) (cleaned up); see also HART & WECHSLER, supra, at 661 (explaining that “application of state law is the proper default position for federal courts“). The competence to create a federal rule does not negate the “basic aspect of our federalism” that the federal government always acts “against a background of existing state law, and that federal law retains its incomplete and interstitial nature.” Ernest A. Young, Preemption and Federal Common Law, 83 NOTRE DAME L. REV. 1639, 1646 (2008) (cleaned up). Therefore, even areas of unique federal interest “do not inevitably require resort to uniform federal rules.” Kimbell Foods, 440 U.S. at 727–28. In exercising the discretion to create a federal rule, federal courts must consider whether such a rule is appropriate in light of the “nature of the specific governmental interests” at stake and “the effects upon them of applying state law.”4 Id. at 728 (cleaned up).
Applying this framework, we have authority to create a federal rule for FCA
First, by providing the federal government a cause of action to remedy fraudulent claims, the FCA safeguards the public fisc and protects the United States against “proprietary injury.” Vt. Agency of Nat. Res. v. United States ex rel. Stevens, 529 U.S. 765, 771 (2000). The rights to recovery under the FCA are analogous to the federal government‘s contract rights, long recognized as a “uniquely federal interest” that federal courts have authority to protect by creating federal common law. See Boyle, 487 U.S. at 504–05 (citing United States v. Little Lake Misere Land Co., 412 U.S. 580, 592–94 (1973); Nat‘l Metro. Bank v. United States, 323 U.S. 454, 456 (1945)). The rule for calculating settlement offsets under the FCA will affect the federal government‘s prosecution of false claims, the costs and benefits of pursuing settlement, and the ultimate amount of damages returned to the U.S. Treasury.
Second, we find that fashioning a federal rule is appropriate because there is a “significant conflict” between the federal interests protected by the FCA and “the use of state law.” Atherton, 519 U.S. at 218. “[A]dopting state law would adversely affect administration” of the FCA because the state law of settlement offsets varies widely.5 Kimbell Foods, 440 U.S. at 730; see RESTATEMENT (SECOND) OF TORTS § 886A cmt. m(3) (“Case authorities and statutes are … divided [between pro tanto and proportionate share rules] and there is no semblance of a consensus.“).
Variable settlement offset rules will make it difficult for the government to vindicate the important federal interests protected by the FCA. Without a uniform rule, the government would not have a secure baseline against which to negotiate settlements in cases involving multiple defendants and thorny choice-of-law questions. The future liability of settling parties and the government‘s ultimate recovery would remain uncertain. Cf. Clearfield Tr. Co., 318 U.S. at 367 (finding a federal rule appropriate when applying state law “would subject the rights and duties of the United States to exceptional uncertainty“). And as the facts of this case demonstrate, the choice of a settlement offset rule may have a substantial impact on the damages the government can recover. This lack of predictability is in tension with Congress’ precise calibration of FCA liability, which involves treble damages, civil penalties, and enumerated classes of liability.6 See
B.
Having concluded we must decide the offset settlement rule under the FCA as a matter of federal common law, we draw on McDermott to structure our analysis. In McDermott, the Court decided what type of settlement credit was appropriate in the admiralty context. In choosing between the pro tanto and proportionate share rules, the Court outlined three “paramount” considerations: (1) consistency with relevant precedent; (2) promotion of settlement; and (3) judicial economy.7 McDermott, 511 U.S. at 211. With respect to settlement offsets for calculating damages under the FCA, the balance of these factors favors the pro tanto rule.
We first consider which rule promotes consistency with the text and structure of the FCA and the precedent interpreting it. This factor decisively favors the pro tanto approach. At the outset, the pro tanto rule is at least compatible with the FCA.
The Supreme Court applied it without comment in Bornstein, although the Court was not asked to decide between the pro tanto and proportionate share rules in that case. 423 U.S. at 314. Moreover, the government‘s litigating position here—that the pro tanto approach is inappropriate in this case but may be appropriate in others—shows that neither party considers the pro tanto approach incompatible with the FCA.
The pro tanto rule, however, is not just compatible with the FCA; it is a better fit with the statute and the liability rules that have been partnered with it. The FCA has been consistently interpreted to impose joint and several liability without a right to contribution. See Mortgages, Inc., 934 F.2d at 212–13. Together these two rules mean that a person who violates the FCA in a joint scheme may have to pay for all the government‘s trebled damages, and, even if that defendant is the least responsible party, it cannot force the other violators to pay their fair share. When administering joint and several liability with no contribution, the court does not determine the equitable assignment of damages.
In light of this statutory context and accompanying liability rules, adopting the proportionate share rule to allocate settlement offsets would be anomalous for the FCA in at least two ways. First, liability would be joint and several if all parties litigate, but in cases of partial settlement, courts would have to decide relative culpability and assign damages based on fault. This mismatch is usually avoided because jurisdictions and statutory schemes that involve joint and several liability without contribution generally also employ the pro tanto approach. See Lewis A. Kornhauser & Richard L. Revesz,
The government suggests the proportionate share rule is more compatible with the FCA‘s punitive goals. The treble damages in the FCA are “essentially punitive in nature” and not just compensatory. See Vt. Agency of Nat. Res., 529 U.S. at 784. Applying the proportionate share rule, a court would calibrate a party‘s punishment to its relative culpability, furthering, at least to some extent, the punitive effect of the statute.
We recognize that in cases such as this where the government has already recouped its full damages from settling parties, a non-settling party like Honeywell will escape paying damages under the pro tanto rule.8 Nevertheless, consistent with the FCA, the pro tanto rule leaves the government in the driver‘s seat to pursue and punish false claims according to its priorities. The text and structure of the FCA make clear that Congress left substantial control for prosecuting false claims to the federal government. In addition to traditional agency enforcement discretion, even when a person brings a qui tam action under the FCA, the government may decide to maintain the action itself or intervene and exercise substantial control over the litigation. See, e.g.,
Furthermore, our holding does not disturb the FCA‘s civil penalties—here potentially exceeding $500,000 according to the government—which serve a punitive purpose and may in some cases even exceed the statutory damages. See, e.g., United States ex rel. Bunk v. Gosselin World Wide Moving, N.V., 741 F.3d 390, 395 (4th Cir. 2013). In sum, the first McDermott factor strongly favors the pro tanto approach.
The second McDermott factor, the promotion of settlement, is too inconclusive to provide guidance. Cf. Honeywell, 502 F. Supp. 3d at 482–83 (finding no “clear advantage” for either approach) (cleaned up); McDermott, 511 U.S. at 216 (same). We see no way to determine which rule better promotes settlement. See RESTATEMENT (THIRD) OF TORTS: APPORTIONMENT OF LIABILITY § 16 cmt. c (2000) (“[T]o the extent that facilitating the complete voluntary resolution of a suit before trial for the sake of judicial efficiency is the goal, there is no reason to believe that either a pro tanto or a comparative-share credit is preferable.“). The promotion of settlement often turns on a complex intersection of factors that are not readily ascertained nor easily balanced by courts.9
The McDermott Court‘s conclusion, in the context of admiralty law, that “[t]he pro tanto rule … has no clear advantage with respect to judicial economy” is not to the contrary. 511 U.S. at 217. That conclusion was explicitly premised on the need for “good faith hearings.” Id. at 216 (“The pro tanto rule, if adopted without the requirement of a good faith hearing, would be easier to administer.“) If the pro tanto rule were applied in admiralty cases, the court would still need to conduct “good faith hearings” because in admiralty there is a right to contribution and a court must ensure that early settlements are fairly negotiated approximations of expected liability. Id. at 213. But there is no right to contribution under the FCA, and so there is no need for good faith hearings. See Kornhauser & Revesz, 68 N.Y.U. L. REV. at 444 (recognizing “[t]here is no reason to have good faith hearings in the absence of a right of contribution“). Therefore, in the FCA context, the pro tanto rule would substantially favor judicial economy.
In conclusion, pro tanto is the settlement offset rule that best coheres with the FCA and the precedents interpreting it, and applying this rule will generally promote judicial economy. As a matter of federal common law, we find it is the appropriate method to apply when offsetting settlement credits in FCA cases.
***
In the False Claims Act, Congress created a vital mechanism for the federal government to protect itself against fraudulent claims. The FCA, however, provides no rule for allocating settlement credits among joint fraudsters. Because the FCA guards the federal government‘s vital pecuniary interests, and because state courts widely diverge over the correct rule for settlement offsets, we find it appropriate to establish a federal common law rule. The pro tanto rule best fits with the FCA and the joint and several liability applied to FCA claims. Honeywell is entitled to offset its common damages in the amount of the government‘s settlements from the other parties. We therefore reverse and remand for further proceedings consistent with this opinion.
So ordered.