Opinion
21-11136
05-23-2023
AFRICA GROWTH CORPORATION, Plaintiff-Appellant, v. REPUBLIC OF ANGOLA, Defendant-Appellee.
DO NOT PUBLISH
Appeal from the United States District Court for the Southern District of Florida D.C. Docket No. 1:19-cv-21995-KMW
Before WILSON and ROSENBAUM, Circuit Judges, and COVINGTON, [*] District Judge.
PER CURIAM:
Between 2016 and 2017, an Angolan Army general conspired with Angolan officials to steal tens of millions of dollars of property from Plaintiff-Appellant Africa Growth Corp ("AFGC"). After AFGC failed to recover its property using the Angolan legal system, it began a global lobbying campaign against Angola. And it sued Angola in the District of Columbia.
We accept AFGC's well-pleaded facts as true for purposes of reviewing the district court's order granting Angola's motion to dismiss. Am. Dental Ass'n v. Cigna Corp., 605 F.3d 1283, 1288 (11th Cir. 2010). The actual facts may or may not be as stated.
While that lawsuit was ongoing, the parties met in Lisbon to discuss settling. At the meeting, Angola agreed to pay $47.5 million dollars in exchange for AFGC's release of its claims against Angola and AFGC's promise to stop lobbying against Angola. When Angola didn't pay, AFGC sued again, this time in the southern District of Florida.
The parties dispute whether Angola has foreign sovereign immunity. The answer turns on what the "gravamen" of the action is. Under the Foreign sovereign immunities Act, if the "gravamen" of the action is Angola's expropriation of AFGC's property, then Angola has immunity. But if the "gravamen" of the action is the breach of the settlement agreement-and if the breach of the settlement agreement is commercial, not sovereign, in nature-then the Foreign Sovereign Immunities Act's exception for commercial activity applies and Angola isn't immune from suit.
So to resolve this appeal, we must determine what the gravamen of the action is. While the case law nationwide is in some tension with itself, ultimately our precedent provides the answer. If a sovereign expropriates property and then breaches a contract to provide compensation for the taking, the gravamen of the ac-tion-what injured the plaintiff-is the taking, not the breach of the contract.
After a thorough review of the record, and with the benefit of oral argument, we conclude that Angola's taking of AFGC's property was what injured AFGC here. For that reason, we affirm.
I. BACKGROUND
In 2015, AFGC, a Nevada corporation, bought four real-estate and commercial properties-through subsidiaries-in the Angolan capital, Luanda. The next year, the Angolan government used "fraudulent documents, abuse of power, forgery, intimida-tion[,] and force of arms" to take control of the subsidiary companies and thereby, the properties. Although AFGC obtained Angolan court orders in its favor, the Angolan government "actively refused to enforce [the] orders."
So AFGC mounted a two-pronged attack. On the publicrelations side, AFGC began a "global lobbying effort" against Angola. Meanwhile, on the legal side, AFGC sued Angola, in the District of Columbia, for expropriation. Afr. Growth Corp. v. Republic of Angl., 2019 WL 3253367 (D.D.C. July 19, 2019).
While the D.C. case was pending, AFGC offered to meet with Angolan President Joao Lourengo to settle the dispute. Dr. Eduarda Rodrigues Neto-Angola's Deputy Attorney General- responded that President Lourengo had directed her to meet with AFGC instead.
The next month, the parties met in Lisbon. Dr. Rodrigues Neto represented "that she was the specially authorized representative for the Office of the Angolan Attorney General and was attending the settlement meeting at the express direction of President Lourengo." The parties reached an agreement that Angola would pay AFGC $47.5 million in exchange for AFGC relinquishing all rights and claims to the properties and-upon payment in full-a promise that AFGC would dismiss its D.C. lawsuit and end its lobbying efforts against Angola. The parties agreed to meet the next week to sign a written settlement agreement. Dr. Rodrigues Neto added that "the sole reason for creating the subsequent writing for signature was to accommodate certain formalities required by the National Bank of Angola for the payment of funds outside Angola."
But the parties never reconvened. And when AFGC advised Dr. Rodrigues Neto that it expected Angola to uphold its end of the bargain, she denied agreeing to anything, explaining that she had "clearly stated" that she had no power to bind the Angolan state.
When Angola failed to pay AFGC by the agreed-upon date, AFGC sued again, this time in the Southern District of Florida.AFGC asserted a cause of action for breach of contract, seeking the $47.5 million owed for the expropriated property and "further consequential damages, which collectively total in excess of USD 95 million." AFGC explained that the extra $95 million was comprised of $72 million in lost shareholder value and $23 million in lost opportunities, lost profits, and costs. In the alternative, AFGC asserted a claim against Angola for unjust enrichment.
The D.C. lawsuit was dismissed on sovereign-immunity grounds, and AFGC voluntarily dismissed its appeal-both long after the events at issue in this case. Afr. Growth Corp. v. Republic of Angl., 17-cv-02469 (D.D.C.), ECF Nos. 7273, 76-1.
We aren't sure whether the "collectively" refers to both the expropriated property and the consequential damages or just to the consequential damages. If it refers to both, then AFGC is seeking collectively $95 million. If it refers to only the consequential damages, then AFGC is seeking about $142.5 million ($47.5 million in expropriated property and $95 million in consequential damages). Either way, as we explain below, AFGC alleged that all its harm occurred before Angola breached the contract.
Angola moved to dismiss, arguing that the district court lacked subject matter jurisdiction because of the Foreign Sovereign Immunities Act ("FSIA" or "the Act"), 28 U.S.C. § 1602, et seq. In Angola's view, the Act's broad grant of immunity applied, and AFGC's pleading didn't satisfy the commercial-activity exception. See id. § 1605(a)(2). Angola said that AFGC's suit was "based on" the expropriation of AFGC's property, not a breach of contract. So because expropriation wasn't commercial activity, Angola concluded, the general FSIA immunity applied, not subject to a limited exception.
The district court agreed and granted Angola's motion to dismiss. In explaining the basis for its decision, the district court said that "the gravamen of this case is the expropriation of real property," so Angola retained its immunity. AFGC now appeals.
II. STANDARD OF REVIEW
We review de novo a district court's determination of whether it had jurisdiction under the FSIA. Devengoechea v. Bolivarian Republic of Venez., 889 F.3d 1213, 1220 (11th Cir. 2018).
III. DISCUSSION
AFGC contends that the gravamen of its complaint is the breach of contract and that the breach of contract was commercial, not sovereign in nature. We disagree.
We proceed in three steps. First, we outline how the Foreign Sovereign Immunities Act works and how its exceptions apply. Second, we conclude that this lawsuit is "based on" Angola's expropriation of AFGC's property. Third, we determine that Angola's expropriation isn't commercial activity that satisfies the FSIA's commercial-activity exception.
A. The Foreign Sovereign Immunities Act in General
Under the Foreign Sovereign Immunities Act, "a foreign state is presumptively immune from the jurisdiction of United States courts; unless a specified exception applies, a federal court lacks subject-matter jurisdiction over a claim against a foreign state." Saudi Arabia v. Nelson, 507 U.S. 349, 355 (1993); see 28 U.S.C. § 1605 (listing exceptions). Because AFGC argues only that the commercial-activity exception applies, we limit our discussion to that exception. See 28 U.S.C. § 1605(a)(2).
As relevant here, the third clause of the commercial-activity exception provides jurisdiction over a foreign state when "the action is based . . . upon an act outside the territory of the United States in connection with a commercial activity of the foreign state elsewhere and that act causes a direct effect in the United States." Id. § 1605(a)(2).
The commercial-activity exception has other clauses, but because AFGC doesn't argue that those clauses supply jurisdiction, we don't discuss them. See 28 U.S.C. § 1605(a)(2).
We've explained that, to invoke jurisdiction under this clause, "(1) the lawsuit must be based upon an act that took place outside the territory of the United States; (2) the act must have been taken in connection with a commercial activity; and (3) the act must have caused a direct effect in the United States." Deven-goechea, 889 F.3d at 1224 (quoting de Csepel v. Republic of Hung., 714 F.3d 591, 598 (D.C. Cir. 2013)) (cleaned up).
Only the second and third prongs are at issue. The parties dispute which act the lawsuit is "based on" and whether the act (whatever it was) was taken in connection with a commercial activity as well as whether that act (again, whatever it was) caused a direct effect in the United States. We first must decide what act this lawsuit is "based on."
Either way, the act took place outside the United States. If the act in question was the expropriation, then it took place in Luanda, Angola. And if the act was the breach of the settlement agreement, then it took place in Lisbon, Portugal.
B. AFGC's lawsuit is "based on" the expropriation of its property.
"[A]t the first step in a commercial-activity-exception case, we must identify the conduct upon which the suit is based." De-vengoechea, 889 F.3d at 1222 (citing OBB Personenverkehr AG v. Sachs, 577 U.S. 27, 33 (2015)). Unfortunately, "[t]he Act itself does not elaborate on the phrase 'based upon,'" Sachs, 577 U.S. at 33, and the Supreme Court has noted that "the relatively sparse legislative history offers no assistance." Nelson, 507 U.S. at 357.
When assessing what act a suit is "based upon," we must look at its "basis" or "foundation." Id. (citations omitted). "That, in turn, requires us to look at 'the "particular conduct" that constitutes the "gravamen" of the suit.'" Devengoechea, 889 F.3d at 1222 (quoting Sachs, 577 U.S. at 33). "In other words, we focus on the 'core' of the suit-the foreign state's 'acts that actually injured' the plaintiff." Id. (quoting Sachs, 577 U.S. at 33); see also Gravamen, Black's Law Dictionary (11th ed. 2019) (defining "gravamen" as "[t]he substantial point or essence of a claim, grievance, or complaint."). We look to the "elements of a claim that, if proven, would entitle a plaintiff to relief." Nelson, 507 U.S. at 357. And we compare the plaintiff's "theory of the case" and the facts her "claims turn on" with the act she claims lies at the basis of her suit. Sachs, 577 U.S. at 35.
We don't engage in a "one-element approach." Under that approach, if an act establishes "a single element of a claim," then a suit raising the claim is "based upon" the act. Sachs, 577 U.S. at 395-96.
To explain what this means, we consider two Supreme Court decisions that flesh out this analysis. We concede, though, that neither gets us all the way to an answer in this case.
First, in Sachs, the plaintiff bought a European train ticket in California. Id. at 30. Unfortunately, she fell and injured herself in Austria. Id. The Supreme Court determined that the "gravamen" of her tort claim was the wrongful failure to warn and the dangerous conditions that resulted in her injury-both of which occurred in Austria. Id. at 35-36 (explaining that there was nothing wrongful about the sale of the ticket standing alone so, without the unsafe boarding conditions, there would have been nothing to warn the plaintiff about). So despite a commercial aspect (buying the train ticket), the Supreme Court determined that the gravamen of the suit was the slip-and-fall in Austria. Id.
Next, in Nelson, the plaintiff signed an employment agreement to work at a government-run hospital in Saudi Arabia. 507 U.S. at 358. Regrettably, during the plaintiff's stay in that country, the Saudi government arrested and tortured the plaintiff. Id. at 352-53. The plaintiff sued for tort claims in connection with his employment agreement, but the Supreme Court again determined that the "gravamen" of the plaintiff's claims centered on Saudi Arabia's tortious conduct. Id. at 358.
These cases are helpful, but as we mentioned, they don't help us to fully resolve the case before us. That's because in those cases, only one wrongful act-the tortious act-occurred. See Sachs, 577 U.S. at 35 ("[T]here is nothing wrongful about the sale of the Eurail pass standing alone."). Here, though, two wrongful acts-the expropriation and the breach of the settlement agreement-happened. Id. at 36 n.2 ("[W]e consider here only a case in which the gravamen of each claim is found in the same place."). So a case with just a single wrongful act does not fully enlighten us as to what the "gravamen" of a case is when two wrongful acts are alleged.
As it turns out, though, our precedent does. Indeed, it compels the outcome here. In Beg v. Islamic Republic of Pakistan, the Pakistani government seized millions of dollars of property from the plaintiff there. 353 F.3d 1323, 1324 (11th Cir. 2003). The Pakistani government offered the plaintiff "an alternative parcel of land," which Beg accepted. Id. But the Supreme Court of Pakistan refused to enforce the agreement because the Pakistani government didn't have title to the alternative parcel. Id. Though Beg sued, we determined that the gravamen of his claim was the expropriation- which was a sovereign taking, not a commercial activity. Id. We rejected Beg's argument that agreement between himself and the government was enough to invoke the commercial-activity exception because "[d]etermining whether or how to compensate property owners for takings is also a sovereign function, not a market transaction." Id. at 1327. In Beg, then, the government expropriated the plaintiff's land, agreed to compensate him, and then reneged.
This case is materially indistinguishable. So we must conclude that AFGC's suit is "based upon" Angola's expropriation. That's so for two reasons.
First, in both Beg and AFGC's case, the government took property and then reneged on its deal with the plaintiff to compensate the plaintiff for having taken the property. So while AFGC's claims and its theory of the case partly turn on Angola's breach of the settlement agreement, the settlement agreement was the means of compensating AFGC for the property Angola allegedly expropriated. Id. (explaining that a contract offer was a method of determining whether or how to compensate a property owner for a taking). As AFGC's complaint explains, the agreement was reached to resolve the "dispute relating to Angola's unlawful seizure and expropriation of assets and real properly lawfully owned by AFGC and its Angolan subsidiaries[.]" Doc. 50 ¶ 2. The settlement agreement, according to AFGC, was designed to remedy the injury caused by Angola's expropriation. Id. ¶ 58. ("Angola agreed to pay AFGC USD 47.5 million in exchange for and as compensation for the AFGC Angolan Assets, in addition to other negotiated terms.") (emphasis added). We don't see how this scenario is distinguishable from the one in Beg-the government took property and then reneged on a contract to compensate. AFGC's attempts to distinguish Beg are unpersuasive. For instance, while AFGC argues that Beg didn't make a contract claim in his complaint, our precedent-which we must follow-makes clear that the precise language Beg used didn't matter. Beg, 353 F.3d at 1327 ("Beg contends that the Punjabi regional government's agreement to compensate him is the equivalent of a contract and, therefore, is commercial activity. This analogy is not persuasive.").
Second, in both Beg and AFGC's case, the plaintiff was "actually injured" by the expropriation-not the breach. See id. In both cases, the agreement was the means of redressing an injury previously inflicted. When we determine what act a lawsuit is "based upon," we must "zero[] in on the . . . acts that actually injured [the plaintiff]." Sachs, 577 U.S. at 35. Here, the breach of contract did not injure AFGC; the expropriation did. Angola's obligation under the settlement agreement was to remedy the expropriation injury. Cf. France.com, Inc. v. French Republic, 992 F.3d 248, 253 (4th Cir. 2021) (explaining that the conduct that "actually injured" the plaintiff was the seizure of France.com, not the later use of the website for tourism).
Although AFGC characterizes the initial expropriation as irrelevant to its theory of the case, its complaint tells a different story. AFGC doesn't just seek $47.5 million in compensatory damages; it seeks $142.5 million-triple the amount of its alleged settlement agreement. To be sure, AFGC says in one paragraph that the extra $95 million is an estimation of "further consequential damages . . . resulting from the breach."
But allegations earlier in the complaint belie that characterization. Paragraph 25 brings the reader back to the negotiating table between AFGC and Angola:
25. AFGC explained at that point how it had arrived at its compromise settlement offer of USD 55 million, as stated in the February 1 Letter. As related to Dr. Rodrigues Neto and her colleague, the foreseeable and actual loss to AFGC arising from Angola's seizure of the AFGC Assets, if AFGC was not properly and promptly compensated by Angola, was approximately USD 95 million. This USD 95 million amount is comprised of a continuing loss of shareholder value at USD 72 million (8 million issued shares valued at the time of merger at USD 10/share, then trading at USD 1/share), plus the loss of new and pending investment opportunities into AFGC, lost reve-nue/profits, reputational damage and significant fees and expenses incurred by AFGC in prosecuting its claims against Angola.Id. ¶ 25. AFGC, in other words, is seeking $95 million in consequential damages resulting from the expropriation of its property. Here's how we know: AFGC and Angola hadn't reached a deal when AFGC says it suffered the $95 million loss. So Angola's later breach of the settlement agreement cannot be the source of the $95 million dollars in consequential damages because AFGC alleges that those damages had already occurred when it was negotiating the settlement agreement.
In sum, Beg controls this case.
We respectfully disagree with Judge Wilson's thoughtful dissent because we don't read Beg the same way as he does. Judge Wilson distinguishes Beg in two ways. He says first, that Beg didn't speak to the question at hand and second, that AFGC has alleged a harm that the Beg plaintiff didn't. Wilson Dissent at 8-9, and 11-12.
On the first point, as we've explained, we think Beg is materially indistinguishable. Both in Beg and here, the sovereign country took property from the plaintiff. Beg, 353 F.3d at 1325. Similarly, in both Beg and here, the plaintiff accepted compensation in principle for the taking. Id. And in both Beg and here, the sovereign country reneged on compensating the plaintiff. Id. The only difference, as we see it, is that Pakistan offered Beg land in exchange and Angola offered AFGC money. But we think that is a meaningless distinction here and doesn't alter the "gravamen" of the action. To be sure, Judge Wilson (rightly) points out that Beg didn't explicitly hold that the gravamen of the case was the expropriation. Wilson Dissent at 8. Rather in Beg, we held that expropriation was a sovereign rather than commercial act. Id. But in so doing, we necessarily decided that the gravamen of the case was the expropriation rather than the breach of contract. Id. Indeed, Beg's analysis would have looked quite different if the Beg panel thought the gravamen of the case was the breach of the contract with the plaintiff. If that were the case, the analysis would have resembled Judge Wilson's rather than the Beg panel's analysis of expropriation. But that's not what happened, and so, in our view, Beg is materially indistinguishable from our case.
As to Judge Wilson's second point-that AFGC was bargaining for something different than the Beg plaintiff because AFGC dismissed its lawsuit and ended its lobbying efforts-we read AFGC's complaint differently. Wilson Dissent at 11. When we "zero[] in" on the gravamen-the basis-of the suit, it is for expropriation. Sachs, 577 U.S. at 35. AFGC's harm stems from the expropriation, not the breach, because AFGC admits that the harm occurred before the contract was formed-in other words, from the expropriation, not from the breach of the contract. Nor does AFGC assert that it forewent lobbying against Angola or dismissed its lawsuit because of Dr. Neto's alleged agreement (nor could it, given that AFGC promised to do both only upon payment in full, which it never received). So AFGC hasn't explained how the alleged breach of the contract was what "actually injured" it.
AFGC tries to parry this conclusion with five responses. None is persuasive.
One, AFGC says that the reason that Angola entered the contract isn't relevant. The Supreme Court, AFGC continues, rejected the proposition that a court should enquire into the "purpose" of a transaction to determine whether the transaction is commercial. That's true, but it's not relevant here.
Indeed, in Republic of Argentina v. Weltover, Inc., 504 U.S. 607 (1992), the Supreme Court explained that the purpose-or the why-of an act was irrelevant for the second-order question of whether an act is commercial. Id. at 614. But that analysis doesn't answer our first-order question: whether an act is the basis of a lawsuit. And we don't have to inquire into the purpose of the settlement agreement to see that it amounts to an agreement to compensate AFGC for the expropriation. That's simply what the agreement is: a promise to pay for property taken in exchange for a release of claims for the taking of the property.
Two, AFGC relies on Guevara v. Republic of Peru, which held that Peru's offer of compensation for information on a fugitive was "commercial activity." 468 F.3d 1289, 1294 (11th Cir. 2006). But again, Guevara also dealt with only a single wrongful act: the breach of contract. So it cannot help us determine which act is the gravamen of AFGC's complaint.
Three, AFGC points to our Devengoechea v. Bolivarian Republic of Venezuela decision for the proposition that the conduct that actually injured AFGC is the breach of the contract. 889 F.3d at 1222. But we think that Devengoechea is also distinguishable. There, the plaintiff owned a collection of rare antiquities that had belonged to Simon Bolivar. Id. at 1217. The plaintiff allowed Venezuela to borrow the collection to see if it would purchase the collection from him. Id. Venezuela took the collection back to Venezuela and then refused to return it. Id. Devengoechea is distinguishable because the wrongful act in Devengoechea was breaking a bailment, not expropriation. Id. at 1221. Indeed, Venezuela came into possession of the collection only because the plaintiff gave Venezuela the collection under a bailment arrangement. So "[e]ven Venezuela [had] not characterized itself as having exercised sovereign powers to obtain or retain the Collection." Id.
Four, the D.C. Circuit's decision in de Csepel v. Republic of Hungary is to similar effect. 714 F.3d 591 (D.C. Cir. 2013). There, during the Second World War, the defendant confiscated the plaintiff's valuable art. Id. at 595-96. But after the war, Hungarian museums and a university "arranged with representatives of the [plaintiff's] Heirs to retain possession . . . so that the works could continue to be displayed in Hungary." Id. Then, fifty years later, the plaintiffs demanded possession back and Hungary refused. Id.
The D.C. Circuit explained that "by entering into bailment agreements to retain possession of the expropriated artwork and later breaching those agreements by refusing to return the artwork, Hungary took affirmative acts beyond the initial expropriation to deprive the family of their property rights in the Collection." Id. at 600. In other words, by entering the bailment arrangements, Hungary's expropriation of the art ended, and Hungary effectively recognized that the plaintiffs owned the property, even though the property remained in Hungary's possession at that time. Id. So when Hungary refused to return the art, it wasn't continuing an expropriation, as that had already ended. Rather, it was breaching the subsequent bailment agreement. As in in Devengoechea, then, in de Csepel, the harm-the gravamen, the focus-was the breach of the bailment, not the expropriation. Here, though, Angola's failure to compensate AFGC did not go "beyond the initial expropriation," id.; it was part and parcel of it. Cf. Beg, 353 F.3d at 1327 (explaining that a government's failure to compensate a party for a taking and the taking itself are two sides of the same expropriation coin, which happens to be a sovereign activity). Indeed, the expropriation never ended here.
Five, AFGC relies on Petersen Energia Inversora S.A.U. v. Argentine Republic and YPF S.A., 895 F.3d 194 (2d Cir. 2018). In Petersen, YPF was a company that was wholly owned by Argentina. Id. at 199. Argentina offered the public the opportunity to buy shares in the company-an initial public offering. Id. To encourage investment, Argentina and YPF amended YPF's bylaws to protect investors from "attempts by Argentina to renationalize the company." Id. For instance, if Argentina later reacquired 49% or more of YPF's shares, then Argentina was required to make a tender offer for the remaining shares. Id. at 199-200. Put differently, if Argentina became (close to) a majority shareholder of YPF, it was required to give the remaining owners a chance to exit the company. Id. Later, Argentina expropriated from a third party a controlling position in YPF but refused to make a tender offer to the remaining shareholders, violating the bylaw. Id. at 202 (Argentina's Deputy Economy Minister describing as "fools . . . those who think that the State has to be stupid and buy everyone according to YPF's own law, respecting its by-law."). One shareholder sued, arguing that Argentina had breached its promise to comply with the bylaws. Id. at 203. Argentina, of course, countered that the suit was for expropriation. Id.
The Second Circuit concluded that the plaintiff's harm was the breach of the tender-offer provision, not the expropriation. Id. at 207. This conclusion is unsurprising, given that none of the plaintiff's property was expropriated (only a third party's property was expropriated). So Petersen doesn't change our conclusion because it involved only one harm to the plaintiff-the breach of the bylaws.
***
To recap, if a sovereign power forms a contract with a private party and then reneges on it, the sovereign power is engaging in commercial activity-even if the way that the sovereign reneges is through what may look like expropriation. So if a sovereign power accepts performance by the private party and then refuses to perform, perhaps it could be said to have expropriated the performance, but the gravamen of the subsequent lawsuit is for breach of contract, regardless. Cf. Devengoechea, 889 F.3d at 1222 (retaining possession of artifacts); de Csepel, 714 F.3d at 600 (same).
On the other hand, when a sovereign power expropriates property and promises to repay the property holder but then breaches that promise, we think that Beg compels the conclusion that the gravamen of the lawsuit is the expropriation. "But even if we thought [Beg] wrong, the prior panel precedent rule is not dependent upon a subsequent panel's appraisal of the initial decision's correctness." Smith v. GTE Corp., 236 F.3d 1292, 1301-02 (11th Cir. 2001) (cleaned up). Given that we must follow Beg, whether we agree with it just isn't relevant.
Having identified the gravamen of the suit, we next determine whether the expropriation satisfies the FSIA's commercial-activity exception. It does not.
C. Angola's expropriation and failure to compensate AFGC were sovereign acts not taken "in connection with a commercial activity."
Given our conclusion that AFGC's suit is "based upon" Angola's expropriation and failure to compensate, our prior precedent compels the conclusion that these actions are sovereign. See Beg, 353 F.3d at 1326 ("The power of eminent domain is a sovereign power.") (citing United States v. Carmack, 329 U.S. 230, 236-37 (1946)); id. at 1327 ("Determining whether or how to compensate property owners for takings is also a sovereign function, not a market transaction.").
Under the FSIA, an activity is "commercial activity" if it is "either a regular course of commercial conduct or a particular commercial transaction or act." Devengoechea, 889 F.3d at 1220 (quoting 28 U.S.C. § 1603(d)). To decide whether an activity is commercial, we look at "whether the particular actions that the foreign state performs (whatever the motive behind them) are the type of actions by which a private party engages in 'trade and traffic or commerce.'" Weltover, 504 U.S. at 614 (quoting Black's Law Dictionary 270 (6th ed. 1990)). Assessing the commercial character of an act "is a question of behavior, not motivation." Nelson, 507 U.S. at 360 .
A country engages in commercial activity "only where it acts in the manner of a private player within the market" or "where it exercises only those powers that can also be exercised by private citizens, as distinct from those powers peculiar to sovereigns." Id. (internal quotation marks omitted). On the other side of the same coin, "[p]ublic acts . . . require sovereign power and thus cannot be performed by a private party." Beg, 353 F.3d at 1325 (citing Wel-tover, 504 U.S. at 614-16). Public acts "make use of the state's sovereign authority." Id.
Having decided the gravamen of the lawsuit is the expropriation, the answer is straightforward under our precedent: Angola's acts are not commercial activity. Expropriations and compensation for expropriations are sovereign functions. Beg, 353 F.3d at 1326 ("The power of eminent domain is a sovereign power."); id. at 1327 (compensating for a taking is a "sovereign function"); Garb v. Republic of Poland, 440 F.3d 579, 586 (2d Cir. 2006). ("[E]xpropria-tions . . . do not fall within the 'commercial activity' exception of the FSIA [because] [e]xpropriation is a decidedly sovereign-rather than commercial-activity.").
AFGC's authorities are well-taken but misplaced because not one of its slew of cases involves the use of a contract to compensate a plaintiff for the injuries caused by a foreign country's unquestionably sovereign acts. More importantly, not one of its cases involves the use of a contract where the contract itself amounts to an execution of a sovereign function. See, e.g., Hond. Aircraft Registry v. Gov't of Hond., 129 F.3d 543, 548-49 (11th Cir. 1997) (breach of contract to pay for "goods and services"-a "marketplace activity" available to "private persons"); Guevara, 468 F.3d at 1301 ("[T]he oldest American decision recognizing a commercial activities exception to sovereign immunity involved a reward for capture."); Weltover, 504 U.S. at 615 (explaining that "[t]he commercial character" of the bonds was "confirmed by the fact that they [were] in almost all respects garden-variety debt instruments[]"); Deven-goechea, 889 F.3d at 1221-22 ("Nothing about [breaching a bailment agreement] is uniquely or peculiarly sovereign in nature.").
This is not a case that involves the use of a contract to achieve a sovereign function. Cf. Weltover, 504 U.S. at 614-15 ("[A] contract to buy army boots or even bullets is a 'commercial' activity, because private companies can similarly use sales contracts to acquire goods."). Rather, this is a case in which the formation and breach of the contract is itself a sovereign function. See MOL, Inc. v. Peoples Republic of Bangladesh, 736 F.2d 1326, 1329 (9th Cir. 1984) (holding that when a country terminates "an agreement that only a sovereign could have made," that breach qualifies as the exercise of a sovereign function).
Contrary to AFGC's suggestion, rather than "look[ing] through" the alleged settlement agreement to see whether a sovereign act underlies it, we need look at only the agreement. In its complaint, AFGC characterized the agreement as a contract to compensate it for the expropriation it suffered. That is a quintes-sentially a sovereign act because private parties are generally not in the business of providing compensation for expropriated property. Beg, 353 F.3d at 1327.
We do not deal with a situation in which a private party exercises the power of eminent domain through a grant of delegated authority. See, e.g., Ga. Power Co. v. Sanders, 617 F.2d 1112 (5th Cir. 1980) (en banc).
As we've noted, we respectfully disagree with Judge Wilson that Beg is distinguishable. Judge Wilson distinguishes Beg because there, the plaintiff sought compensation for expropriated property, while in this case, AFGC allegedly accepted compensation in exchange for dismissing its lawsuit, giving up its claims to land, and ceasing its lobbying efforts. Wilson Dissent at 11. But that analysis misplaces the gravamen of the lawsuit, which as we've explained, is the failure to compensate for the expropriation, as evidenced by the fact that all $95 million in damages to which Dr. Neto allegedly agreed occurred before Angola ever breached the alleged contract. In short, the gravamen of the lawsuit-as our precedent requires us to view it-is the failure to compensate for expropriation, not the settling of the lawsuit. And that is a sovereign act. Id.
IV. CONCLUSION
For these reasons, we affirm the district court's order dismissing AFGC's complaint for lack of subject-matter jurisdiction.
AFFIRMED.
WILSON, Circuit Judge, dissenting:
I respectfully dissent. In my view, plaintiff Africa Growth Corporation's (AFGC) case against the Republic of Angola is based upon a breach of a contract, which in this case falls within the Foreign Sovereign Immunity Act's (FSIA) commercial activity exception to immunity from suit.
As the majority notes, under the FSIA, "a foreign state shall be immune from the jurisdiction of the courts of the United States and of the States" unless an identified exception applies. 28 U.S.C. § 1604. One such exception is a claim based on "commercial activity." Id. § 1605(a)(2). Under that exception, foreign states are not immune from the jurisdiction of federal or state courts if an action is 1) "based . . . upon an act" that took place "outside the territory of the United States," 2) "in connection with a commercial activity of the foreign state elsewhere," and 3) caused "a direct effect in the United States." Id. (emphasis added); see also Devengoechea v. Bolivarian Republic of Venezuela, 889 F.3d 1213, 1224 (11th Cir. 2018). I will address my views regarding how each of these elements apply to this particular case in turn.
As the majority rightly notes, while the exception in 28 U.S.C. § 1605(a)(2) describes other scenarios related to commercial activity that would bar immunity for foreign states, they are not invoked by AFGC. See Maj. Op., at 7 n.4.
I. "Based Upon"
I diverge from the majority on the initial question: What act is AFGC's action based upon? The majority's analysis comes to the conclusion that AFGC's action in the district court is "based upon" the expropriation of its property by the Angolan government-a decidedly sovereign activity that would be immune from suit under the FSIA. My analysis reaches a different conclusion. In my view, the action is "based upon" a breach of contract, which as I will describe below, is a commercial activity subject to the jurisdiction of our courts.
The initial element in our FSIA commercial activity framework also requires a determination that the act at issue took place "outside the territory of the United States." 28 U.S.C. § 1605(a)(2). I agree with the majority that regardless of whether the act is based on the expropriation of property or the breach of a contract, the act took place outside of the territory of the United States. Maj. Op., at 7 n. 5.
Three questions guide me to the conclusion that this case is "based upon" a breach of contract. First, what does it mean to be "based upon" in the context of 28 U.S.C. § 1605(a)(2)? Second, what are the relevant facts of this particular dispute? And third, how does the answer to the first question apply to that of the second?
Regarding the first question, the Supreme Court has provided guidance as to the meaning of "based upon," concluding that "the phrase is read most naturally to mean those elements of a claim that, if proven, would entitle a plaintiff to relief under his theory of the case." Saudi Arabia v. Nelson, 507 U.S. 349, 357 (1993). Put another way, what forms the "basis," "foundation," and "gravamen" of the complaint? OBB Personenverkehr AG v. Sachs, 577 U.S. 27, 33-34 (2015) (quoting Nelson, 507 U.S. at 356-57).
As an example, in Nelson, plaintiff Scott Nelson contracted to work at a state-run hospital in Saudi Arabia. 507 U.S. at 351-52. According to the complaint, agents of the Saudi Government later arrested, imprisoned, and tortured him. Id. at 352-53. After his release, he and his wife Vivian attempted to sue the Saudi Government for various intentional torts, negligent failure to warn, and derivative injuries. Id. at 353-54. In their argument to the Supreme Court, the plaintiffs argued that their claims were based upon the recruitment and hiring activity conducted by the Saudi Government and that there was a "sufficient nexus between those commercial activities and the wrongful acts that had allegedly injured [them]." Id. at 355. The Court roundly rejected that argument and found the Saudi Government was entitled to immunity under the FSIA. The Court explained:
In this case, the Nelsons have alleged that petitioners recruited Scott Nelson for work at the hospital, signed an employment contract with him, and subsequently employed him. While these activities led to the conduct that eventually injured the Nelsons, they are not the basis for the Nelsons' suit. Even taking each of the Nelsons' allegations about Scott Nelson's recruitment and employment as true, those facts alone entitle the Nelsons to nothing under their theory of the case. The Nelsons have not, after all, alleged breach
of contract . . . but personal injuries caused by petitioners' intentional wrongs and by petitioners' negligent failure to warn Scott Nelson that they might commit those wrongs. Those torts, and not the arguably commercial activities that preceded their commission, form the basis for the Nelsons' suit.Id. at 358.
With the meaning of "based upon" established, I move now to the second question: What are the relevant facts of this particular dispute? The majority does a sound job of laying out these facts, but I will repeat the relevant parts from AFGC's amended complaint for ease of reference.
As the majority rightly reminds us, Maj. Op., at 2 n.1, because this is an appeal of the district court's order granting Angola's motion to dismiss, we are required to accept all AFGC's well-pleaded allegations as true. Kennedy v. Floridian Hotel, Inc., 998 F.3d 1221, 1230 (11th Cir. 2021).
In 2015, AFGC (through its wholly-owned subsidiaries) acquired various pieces of real estate and commercial property in Luanda, Angola. A year later, the Angolan Government unlawfully seized and expropriated AFGC's assets in the country. AFGC subsequently filed suit in the United States District Court of the District of Columbia seeking to recover 1) the fair market value of and rents derived from the stolen properties, 2) lost profits from those properties, 3) damages from investment opportunities that were lost as a result of the expropriation, 4) damages from AFGC's tarnished reputation, and 5) treble damages under civil RICO statutes.
The parties eventually agreed to send representatives to Lisbon, Portugal for the purpose of settling the dispute. Negotiations went back and forth as to the amount Angola would pay AFGC, but (according to AFGC) the parties eventually came to an agreement, which an AFGC representative reduced to writing. That agreement provided that Angola would deposit 47.5 million dollars into the Florida bank account of AFGC's attorneys within 15 days of the meeting. As consideration, AFGC would waive and relinquish its rights and claims to its properties in Angola, withdraw its pending lawsuit in the District of Columbia, and end its global lobbying efforts against Angola.
AFGC asked the Angolan representative to sign the agreement; however, while the representative assured AFGC that the agreement was valid and had the approval of the Angolan Government, she stated that her government would memorialize the agreement in a more formal writing and reconvene in Lisbon the following week to execute it. The subsequent meeting never happened, Angola never signed the formal agreement, and Angola never paid AFGC.
AFGC then filed this new lawsuit in the United States District Court for the Southern District of Florida. Its claims are straightforward: breach of contract (Count I) and unjust enrichment (Count II)-both premised on the formation of a settlement agreement that Angola subsequently breached.
The facts in place, I turn now to the third question: Applying the meaning of "based upon" in 28 U.S.C. § 1605(a)(2) to the facts at hand, what forms the foundation of AFGC's complaint? That is, what are the "elements of [AFGC's] claim that, if proven, would entitle [it] to relief under [its] theory of the case"? Nelson, 507 U.S. at 357; see also Devengoechea, 889 F.3d at 1222 .
It seems to me that AFGC's claims-breach of contract and unjust enrichment-are based upon the alleged breach of a settlement agreement, rather than the alleged expropriation that brought the parties to the negotiation table. This becomes clearer once we consider what AFGC would and would not have to prove in order to obtain relief should it proceed to trial. Take AFGC's breach of contract claim: AFGC would have to prove that 1) the parties formed a valid contract-whereby (according to the amended complaint) Angola committed to pay 47.5 million dollars in consideration for AFGC agreeing to waive all rights and claims to its Angolan properties (regardless of how rightful those rights and claims were), withdraw its pending lawsuit in the District of Columbia, and end its lobbying efforts against the country; 2) that Angola breached that contract; and 3) that AFGC suffered damages as a result of Angola's breach. AFGC would not have to prove that Angola actually seized and expropriated its property unlawfully. Indeed, whether or not Angola wrongfully took AFGC's property is irrelevant to AFGC's claims. While AFGC was surely harmed by Angola's alleged expropriation, it was also separately harmed by Angola's breach of contract. Yet, AFGC only brings suit for the latter harm, so it is the latter harm that its suit is based upon.
To briefly parrot the Supreme Court, while Angola's seizure and expropriation "led to the conduct that eventually injured [AFGC], they are not the basis for [AFGC's] suit. Even taking each of [AFGC's] allegations about [Angola's seizure and expropriation] as true, those facts alone entitle [AFGC] to nothing under [its] theory of the case." Nelson, 507 U.S. at 358. Rather, what entitles AFGC to relief-what it has alleged and what forms the basis of its suit-is that a settlement agreement was created, Angola breached that agreement, and AFGC was harmed as a result.
To reach the opposite conclusion that AFGC's lawsuit is based on the expropriation of its property, the majority relies heavily on, and provides a thoughtful analysis of, our case in Beg v. Islamic Republic of Pakistan, 353 F.3d 1323 (11th Cir. 2003). However, Beg is distinguishable from the dispute before us today.
According to the allegations in Beg, the Government of Pakistan expropriated the plaintiff's land in the Punjab region of the country. 353 F.3d at 1324. Later, the Pakistani Government sent an official to the United States to offer the plaintiff an alternative parcel of land, which the plaintiff accepted. Id. Unfortunately for the plaintiff, the Supreme Court of Pakistan determined the Government of Punjab did not have good title to the new parcel and rejected the plaintiff's claims of title. Id. The plaintiff subsequently sought compensation through litigation in United States federal court. Id.
Assessing the claims in Beg, we largely skipped over any analysis of what the plaintiff's suit was "based upon," focusing instead on the nature of the conduct at issue. See id. at 1325. In doing so, we held that Pakistan did not engage in commercial activity because the lawsuit involved the Government's power of eminent domain, a uniquely sovereign power. Id. at 1326. We further held that "[d]etermining whether or how to compensate property owners for takings is also a sovereign function, not a market transaction." Id. at 1327. The plaintiff argued that the Pakistani Government's agreement to compensate him was "the equivalent of a contract and, therefore, [was] a commercial activity." Id. We found this argument unpersuasive, emphasizing that "the dispositive issue in determining whether an activity is commercial is whether private actors could undertake this type of activity in a market." Id. (citing Republic of Argentina v. Weltover, 504 U.S. 607, 614 (1992)).
At first glance, it might appear that Beg controls here. After all, both involve a government's exercise of eminent domain and an eventual failure to pay the property's owner. However, we have to remind ourselves of where we are in the analysis. At this stage, all we are asking is, "What is the action based upon?" For this question, Beg offers little guidance. Indeed, Beg provides almost no input on how to determine what an action is based upon when a sovereign power breaks an agreement to compensate someone for the use of its eminent domain power. Rather, Beg stands only for the proposition that when a party is ultimately seeking compensation for a foreign government's takings abroad, its claims are based upon an act (whatever that act might be) that is sovereign in nature.
So, while Beg might have much to say about whether the conduct at issue is commercial or sovereign, it has little to say about what the conduct at issue is. For that, we must look to cases like Nelson and Sachs, which task us with determining what elements a plaintiff must prove to obtain relief. Nelson, 507 U.S. at 357; Sachs, 577 U.S. at 33-34. Here, AFGC must prove that an agreement existed, Angola breached that agreement, and AFGC was harmed by that breach. Thus, it is the breach of the settlement agreement that AFGC's action is "based upon."
Of course, this still leaves the question of whether Angola's conduct in breaching the parties' agreement was commercial or sovereign.
II. Commercial or Sovereign?
As noted above, 28 U.S.C. § 1605(a)(2) excepts from immunity cases "in which the action is based . . . upon an act . . . in connection with a commercial activity of the foreign state elsewhere." (emphasis added). The FSIA provides that a commercial activity is "either a regular course of commercial conduct or a particular commercial transaction or act. The commercial character of an activity shall be determined by reference to the nature of the course of conduct or particular transaction or act, rather than by reference to its purpose." 28 U.S.C. § 1603(d). Therefore, "the question is not whether the foreign government is acting with a profit motive or instead with the aim of fulfilling uniquely sovereign objectives." Weltover, 504 U.S. at 614. "Rather, the issue is whether the particular actions that the foreign state performs (whatever the motive behind them) are the type of actions by which a private party engages in 'trade and traffic or commerce.'" Id. Here, I believe Angola's actions are "commercial" within the meaning of the FSIA.
As a starting point, we have previously held that a foreign state's breach of a contract with a private party can qualify as commercial activity under the FSIA. In Guevara v. Republic of Peru, the issue we faced was "whether a foreign state's offer of a reward in return for information enabling it to locate and capture a fugitive" fell within the commercial activity exception. 468 F.3d 1289, 1292 (11th Cir. 2006). The district court found that "offering a reward for capturing a fugitive involves uniquely sovereign objectives," and was not a commercial activity. Id. at 1295. On appeal, we reversed and reasoned that "[t]he location and capture of a fugitive . . . may be a sovereign act, but that is not what this case is about." Id. at 1298. We noted that the plaintiff was not seeking to compel Peru to capture the fugitive; he was only seeking to enforce Peru's promise that it would pay 5 million dollars for information leading to his capture. Id. at 1298-99. We held that "[t]he underlying activity at issue-the exchange of money for information-is 'commercial in nature and of the type negotiable among private parties.'" Id. at 1299.
Similarly, here, while taking property and "[d]etermining whether or how to compensate property owners" for that taking may be a sovereign act, Beg, 353 F.3d at 1327, that is not what this case is about. Rather, this case is about Angola entering into and breaching a contract with AFGC, and this is precisely the type of action "by which a private party engages in 'trade and traffic or commerce.'" Weltover, 504 U.S. at 614.
Now, here is where Beg fits into the analysis and muddies the waters. However, heeding 28 U.S.C. § 1603(d)'s directive that the "commercial character of an activity shall be determined by reference to the nature of the . . . particular transaction or act, rather than by reference to its purpose," I believe there is an important distinction between that case and the dispute before us. In Beg, the plaintiff argued that the Government of Pakistan reneged on a deal to compensate him directly for the land it expropriated. 353 F.3d at 1327. That is, the Pakistani Government took the plaintiff's land, promised to give him an alternative parcel of land in exchange, and then never did. Id. Here, however, the parties were bargaining for something else. AFGC, after having its property taken away, brought suit in the United States District Court for the District of Columbia and apparently started a lobbying effort against Angola. Angola then (allegedly) agreed to pay 47.5 million dollars in exchange for AFGC withdrawing its lawsuit, waiving any claims it thought it had to the Angolan property, and ending its lobbying efforts. So, while the nature of the act in Beg was a deal to pay the plaintiff for the property that was taken, the nature of the act here is a deal to pay the plaintiff for dropping its lawsuit, rights, and lobbying efforts. The former is an act of a sovereign; the latter is an act that I can imagine any private entity undertaking. And, in my view, digging any deeper into the purpose of Angola's action runs afoul of 28 U.S.C. § 1603(d).
For these reasons, I believe that AFGC's action is based upon Angola's breach of a settlement agreement, and I further believe that Angola's act was in connection with commercial activity.
This brings me to the final portion of this FSIA analysis, which requires a court to determine whether the act of the sovereign defendant caused a direct effect in the United States.
III. "Direct Effect" in the United States
As mentioned just above, the commercial activity exception only applies if the alleged act of the foreign sovereign "cause[d] a direct effect in the United States." 28 U.S.C. § 1605(a)(2); see also Weltover, 504 U.S. at 617-19. Based on the allegations in AFGC's amended complaint, this element is clearly met. According to AFGC, the contract it entered into with Angola required Angola to deposit 47.5 million dollars into an attorney trust bank account in the State of Florida, which would be held for the benefit of AFGC, a business incorporated in the State of Nevada. This allegation is sufficient to establish a direct effect in the United States. See Wel-tover, 504 U.S. at 617 ("Because New York was thus the place of performance for Argentina's ultimate contractual obligations, the rescheduling of those obligations necessarily had a 'direct effect' in the United States: Money that was supposed to have been delivered to a New York bank for deposit was not forthcoming.").
IV. Conclusion
For the above reasons, I would hold that the commercial activities exception applies, and that the Republic of Angola is not immune from this particular suit. Therefore, I respectfully dissent.
[*] Honorable Virginia M. Covington, United states District Judge for the Middle District of Florida, sitting by designation.