United States v. Hoskins

See case details

UNITED STATES OF AMERICA, Appellant, v. LAWRENCE HOSKINS, Defendant - Appellee.

Docket No. 16-1010-cr

UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT

August 24, 2018

August Term, 2016
(Argued: March 2, 2017)

Before: KATZMANN, Chief Judge, POOLER and LYNCH, Circuit Judges.

In this interlocutory appeal from a district court‘s dismissal of part of one count in a multi-count indictment, we are asked to decide whether the government may employ theories of conspiracy or complicity to charge a defendant with violating the Foreign Corrupt Practices Act (“FCPA“), even if he is not in the categories of persons directly covered by the statute.

We first decide that we have jurisdiction over the appeal under 18 U.S.C. § 3731. That statute, as amended, permits interlocutory appeals by the government when a part of a count of an indictment is dismissed.

We then address the merits of the district court‘s dismissal of the government‘s FCPA charges. We determine that the FCPA defines precisely the categories of persons who may be charged for violating its provisions. The statute also states clearly the extent of its extraterritorial application. Because we agree with the district court that the FCPA‘s carefully-drawn limitations do not comport with the government‘s use of the complicity or conspiracy statutes in this case, we AFFIRM the district court‘s ruling barring the government from bringing the charge in question. We REVERSE the district court‘s holding on the Second Object of the Conspiracy, because the government‘s intention to prove that Hoskins was an agent of a domestic concern places him squarely within the terms of the statute and takes that provision outside our analysis on the other counts.

Affirmed in part and Reversed in part.

JUDGE LYNCH concurs in the opinion and files a separate concurring opinion.

SANGITA K. RAO, Department of Justice (Michael J. Gustafson, First Assistant United States Attorney, Leslie R. Caldwell, Assistant Attorney General, David E. Novick, Assistant United States Attorney, Sung-Hee Suh, Deputy Assistant Attorney General, Andrew Weissmann, Chief, Fraud Section, Daniel S. Kahn and Jeremy Sanders, Attorneys, Fraud Section, on the brief), Washington, D.C., for Appellant United States of America.

CHRISTOPHER J. MORVILLO, Clifford Chance US LLP (Daniel Silver and Benjamin Peacock, on the brief), New York, N.Y., for Defendant-Appellee Lawrence Hoskins.

Ira M. Feinberg and Derek Musa, on the brief, Hogan Lovells US LLP, New York, N.Y., for Amicus Curiae the New York Council of Defense Lawyers in support of Defendant-Appellee Lawrence Hoskins.

Jonathan Bach and Adam S. Gershenson, on the brief, Cooley LLP, New York, N.Y., for Amicus Curiae the New York Council of Defense Lawyers in support of Defendant-Appellee Lawrence Hoskins.

POOLER, Circuit Judge:

In this case, we are asked to decide whether the government may employ theories of conspiracy or complicity to charge a defendant with violating the Foreign Corrupt Practices Act (“FCPA“), even if he is not in the category of persons directly covered by the statute.¹ We determine that the FCPA defined precisely the categories of persons who may be charged for violating its provisions. The statute also stated clearly the extent of its extraterritorial application.

The FCPA establishes three clear categories of persons who are covered by its provisions: (1) Issuers of securities registered pursuant to 15 U.S.C. § 78l or required to file reports under Section 78o(d), or any officer, director, employee, or agent of such issuer, or any stockholder acting on behalf of the issuer, using interstate commerce in connection with the payment of bribes, 15 U.S.C. § 78dd-1; (2) American companies and American persons using interstate commerce in connection with the payment of bribes, 15 U.S.C. § 78dd-2; and (3) foreign persons or businesses taking acts to further certain corrupt schemes, including ones causing the payment of bribes, while present in the United States, 15 U.S.C. § 78dd-3.

Because we agree with the district court that the FCPA‘s carefully-drawn limitations do not comport with the government‘s use of the complicity or conspiracy statutes in this case, we AFFIRM the district court‘s ruling barring the government from bringing the charge in question. We REVERSE the district court‘s holding on the Second Object of the Conspiracy, because the government‘s intention to prove that Hoskins was an agent of a domestic concern places him squarely within the terms of the statute and takes that provision outside our analysis on the other counts.

BACKGROUND

I. The Allegations

The government alleges that several defendants, including Hoskins, were part of a scheme to bribe officials in Indonesia so that their company could secure a $118 million contract from the Indonesian government. Hoskins worked for Alstom S.A. (“Alstom“), a global company headquartered in France that provides power and transportation services During the relevant time, which was from 2002 to 2009, Hoskins was employed by Alstom‘s UK subsidiary, but was assigned to work with another subsidiary called Alstom Resources Management, which is in France.

The alleged bribery scheme centers on Alstom‘s American subsidiary, Alstom Power, Inc. (“Alstom U.S.“), headquartered in Connecticut. The allegations are that Alstom U.S. and various individuals associated with Alstom S.A. retained two consultants to bribe Indonesian officials who could help secure the $118 million power contract for the company and its associates. Hoskins never worked for Alstom U.S. in a direct capacity. But the government alleges that Hoskins, while working from France for Alstom Resources Management, was “one of the people responsible for approving the selection of, and authorizing payments to, [the consultants], knowing that a portion of the payments to [the consultants] was intended for Indonesian officials in exchange for their influence and assistance in awarding the [contract.]” Third Superseding Indictment (hereinafter “Indictment“) ¶¶ 3, 8.

The government alleges that several parts of the scheme occurred within the United States. The indictment alleges that one of the consultants kept a bank account in Maryland.² In some cases, funds for bribes allegedly were paid from bank accounts held by Alstom and its business partners in the United States, and deposited in the consultant‘s account in Maryland, for the purpose of bribing Indonesian officials. The indictment also states that several executives of Alstom U.S. held meetings within the United States regarding the bribery scheme and discussed the project by phone and email while present on American soil.

The government concedes that, although Hoskins “repeatedly e-mailed and called . . . U.S.-based coconspirators” regarding the scheme “while they were in the United States,” Hoskins “did not travel here” while the bribery scheme was ongoing. Appellant‘s Br. at 7.

II. The Indictment

The Third Superseding Indictment, the operative one in the case, brings twelve counts against Hoskins. This appeal concerns the first seven counts of the indictment.

Count one charges Hoskins with conspiring to violate the FCPA. It alleges that Hoskins is liable because he was an agent of Alstom U.S., an American company, and, in that capacity, committed acts that violated the statute. It also alleges that, independently of his agency relationship with an American company, Hoskins conspired with the company and its employees, as well as foreign persons, to violate the FCPA, and also aided and abetted their violations. The Count focuses on two objects of the conspiracy, which correspond to two provisions of the FCPA that Hoskins supposedly violated as an accomplice and also conspired to violate. The first of the two FCPA provisions prohibits American companies and American persons, as well as their agents, from using interstate commerce in connection with the payment of bribes. 15 U.S.C. § 78dd-2. The second prohibits foreign persons or businesses from taking acts to further certain corrupt schemes, including ones causing the payment of bribes, while present in the United States. 15 U.S.C. § 78dd-3.

Counts two through seven charge substantive violations of the FCPA, focusing on particular wire transfers from Alstom U.S.‘s bank account to the consultants’ accounts. These counts all charge Hoskins with violations of 15 U.S.C. § 78dd-2. The counts allege that Hoskins violated this provision as “an agent” of an American company or person, and also “by aiding and abetting” such a company or person.³

III. Proceedings Below

Before the district court Hoskins moved for dismissal of the first count of the indictment. See United States v. Hoskins, 123 F. Supp. 3d 316 (D. Conn. 2015). He noted that the FCPA prescribes liability only for narrowly-circumscribed groups of people—American companies and citizens, and their agents, employees, officers, directors, and shareholders, as well as foreign persons acting on American soil. Hoskins argued that the government could not circumvent those limitations by charging him with conspiring to violate the FCPA, or aiding and abetting a violation of it, if he did not fit into one of the statute‘s categories of defendants. He thus moved for dismissal of Count One, as it charged that he was liable even if he did not fit into one of the statute‘s categories. Hoskins, 123 F. Supp. 3d at 317, 319.

The government filed a closely-related motion in limine regarding Counts Two through Seven. Id. at 317. The motion sought to preclude Hoskins from arguing at trial that he could only be convicted of violating the statute under a conspiracy or aiding-and-abetting theory if the government first proved that he fell within one of the FCPA‘s enumerated categories of defendants.

The district court granted Hoskins‘s motion in part and denied the government‘s motion. See id. at 327. The court explained that, under Gebardi v. United States, 287 U.S. 112 (1932), “where Congress chooses to exclude a class of individuals from liability under a statute, the Executive may not override the Congressional intent not to prosecute that party by charging it with conspiring to violate a statute that it could not directly violate.” Hoskins, 123 F. Supp. 3d at 321 (internal quotation marks and alterations omitted). Upon a thorough consideration of the text, structure, and legislative history of the FCPA, the district court concluded that “Congress did not intend to impose accomplice liability on non-resident foreign nationals who were not subject to direct liability” under one of the statute‘s provisions. Id. at 327.

The court thus dismissed Count One of the indictment to the extent that it sought to charge Hoskins with conspiring to violate Section 78dd-2 of the FCPA without demonstrating that Hoskins fell into one of the FCPA‘s enumerated categories. Id. The court also dismissed Count One to the extent it alleged that Hoskins conspired to violate Section 78dd-3, which prohibits acts “while in the territory of the United States,” because Hoskins had never entered the United States during the relevant period. Id. at 327 n.14 (“[I]f Congress intended to limit liability under this section to those within the territory of the United States, the Government cannot circumvent this intention by resort to the conspiracy statute.“). The district court denied Hoskins‘s motion in part, however, because the indictment charged him with conspiring to violate the FCPA, or aiding and abetting a violation, as an agent of an American company, a category covered by Section 78dd-2. Id. at 318 n.1, 327. The court also denied the government‘s motion in limine. Id. at 327.

The government appeals all of the district court‘s rulings.

DISCUSSION

I. Jurisdiction under 18 U.S.C. § 3731

Hoskins first argues that the court has no jurisdiction over this interlocutory appeal. In the government‘s view, the court has jurisdiction under 18 U.S.C. § 3731, which permits the United States certain interlocutory appeals in criminal cases. The statute reads as follows:

In a criminal case an appeal by the United States shall lie to a court of appeals from a decision, judgment, or order of a district court dismissing an indictment or information or granting a new trial after verdict or judgment, as to any one or more counts, or any part thereof, except that no appeal shall lie where the double jeopardy clause of the United States Constitution prohibits further prosecution.... The provisions of this section shall be liberally construed to effectuate its purposes.

18 U.S.C. § 3731. The previous version of Section 3731 did not include the phrase “or any part thereof,” which was added by Congress in 2002. The legislative history of the provision makes clear that, in making the change, Congress intended to broaden the scope of interlocutory appeals the government could bring:

This section clarifies that 18 U.S.C. § 3731 authorizes an appeal by the United States, consistent with the Double Jeopardy clause, whenever a district court enters an order dismissing or striking part of an indictment or information. . . . [The pre-2002 version of] the statute has generally been generously interpreted to allow government appeals, even when its literal language does not clearly extend to the case, such as where a district court has dismissed only a portion of a count such as a predicate act in a RICO count or an overt act in a conspiracy count. . . . However, one federal circuit has held that section 3731 does not permit any government appeals from the dismissal of only part of a count. See United States v. Louisiana Pacific Corporation, 106 F.3d 345 (10th Cir. 1997). In other cases, appellate review of orders dismissing predicate acts or overt acts has been denied where the dismissed acts could not themselves have been charged in separate counts. See United States v. Terry, 5 F.3d 874 (5th Cir. 1993); United States v. Tom, 787 F.2d 65 (2d Cir. 1986).

It is time to resolve these conflicting results definitively. The reach of section 3731 should clearly be extended to orders dismissing portions of counts. Although the Solicitor General, who must approve all appeals by the United States to a court of appeals, only seldom authorizes appeals from partial dismissals of counts in criminal cases, there is no reason not to permit the government to appeal when the issue involved is important and determined by the Solicitor General to be worthy of presentation to a higher court. Indeed, there are some cases where the dismissal of a predicate act or overt act may substantially weaken the government‘s ability to prove its case. The proposed amendment would therefore insert the phrase “or any part thereof” in section 3731 so as to make clear that dismissals of any part of a count are subject to appeal by the United States in appropriate circumstances.

H.R. Rep. No. 107-685, at 188 (2002) (Conf. Rep.), reprinted in 2002 U.S.C.C.A.N. 1120, 1140-41 (hereinafter “2002 Conference Report“). Following the 2002 amendment, the First and Ninth Circuits have recognized the statute‘s clear reach over government requests to appeal dismissals of portions of counts in an indictment. United States v. DeCologero, 364 F.3d 12, 20 (1st Cir. 2004) (“The first paragraph of the statute allows (in pertinent part) an appeal from a district court‘s dismissal of an indictment ‘as to any one or more counts, or any part thereof’ — the ‘any part’ language having been added in 2002 in part to resolve a circuit split . . . .” (quoting 18 U.S.C. § 3731 (2003))); United States v. Morales, 465 F. App‘x 734, 736 (9th Cir. 2012) (“The statute was amended in 2002 to permit appeals from any “order of a district court [. . . ] dismissing an indictment [...] or any part thereof.“).

Hoskins argues, based on our rulings in United States v. Margiotta, 662 F.2d 131 (2d Cir. 1981), and United States v. Tom, 787 F.2d 65 (2d Cir. 1986), that the court lacks jurisdiction to hear the appeal under Section 3731. Margiotta and Tom laid down the rule that “the Government may appeal when an order precludes consideration of an independent ground for a conviction.” Tom, 787 F.2d at 70 (internal quotation marks omitted). This rule permitted some appeals when a district court dismissed something less than a full count of an indictment. We explained that “an independent ground for a conviction” need not always be “formally pleaded as a separate count in the indictment,” Margiotta, 662 F.2d at 140, and that dismissal of a “ground for a conviction” would be appealable even if not set out as a separate count.

But, under the rule stated in Margiotta and Tom dismissal of a theory of liability did not always remove a separate ground for a conviction, and thus did not always give rise to a government appeal. As long as some path remained for the defendant to be convicted under a given charge, the appeal was impermissible. For example, in Margiotta, we did not allow the government to appeal from the district court‘s dismissal of the theory that the defendant was liable as a principal for violation of the Hobbs Act when he was also charged with aiding and abetting Hobbs Act violations. Id. at 141. We reasoned that “[a]ider and abettor activity . . . is . . . punishable . . . to the same extent as activity of a principal,” and so “the court‘s jury instruction concerning the Hobbs Act charges does not strike from the case an independent basis of liability.” Id. Similarly, in Tom, we dismissed an appeal from a district court‘s dismissal of some, but not all, allegations of racketeering acts that undergirded a charge under the Racketeer Influenced and Corrupt Organizations Act (“RICO Act“). 787 F.2d at 67, 71.

In light of Congress‘s 2002 amendments to Section 3731, Tom and Margiotta are no longer authoritative regarding appeals from dismissals of portions of indictments. First, Congress‘s statement that appeals may be taken from dismissal of “any part” of a count of an indictment plainly conflicts with the rules stated in those cases, which did not permit appeals from dismissals of some parts of indictments. Second, the 2002 Conference Report rejected the rule established in Tom. See 2002 Conference Report at 188. Given that Margiotta and Tom stated the same rule, Congress‘s displeasure with both cases can be understood from the Conference Report.

Under the revised version of Section 3731, the government‘s appeal in this case may go forward. The district court dismissed portions of the indictment charging Hoskins with conspiracy to violate 15 U.S.C. § 78dd–2, or violation of that provision as an accomplice, to the extent that the government could not also show Hoskins was an agent of an American company or person, or a director, employee, or stockholder of an American company. Hoskins, 123 F. Supp. 3d at 317, 327. Additionally, the court dismissed part of the indictment alleging that he conspired to violate 15 U.S.C. § 78dd–3, or violated that provision as an accomplice, because the government conceded that the defendant did not enter the United States during the relevant time period. Id. at 327 n.14. Although the court did not dismiss an entire count of the indictment, it dismissed two significant parts of a count. That suffices for an appeal, given Congress‘s statement that review may be sought after dismissal of “any part” of a count of an indictment.

Because we have jurisdiction under 18 U.S.C. § 3731 to review the dismissal of portions of Count One, we will exercise pendent appellate jurisdiction to review the district court‘s denial of the government‘s motion in limine. The doctrine of pendent appellate jurisdiction permits review of “all matters inextricably bound up with” an issue over which the court has jurisdiction. Lamar Advert. of Penn, LLC v. Town of Orchard Park, N.Y., 356 F.3d 365, 371 (2d Cir. 2004); see also United States v. Zabawa, 39 F.3d 279, 283 (10th Cir. 1994) (applying rules of pendent appellate jurisdiction to appeal taken under 18 U.S.C. § 3731). Because “the district court here denied” the government‘s motion in limine “for the very same reasons” that it granted in part Hoskins‘s motion to dismiss the indictment, the issues “are indeed inextricably intertwined with [the issues] over which we have appellate jurisdiction.” Lamar, 356 F.3d at 372. And because “[i]t is surely in the interest of judicial economy” to consider both of the motions in the same appeal, we will do so. Id.

II. The FCPA and the First Object of the Conspiracy

The central question of the appeal is whether Hoskins, a foreign national who never set foot in the United States or worked for an American company during the alleged scheme, may be held liable, under a conspiracy or complicity theory, for violating FCPA provisions targeting American persons and companies and their agents, officers, directors, employees, and shareholders, and persons physically present within the United States. In other words, can a person be guilty as an accomplice or a co-conspirator for an FCPA crime that he or she is incapable of committing as a principal?

A. Conspiracy Liability

For purposes of this appeal, we assume that Hoskins was neither an employee nor an agent of a domestic concern and therefore does not fall within the terms of the statute. But accomplice and conspiracy liability are generally not so limited. A get-away driver for a bank robbery team can still be prosecuted even though he has not “by force and violence . . . take[n] . . . from the person or presence of another . . . any property . . . belonging to . . . any bank.” 18 U.S.C. § 2113(a). As the common law has long recognized, persons who intentionally direct or facilitate the crimes physically executed by others must be held accountable for their actions. This recognition was effectuated by developing the doctrines of conspiracy and complicity, principles that are now codified in statutes. Under 18 U.S.C. § 2(a), a person who does not personally commit the acts constituting an offense is liable as a principal if he or she “aids, abets, counsels, commands, induces or produces” the commission of those acts by another. In addition, 18 U.S.C. § 371 punishes anyone who “conspire[s]” with another to commit the offense. Thus, by the plain language of the general statutes regarding conspiracy and accessorial liability—which nothing in the language of the FCPA purports to overrule or limit—if Hoskins did what the indictment charges, he would appear to be guilty of conspiracy to violate the FCPA and (as an accomplice) of substantive violations of that statute.

Conspiracy and complicity statutes do not cease to apply simply because a statute specifies particular classes of people who can violate the law. It is well established in federal criminal law that “[a] person . . . may be liable for conspiracy even though he was incapable of committing the substantive offense.” Salinas v. United States, 522 U.S. 52, 64 (1998). That principle was already deeply ingrained when the Supreme Court unanimously ruled in 1915 that persons not themselves bankrupt could be guilty of conspiring with someone who had declared bankruptcy to hide assets of the bankrupt‘s estate from the bankruptcy trustee, even if a non-bankrupt party could not be convicted of the principal offense. United States v. Rabinowich, 238 U.S. 78, 86 (1915). With respect to complicity, the same principal was so clearly entrenched as a matter of the common law of crimes that the Supreme Court saw no need to cite a particular precedent when it unanimously recognized in 1833 that someone who “procure[d], advise[d] and assist[ed]” a postmaster to remove from the mail and destroy a letter was guilty of violating, as an accomplice, a statute prohibiting postal employees from taking mail entrusted to them for delivery. United States v. Mills, 32 U.S. 138, 141 (1833).

Thus the firm baseline rule with respect to both conspiracy and complicity is that where the crime is so defined that only certain categories of persons, such as employees of a particular sort of entity, may commit the crime through their own acts, persons not within those categories can be guilty of conspiring to commit the crime or of the substantive crime itself as an accomplice.⁴

Longstanding principle and precedent thus reinforces what the plain language of the conspiracy and aiding and abetting statutes command.

B. The Affirmative-Legislative-Policy Exception

There is a narrowly circumscribed exception to this common-law principle. In certain cases it is clear from the structure of a legislative scheme that the lawmaker must have intended that accomplice liability not extend to certain persons whose conduct might otherwise fall within the general common-law or statutory definition of complicity. A classic illustration is statutory rape, which makes it a crime to have sexual relations with a person who is under a statutorily defined age of consent. Applying the literal definitions of accomplice liability, a youthful participant who voluntarily consents to the act would be guilty of rape as well, because he or she intentionally aided or solicited the commission of the criminal act. But the legislature, in criminalizing the conduct of the adult participant and not that of the juvenile, obviously conceptualized the under-age party as the victim of the crime, and not a co-participant. Despite the common-law recognition of conspiracy and accomplice liability, and of the general principle that one could be guilty as a conspirator or accomplice even if the statute were defined in such a way that one was not capable of committing it as a principal, the common-law courts had no difficulty in recognizing an exception in those circumstances. See, e.g., Regina v. Tyrell, [1894] 1 Q.B. 710.

Here the government concedes that the common-law principle of conspiracy liability admits of exceptions but argues that the FCPA falls outside those exceptions. Hoskins, by contrast, contends that the FCPA demonstrates “an affirmative Congressional intent to exclude certain persons from liability” under the statute. Appellee‘s Br. at 20 (emphasis omitted). The parties’ dispute focuses on two cases, Gebardi v. United States, 287 U.S. 112 (1932), and United States v. Amen, 831 F.2d 373 (2d Cir. 1987), and it is thus profitable to consider both in some detail.

1. Gebardi

In Gebardi, the Supreme Court considered a conviction under the Mann Act, a statute that imposes a penalty upon

any person who shall knowingly transport or cause to be transported, or aid or assist in obtaining transportation for, or in transporting, in interstate or foreign commerce any woman or girl for the purpose of prostitution or debauchery, or for any other immoral purpose.

287 U.S. at 118 (quoting 18 U.S.C. § 398 (1932)). The Mann Act criminalizes such transportation “with or without [the woman‘s] consent.” Id. The government convicted both a man and woman for conspiracy to violate the Mann Act, on the theory that the woman conspired to transport a person—herself—merely by consenting to the man‘s transportation of her.

The Supreme Court reversed the convictions. The Court first noted that the Mann Act plainly covered cases where “the woman consents to her own transportation,” rather than just cases where her transportation was forced, “[y]et it does not specifically impose any penalty upon her, although it deals in detail with the person by whom she is transported.” Id. at 119. Because it would be obvious that women would participate in many violations of the statute, but the statute discussed no punishment for the women, the Court concluded that Congress intended for the women not to be liable for at least some class of violations of the Act. In particular, the Court determined it could not “infer that the mere acquiescence of the woman transported was intended to be condemned by the general language punishing those who aid and assist the transporter.” Id. “The penalties of the statute are too clearly directed against the acts of the transporter” to support the view that Congress intended the woman always to be liable. Id.

Having decided that Congress intended to leave the woman unpunished when she merely acquiesced in her own illegal transportation, the Court next considered whether she could be convicted of conspiring to violate the statute in such circumstances. Id. at 119-23. The Court concluded that she could not. The Court emphasized, again, that “Congress set out in the Mann Act to deal with cases which frequently, if not normally, involve consent and agreement on the part of the woman to the forbidden transportation,” but that “this acquiescence . . . was not made a crime under the Mann Act itself.” Id. at 121. Consequently, the Court “perceive[d] in the failure of the Mann Act to condemn the woman‘s participation in those transportations which are effected with her mere consent, evidence of an affirmative legislative policy to leave her acquiescence unpunished.” Id. at 123. The Court explained that it was

a necessary implication of that policy that when the Mann Act and the conspiracy statute came to be construed together, as they necessarily would be, the same participation which the former contemplates as an inseparable incident of all cases in which the woman is a voluntary agent at all, but does not punish, was not automatically to be made punishable under the latter. It would contravene that policy to hold that the very passage of the Mann Act effected a withdrawal by the conspiracy statute of that immunity which the Mann Act itself confers.

Id. at 123. Because the defendant in Gebardi had merely consented to her transportation, the Court ruled that her conviction for conspiracy could not stand; and because she had not conspired to violate the Mann Act, her companion had no one with whom to conspire. Id. Both of their convictions for conspiracy were reversed. Id.

In determining that the woman in Gebardi was not liable as a conspirator because of Congress‘s “affirmative legislative policy” to leave her unpunished, id. at 123, the Gebardi Court distinguished its reasoning from an older common-law limitation on conspiracy liability—a rule widely known as Wharton‘s Rule. See id. at 121-22; see also Iannelli v. United States, 420 U.S. 770, 773-76 (1975) (discussing Wharton‘s Rule and identifying Gebardi as a case that had previously discussed it). Wharton‘s Rule states that “[a]n agreement by two persons to commit a particular crime cannot be prosecuted as a conspiracy when the crime is of such a nature as to necessarily require the participation of two persons for its commission,” such as dueling. Iannelli, 420 U.S. at 773 n.5 (quoting 1 R. Anderson, Wharton‘s Criminal Law and Procedure § 89, at 191 (1957)). The Court in Gebardi alluded to Wharton‘s Rule. See Gebardi, 287 U.S. at 122. But the Court stated that Wharton‘s Rule did not apply, because the Rule requires voluntary consent while “criminal transportation under the Mann Act may be effected without the woman‘s consent as in cases of intimidation or force.” Id. Consequently, the Court “d[id] not rest [the] decision upon [Wharton‘s Rule], nor upon the related one that the attempt is to prosecute as conspiracy acts identical with the substantive offense.” Id. at 122-23. Instead, the Court explicitly situated its ruling “upon the ground that we perceive in the failure of the Mann Act to condemn the woman‘s participation in those transportations which are effected with her mere consent, evidence of an affirmative legislative policy to leave her acquiescence unpunished.” Id. at 123.

2. Amen

We applied the reasoning of Gebardi in United States v. Amen, 831 F.2d 373 (2d Cir. 1987). In Amen, the Court considered the “continuing criminal enterprise” statute, 21 U.S.C. § 848, a provision “designed to reach the ‘top brass’ in the drug rings,” Garrett v. United States, 471 U.S. 773, 781 (1985), or, to put it differently, the “kingpin” in an enterprise. Amen, 831 F.2d at 382. A defendant was convicted on the theory that he conspired with, and aided and abetted, an enterprise‘s “kingpin,” even though the defendant himself was not the “kingpin.” Id.

The government conceded that the statute did not apply to an enterprise‘s employees. Id. at 381. It nevertheless attempted to distinguish between “mere employees and those who otherwise ‘help’ the kingpin,” and to argue that “non-employees who knowingly provide direct assistance to the head of the organization in supervising and operating the criminal enterprise can be . . . punished” for violating the “kingpin” statute under conspiracy and aiding-and-abetting theories. Id. at 381-82.

We explained, however, that the government‘s theory “lack[ed] support in legislative history” and “seem[ed] totally unworkable” because many employees would provide greater assistance to the “kingpin” than non-employee third parties, and that it made little sense to extend the government‘s theory to one group if it concededly could not reach the other. Id. at 382. This application of complicity and conspiracy would disrupt the carefully defined statutory gradation of offenses; the low-level henchman would find himself subject to the more severe penalties applicable to the “kingpin.” Because the Court determined that Congress did not intend for the “kingpin” statute to apply to the class of individuals involved in the case, the defendant‘s conviction was overturned. Id.

3. Identifying an Affirmative Legislative Policy

Accepting Gebardi‘s teaching that conspiracy and complicity liability will not lie when Congress demonstrates an affirmative legislative policy to leave some type of participant in a criminal transaction unpunished, 287 U.S. at 123, the question becomes how to identify such a policy. As the common-law principle outlined above indicates, we cannot identify such a policy whenever a statute focuses on certain categories of persons at the exclusion of others. Gebardi confirms this, emphasizing that its reasoning was “concerned with something more than an agreement between two persons for one of them to commit an offense which the other cannot commit.” Id. at 121. In Gebardi that “something more” was a recognition that because a woman‘s participation was “an inseparable incident of all cases in which the woman is a voluntary agent” capable of entering into a conspiracy, Congress‘s silence as to the women‘s liability was a conferral of immunity. Id. at 121-23. Similarly, in Amen the Court saw that the continuing criminal enterprise provision “was designed to reach the top brass in the drug rings, not the lieutenants and foot soldiers” and broadening the scope of liability with the conspiracy statute would subvert that purpose. 831 F.2d at 381 (brackets and internal quotation marks omitted). In both instances the courts looked to the text of the statute and the purpose that Congress was trying to achieve, thereby honoring their “over-arching obligation to give effect to congressional intent” when interpreting statutes. United States v. Bonanno Organized Crime Family of La Cosa Nostra, 879 F.2d 20, 21 (2d Cir. 1989). In keeping with traditional principles of statutory interpretation, as well as the analysis employed in Gebardi and its progeny, an affirmative legislative policy can be discerned by looking to the statute‘s text, structure, and legislative history.⁵

4. Government‘s Arguments for a Narrower Principle

The government argues for a much narrower reading of Gebardi that would effectively circumscribe the ability of the courts to ascertain congressional intent in enacting criminal statutes. The government argues that Gebardi forecloses liability for conspiracy or complicity only when (1) “the defendant‘s consent or acquiescence is inherent in the [substantive] offense,” or (2) “the defendant‘s participation in the crime is frequently, if not normally a feature of the [substantive] criminal conduct.” Appellant‘s Opening Br. at 24 (internal quotation marks omitted).

A number of problems arise with either of these narrow readings of Gebardi. The government‘s first reading of Gebardi is foreclosed because, at least in the conspiracy context, it is the same as Wharton‘s Rule. As noted, where a substantive offense requires persons to agree in order to commit it, Wharton‘s Rule disallows liability for conspiracy based on the same agreement required for the substantive crime. See Iannelli, 420 U.S. at 773. Here, the government suggests that we should read the Gebardi principle to mean the same thing: that liability for conspiracy is barred when “the defendant‘s consent or acquiescence is inherent in the [substantive] offense.” Appellant‘s Opening Br. at 24 (internal quotation marks omitted). The opinion in Gebardi explicitly stated that its reasoning was not based on Wharton‘s Rule; thus that cannot be the basis for the exception. Gebardi, 287 U.S. at 122-23.⁶

The government‘s argument that the exception is limited to situations where the defendant‘s conduct is inherent in the substantive offense is also inconsistent with Amen. Our holding in Amen, which considered an individual who was not an employee of the criminal enterprise, did not turn on the fact that the defendant was essential to the existence of the criminal transaction under consideration. Amen, 831 F.2d at 381. Although a “criminal enterprise” with a “kingpin” must have employees, and such employees are thus essential to the statute‘s application, the enterprise need not work with non-employee third parties. Amen held that the “kingpin” statute did not apply to third parties, and did so based on the intentions of Congress rather than because third parties were required for a criminal enterprise to exist. See id. at 382.

Second, we do not share the government‘s view that Gebardi asks whether a certain type of defendant‘s conduct is “frequently, if not normally” involved in an offense. Gebardi, 287 U.S. at 121. With respect to the statute giving rise to Gebardi—the Mann Act—there was no question that a woman‘s participation in the crime was “frequently, if not normally” a feature of a violation. Indeed, a woman‘s participation, either willing or unwilling, was required in every violation. But the Court did not merely ask whether her involvement was “frequently, if not normally” a feature of a violation; instead, the Court discerned the legislative policy of the Mann Act, and provided immunity only to the extent it comported with the Act‘s policy. Id. at 123.

Indeed, in United States v. Holte, 236 U.S. 140 (1915), a predecessor case to Gebardi, the Court explicitly held that a woman could be found guilty for conspiring to violate the Mann Act. The Court described a hypothetical case where immunity would not be appropriate:

Suppose, for instance, that a professional prostitute, as well able to look out for herself as was the man, should suggest and carry out a journey within the [Mann Act] in the hope of black-mailing the man, and should buy the railroad tickets, or should pay the fare from Jersey City to New York,—she would be within the letter of the [Mann Act], and we see no reason why the act should not be held to apply. We see equally little reason for not treating the preliminary agreement as a conspiracy that the law can reach, if we abandon the illusion that the woman always is the victim.

236 U.S. at 145. The Court‘s analysis in Holte, much like in Gebardi, did not merely ask whether a woman would “frequently if not normally” be present for violations of the Mann Act. Instead, the Court determined Congress‘s policy in enacting the statute, and limited liability consistent with that policy. To be sure, the fact that a woman was invariably part of a violation of the Act was relevant in discerning congressional policy. But the rule the Holte Court adopted was much more nuanced than could be justified by simply observing those offenses for which women would be present: by definition, a woman‘s presence was required for every violation of the Act.

Finally, the government relies on Ocasio v. United States, 136 S. Ct. 1423 (2016), a recent decision that it believes to have drawn narrowly the exception exemplified by Gebardi. The opinion in Ocasio considered an incident of bribery charged under the Hobbs Act, and a charge of conspiracy to violate the Hobbs Act by paying the same bribe. 136 S. Ct. at 1427. Although the language of the Hobbs Act prohibits “extortion” committed by “the obtaining of property from another, with his consent . . . under color of official right,” 18 U.S.C. § 1951(b)(2), the Supreme Court has held that this tortured language is best understood as the “rough equivalent of what we would now describe as ‘taking a bribe,‘” Ocasio, 136 S. Ct. at 1428 (quoting Evans v. United States, 504 U.S. 255, 260 (1992)). In other words, the Hobbs Act‘s text speaks as though a bribe-payer is being “extorted,” when, in reality, the bribe may be a consensual one paid to secure some advantage.

The defendant in Ocasio contended, using the language of the Hobbs Act, that he could not be convicted of conspiracy. He noted that the Hobbs Act criminalized “obtaining of property from another,” 18 U.S.C. § 1951(b)(2) (emphasis added). He then contended that a conspiracy charge was not appropriate, because “the conspirators,” who were the officials taking the bribe and the persons paying it, “had not agreed to obtain money from [“another“—that is, from] a person who was not a member of the conspiracy.” Ocasio, 136 S. Ct. at 1429. The Court rejected this argument, explaining that it did not matter that the defendants who paid the bribes “did not have the objective of obtaining money ‘from another’ because the money in question was their own.” Id. at 1433. The Court simply reasoned that it was sufficient for the defendants to conspire with others who would take money “from another,” even if that “[]other” person happened to be the conspirator himself. Id. at 1434.

The opinion in Ocasio emphasized that the crime in question, Hobbs Act extortion, bears a meaning not readily discernible from its text. Because, as noted, the statute essentially criminalizes “taking a bribe,” the Court was unwilling to indulge the defendant‘s argument that the text indicated an affirmative legislative policy to leave the “extorted” party unpunished, or a desire to punish only the party taking property “from another.” Id. at 1435-36.

Although Ocasio arose in a setting where a statute‘s language arguably suggested that certain persons are spared from liability, the unique features of Hobbs Act extortion limit Ocasio‘s helpfulness to the government. Because the Supreme Court did not agree that the Hobbs Act manifested the “something more” present in Gebardi, namely any intention to limit liability for the payer of a bribe, the Court rejected the argument that conspiracy liability should be circumscribed based on any such limitation. Id. at 1434-35 (“The subtext of [defendant‘s] arguments is that it seems unnatural to prosecute bribery on the basis of a statute prohibiting ‘extortion,’ but this Court held in Evans that Hobbs Act extortion ‘under color of official right’ includes the rough equivalent of what we would now describe as ‘taking a bribe.’ . . . [W]e have no occasion to [overrule Evans.]” (internal quotation marks and citations omitted)). Consequently, the case does not demonstrate a narrowing of the affirmative-legislative-policy exception, but simply a situation where there was no affirmative legislative policy to leave the bribe payers unpunished. Moreover, Ocasio‘s independent ruling that incapacity to commit a substantive offense does not, without more, preclude conspiracy or complicity charges, is merely a reaffirmation of the common-law principle addressed above, not an abdication of the affirmative-legislative-policy exception.

C. The Affirmative Legislative Policy Regarding the FCPA‘s Coverage

Applying the teachings of Gebardi and Amen to the FCPA, we find the “something more” that evinces an affirmative legislative policy to leave the category of defendants omitted from the statutory framework unpunished. In particular, the carefully tailored text of the statute, read against the backdrop of a well-established principle that U.S. law does not apply extraterritorially without express congressional authorization and a legislative history reflecting that Congress drew lines in the FCPA out of specific concern about the scope of extraterritorial application of the statute, persuades us that Congress did not intend for persons outside of the statute‘s carefully delimited categories to be subject to conspiracy or complicity liability. Our conclusion is consistent with the reasoning of other courts that have addressed this question. See United States v. Castle, 925 F.2d 831 (5th Cir. 1991); United States v. Bodmer, 342 F. Supp. 2d 176 (S.D.N.Y. 2004).

1. Text of the FCPA

We begin with the text of the statute. Like the Mann Act, which “[did] not specifically impose any penalty upon” a woman for assisting in her own transportation across state lines, “although it deal[t] in detail with” other persons, Gebardi, 287 U.S. at 119, the FCPA contains no provision assigning liability to persons in the defendant‘s position—nonresident foreign nationals, acting outside American territory, who lack an agency relationship with a U.S. person, and who are not officers, directors, employees, or stockholders of American companies. See 15 U.S.C. §§ 78dd-1; 78dd-2; 78dd-3.

Moreover, in Gebardi, the statute under consideration was less clear as to Congress‘s intent to exclude the defendant from liability, compared to the FCPA‘s utter silence regarding the class of defendants involved in this case. As noted, the Mann Act placed a penalty upon “any person who shall knowingly transport or cause to be transported, or aid or assist in obtaining transportation for . . . any woman or girl for . . . any . . . immoral purpose.” Id. at 118. The Supreme Court explained that, for a woman to be liable under the Mann Act, her role must “be more active than mere agreement on her part to the transportation and its immoral purpose.” Id. at 119. But the Court stated in Gebardi, much as it did in Holte, that the Mann Act would cover the woman to the extent she were to “‘aid or assist’ some one else in transporting or in procuring transportation” for her. Id. Thus, the statute created at least some potential for liability where a woman did more than exhibiting “mere agreement . . . to the transportation.” Id. In the present case, by contrast, there is no text that creates any liability whatsoever for the class of persons in question.

2. Structure of the FCPA

A second piece of evidence—the structure of the FCPA—confirms that Congress‘s omission of the class of persons under discussion was not accidental, but instead was a limitation created with surgical precision to limit its jurisdictional reach. The statute includes specific provisions covering every other possible combination of nationality, location, and agency relation, leaving excluded only nonresident foreign nationals outside American territory without an agency relationship with a U.S. person, and who are not officers, directors, employees, or stockholders of American companies.

The FCPA explicitly lays out several different categories of persons over whom the government may exercise jurisdiction. First, the statute prohibits a company issuing securities regulated by federal law (an “issuer“) from using interstate commerce in connection with certain types of corrupt payments to foreign officials. 15 U.S.C. § 78dd-1(a). The same prohibitions apply to any “domestic concern.” 15 U.S.C. § 78dd-2(a). “Domestic concern” is a broad term that covers “any individual who is a citizen, national, or resident of the United States,” 15 U.S.C. § 78dd-2(h)(1)(A), wherever such a person happens to be in the world. It also covers most businesses—including partnerships, sole proprietorships, and unincorporated organizations—that are organized under state or federal law or have principal places of business in the United States. 15 U.S.C. § 78dd-2(h)(1)(B).

Importantly, the prohibitions on issuers and domestic concerns also apply to “any officer, director, employee, or agent of” the entity, “or any stockholder thereof acting on behalf of” the entity. 15 U.S.C. §§ 78dd-1(a), 78dd-2(a). The statute‘s prohibitions thus apply not only (for example) to partnerships organized under state law, but also to their executives, janitors, and travel agents. And, although a person must be a citizen, national, or resident of the United States to be charged as a domestic concern, no similar requirement limits the liability of officers, employees, or agents of domestic concerns and issuers.

Second, the statute prohibits “any person other than an issuer . . . or a domestic concern” from using interstate commerce in furtherance of corrupt payments to foreign officials, but only while the person is “in the territory of the United States.” 15 U.S.C. § 78dd-3(a). A “person” is “any natural person other than a national of the United States,” as well as any business organized under foreign law. 15 U.S.C. § 78dd-3(f)(1).

In sum, these provisions provide jurisdiction over the following persons, in the following scenarios:

1. American citizens, nationals, and residents, regardless of whether they violate the FCPA domestically or abroad;
2. most American companies, regardless of whether they violate the FCPA domestically or abroad;

(3) agents, employees, officers, directors, and shareholders of most American companies, when they act on the company‘s behalf, regardless of whether they violate the FCPA domestically or abroad;

(4) foreign persons (including foreign nationals and most foreign companies) not within any of the aforementioned categories who violate the FCPA while present in the United States.

The single, obvious omission is jurisdiction over a foreign national who acts outside the United States, but not on behalf of an American person or company as an officer, director, employee, agent, or stockholder.

3. Legislative History

The question thus becomes whether there is “something more,” a policy basis for Congress to exclude Hoskins‘s category of defendants from criminal liability—something akin to the Mann Act‘s decision not to punish the woman who is frequently, if not normally involved in the offense or 21 U.S.C. § 848‘s gradation of punishment based on leadership in a criminal enterprise. We think there is. “It is a basic premise of our legal system that, in general, United States law governs domestically but does not rule the world.” RJR Nabisco, Inc. v. European Cmty., 136 S. Ct. 2090, 2100 (2016) (internal quotation marks omitted). Courts will therefore not apply a U.S. law extraterritorially unless “the affirmative intention of the Congress [is] clearly expressed.” E.E.O.C. v. Arabian Am. Oil Co., 499 U.S. 244, 248 (1991). This principle stems from the risk of “unintended clashes between our laws and those of other nations which could result in international discord.” Id. The legislative history of the FCPA makes it clear that Congress was attuned to these risks and carefully delimited the statute accordingly.

a. The Foreign Corrupt Practices Act of 1977

When President Carter took office in 1977, sponsors of the 1976 precursor to the FCPA exhorted the administration to take an active approach in promoting an anti-bribery statute comparable to the 1976 bill that passed the Senate but failed to pass the House. See Mike Koehler, The Story of the Foreign Corrupt Practices Act, 73 OHIO ST. L.J. 929, 996 (2012). The Carter Administration indicated its support for such a statute, and, in particular, suggested that “specific criminal penalties” for acts of bribery were the correct approach to solving the problem. See Foreign Corrupt Practices and Domestic and Foreign Investment Disclosure: Hearing on S. 305 Before the S. Comm. on Banking, Hous., & Urban Affairs, 95th Cong. 67 (1977) (statement of W. Michael Blumenthal, Secretary of the Treasury).

i. The Administration and the Senate Bill

Although it hoped to pass aggressive anti-bribery legislation, the Administration recognized that a statute focusing on criminalization, rather than disclosure, required a delicate touch where extraterritorial conduct and foreign nationals were concerned. The Secretary of the Treasury, W. Michael Blumenthal, noted as much at a hearing before the Senate Committee on Banking, Housing, and Urban Affairs on March 16, 1977:

[T]he Administration recognizes that great care must be taken with an approach which makes certain types of extraterritorial conduct subject to our country‘s criminal laws. Moreover, a law which provides criminal penalties must describe the persons and acts covered with a high degree of specificity in order to be enforceable, to provide fair warning to American businessmen.

Id. at 70 Secretary Blumenthal emphasized, in particular, the Administration‘s concerns regarding the protection of foreign nationals:

There is a problem of extraterritoriality which needs to be carefully addressed. There is also a question of insuring fairness and due process, not only for American citizens but also for those foreign citizens and foreign countries who may in some way become involved and whose reputations become involved in particular allegations. We have to deal with the question of how we can write the bill in such a way that it includes protections in this regard.

Id. at 94 The Secretary‘s requests, in other words, were for the Committee to enact a bill that clarified the extent of liability, and that provided protection for foreign persons.

In the initial draft of the FCPA, individual liability for bribery was chargeable largely through the conspiracy and complicity statutes. In the initial draft, as in the current version of the FCPA, there were three categories of legal rules:

• first, obligations to create “books, records, and accounts, which accurately and fairly reflect the transactions” of the company, S. 305, 95th Cong., § 102(2)(A) (as introduced Jan. 18, 1977) (hereinafter “S. 305 as Introduced“),;
• second, obligations to “devise and maintain an adequate system of internal accounting controls sufficient to provide reasonable assurances” that transactions are properly authorized and recorded, id. at §§ 102(2)(B); and
• third, provisions prohibiting the payment of bribes to foreign officials, id. at §§ 103-104.

Individual liability was discussed for the first two classes of rules—the “books and records” and “internal accounting controls” provisions. See S. 305 as Introduced, § 102(3) (“It shall be unlawful for any person, directly or indirectly, to falsify, or cause to be falsified, any book, record, account, or document . . . .“). But the anti-bribery provision, spread over Sections 103 and 104, covered only bribery by an “issuer” or a “domestic concern.” Conspicuously absent was any provision creating liability for the employees of an “issuer,” which meant that there would be no liability under a substantive provision of the statute for an employee of a publicly-traded company who approved a bribe. See S. 305 as Introduced, § 103. Although the draft prohibited “domestic concerns, other than an issuer” from offering, paying, promising to pay, or “authoriz[ing] the payment of” bribes, see S. 305 as Introduced, § 104(a), and the draft included among “domestic concerns” both “an individual who is a citizen or national of the United States,” as well as most American companies, see S. 305 as Introduced, § 104(c)(1), the provision did not clarify that an individual employee of a non-issuer company would be liable for the company‘s payment of bribes. The result was draft legislation that clearly did not create direct individual liability for employees of publicly-traded companies, and only arguably created it for employees of other companies. As explained below, the Senate‘s intention in this draft was to create individual liability using the conspiracy and complicity statutes rather than by enumerating particular individuals who could be liable within the statute‘s text.

On April 6, 1977, less than a month after Secretary Blumenthal‘s testimony before a Senate committee, his concerns regarding individual liability under the nascent FCPA were specifically addressed in a markup session held by the same committee. During the session, the Committee on Banking, Housing and Urban Affairs discussed the version of S. 305 that had been introduced to the Senate. See Markup Session on S. 305, Corporate Bribery, S. Comm. on Banking, Hous. and Urban Affairs, 95th Cong. 1-2 (1977). The Committee adopted two amendments that significantly clarified the classes of persons liable under the statute, and did so by reducing the bill‘s reliance on conspiracy and complicity theories:

Amendment number 3, on page 4, line 5 after the word “title” insert the words “or any officer, director, employee or stockholder thereof acting on behalf of such issuer.”

On page 6, line 1 after “1934” insert the words: “or any officer, director, employee or stockholder thereof acting on behalf of such domestic concern.”

Id. at 12. These amendments clarified that particular individuals would be liable for certain violations of the statute. These amendments were explained in the markup hearing as a change to the Senate‘s earlier plan to cover individuals using theories of complicity and conspiracy instead of defining specifically the persons who could be liable under the statute:

[T]his amendment also reflects the Administration‘s position in recommending that individuals be covered. Indeed, I believe that the committee last year intended to cover individuals; however, it wasn‘t specifically stated. They were intended to be covered as aiders, abettors and conspirators and so on and so forth, and this makes clear that they are covered directly and also it makes it clear that they are covered in their capacity in acting on behalf of the company.

Id.

The markup session provides powerful evidence of two points relevant to this case. First, before the Carter Administration‘s concerns and the markup hearing detailed above, the Senate had planned to adopt a bill that largely omitted references to individual liability, and that instead relied on theories of conspiracy and complicity to tie individual action to corporate misdeeds. In response to administration concerns—particularly concerns regarding the clarity of liability and its application to foreign persons—the Senate rejected its prior approach. Instead, it opted for a version of the bill that was not reliant on conspiracy or complicity theories. Rather, it defined, with great precision, who would be liable.

ii. The House Bill and Final Legislation

In the House, Representative Bob Eckhardt initially proposed a bill, the Unlawful Corporate Payments Act of 1977, with broader coverage than the Senate‘s initial legislation. See H.R. 3815, 95th Cong. (as introduced Feb. 22, 1977). The bill created liability not only for officers, directors, and employees of issuers and domestic concerns, id. at §§ 2(a), 3(c)(1), but also for “agents” who “knowingly and willfully carried out” bribes, id. at §§ 2(a), 3(c)(2). The sections covering individuals—including subsection (3)(c)(2), which covered “agents” who “carried out” bribes—appeared to apply regardless of nationality or location.

Several leading authorities, including Harvey L. Pitt, General Counsel of the SEC, suggested to Representative Eckhardt and other Congressmen on the Committee on Interstate and Foreign Commerce that these provisions went too far. In a hearing discussing the bill, Mr. Pitt stated as follows:

At a minimum, I think the language of subsection (c)(2), applying to any agent, might create some jurisdictional problems if the agent is wholly situated overseas and has not been in this country. While I think there are jurisdictional ties that could be asserted, the problems you express in this case might be even worse in terms of prosecution. But, I think you could so something along the lines you are suggesting either by amending this subsection or by report language that would clarify burdens of proof, obligations, and the involvement of agents, to provide a fair opportunity for an agent to present his defense. That does seem to be a very serious concern.

Unlawful Corporate Payments Act of 1977: Hearings Before the Subcomm. on Consumer Prot. and Fin. of the Comm. on Interstate and Foreign Commerce, 95th Cong. 232 (1977) (statement of Harvey L. Pitt, General Counsel, Securities and Exchange Commission). Following these hearings, the Committee on Interstate and Foreign Commerce reported an amended bill to the House. The revised version allowed liability for agents and employees of issuers and domestic concerns only if the company for which they worked was also found to be liable—a change that essentially increased the U.S. nexus required for an offense to be covered.⁷ See H. Rep. No. 95-831, at 13 (1977) (Conf. Rep.) (hereinafter “1977 Conference Report“).

The final version of the FCPA, agreed to in conference, demonstrated a compromise between the House and Senate versions. Like the Senate‘s revised bill—and the House‘s original bill—it named particular categories of individuals who would be liable under the FCPA rather than relying on the use of conspiracy and complicity principles to create such liability. See Foreign Corrupt Practices Act, Pub. L. No. 95-213, § 103(a), 91 Stat. 1494, 1495 (creating liability for bribery committed by an “issuer” and “any officer, director, employee, or agent of such

H.R. Rep. 95-640, at 4-5 (1977). This strong language underlines what is otherwise clear throughout the legislative history: that the House was concerned chiefly with questions of morality and public propriety—and public perception—whereas the Senate appeared more concerned with the SEC‘s ability to obtain accurate disclosures and to police corporate filings. issuer or any stockholder thereof acting on behalf of such issuer“); id. at § 104(a), 91 Stat. at 1496 (creating liability for bribery by a “domestic concern, other than an issuer,” and for “any officer, director, employee, or agent of such domestic concern or any stockholder thereof acting on behalf of such domestic concern“). It did allow liability for agents, but restricted the liability to an agent who was “a United States citizen, national, or resident or is otherwise subject to the jurisdiction of the United States,”⁸ and also required a finding that the employer had been liable. Id. at § 103(a), 91 Stat. at 1496; id. at § 104(b)(3)(a), 91 Stat. at 1497. The bill also rejected liability for foreign affiliates of American companies.

The Conference Report emphasized that the statute drew deliberate lines regarding the liability of foreign persons, both corporate and natural:

[T]he conferees recognized the inherent jurisdictional, enforcement, and diplomatic difficulties raised by the inclusion of foreign subsidiaries of U.S. companies in the direct prohibitions of the bill. However, the conferees intend to make clear that any issuer or domestic concern which engages in bribery of foreign officials indirectly through any other person or entity would itself be liable under the bill. The conferees recognized that such jurisdictional enforcement, and diplomatic difficulties may not be present in the case of individuals who are U.S. citizens, nationals, or residents. Therefore, individuals other than those specifically covered by the bill (e.g., officers, directors, employees, agents, or stockholders acting on behalf of an issuer or domestic concern) will be liable when they act in relation to the affairs of any foreign subsidiary of an issuer or domestic concern if they are citizens, nationals, or residents of the United States. In addition, the conferees determined that foreign nationals or residents otherwise under the jurisdiction of the United States would be covered by the bill in circumstances where an issuer or domestic concern engaged in conduct proscribed by the bill.

1977 Conference Report at 14. This discussion, much like the discussion in the earlier hearings on the Senate‘s 1976 legislation, largely resolves the problem of liability against foreign persons by noting that an American company will be liable if it acts through unreachable foreign affiliates. See id. (noting that “any issuer or domestic concern which engages in bribery of foreign officials indirectly through any other person or entity would itself be liable“). Its mention of situations where “foreign nationals or residents otherwise under the jurisdiction of the United States” would be liable because “an issuer or domestic concern engaged in conduct proscribed by the bill” clearly refers to the statute‘s liability for agents, which permits jurisdiction over foreign nationals. Id. The Conference Report made no mention of conspiracy or aiding-and-abetting theories of liability.

b. The 1998 Revisions⁹

In 1998, Congress amended the FCPA. The Committee Report from the Senate Committee on Banking, Housing, and Urban Affairs noted that “[s]ince the passage of the FCPA, American businesses have operated at a disadvantage relative to foreign competitors who have continued to pay bribes without fear of penalty,” because their countries’ laws did not include comparable prohibitions on bribery. S. Rep. No. 105-277, at 2 (1998) (hereinafter “1998 Senate Report“). In response to this problem, “[i]n 1988, Congress directed the Executive Branch actively to seek to level the playing field by encouraging . . . trading partners to enact legislation similar to the FCPA.” Id. “These efforts eventually culminated in the Organization for Economic Cooperation and Development Convention on Combating Bribery of Foreign Public Officials in International Business Transactions (the ‘OECD Convention‘),” which asked signatory nations to enact anti-bribery laws containing certain minimum requirements. Id.

i. Congress‘s View of the Amendments

As the Committee Report explained, the 1998 statute aimed to “amend[] the FCPA to conform it to the requirements of and to implement the OECD Convention.” Id. The amendments served five major purposes, although only three are pertinent here:

[First], the OECD Convention calls on parties to cover “any person“; the [1977] FCPA cover[ed] only issuers with securities registered under the 1934 Securities Exchange Act and “domestic concerns.” The Act, therefore, expands the FCPA‘s coverage to include all foreign persons who commit an act in furtherance of a foreign bribe while in the United States. . . .

[Additionally], the OECD Convention calls on parties to assert nationality jurisdiction when consistent with national legal and constitutional principles. Accordingly, the Act amends the FCPA to provide for jurisdiction over the acts of U.S. businesses and nationals in furtherance of unlawful payments that take place wholly outside the United States. This exercise of jurisdiction over U.S. businesses and nationals for unlawful conduct abroad is consistent with U.S. legal and constitutional principles . . ..

[F]inally, the Act amends the FCPA to eliminate the current disparity in penalties applicable to U.S. nationals and foreign nationals employed by or acting as agents of U.S. companies. In the [1977] statute, foreign nationals employed by or acting as agents of U.S. companies [were] subject only to civil penalties. The Act eliminates this restriction and subjects all employees or agents of U.S. businesses to both civil and criminal penalties.

Id. at 2-3. The relevant changes to the statute, in short, were liability for foreign persons who committed acts within the United States, assertion of jurisdiction over American businesses and nationals bribing persons wholly outside the United States, and creation of criminal, rather than just civil, penalties for foreign nationals who are employees or agents of American companies. Clearly, none of these goals have an impact on the question at issue in this case—whether a nonresident foreign national, acting entirely outside the United States, and who is not an employee or agent of an American company, may be liable based on a conspiracy or complicity theory.

Moreover, the Committee Report took great pains to emphasize that the foreign nationals covered under the statute fit within three categories: (1) those who acted on American soil, (2) those who were officers, directors, employees, or shareholders of U.S. companies, and (3) those who were agents of U.S. companies. The following examples illustrate the Report‘s care in making the matter clear:

Section 2(d) implements the OECD Convention by amending § 32(c) of the Securities Exchange Act of 1934 to eliminate the current disparity in treatment between U.S. nationals that are employees or agents of issuers and foreign nationals that are employees or agents of issuers. Presently, foreign nationals who are employees or agents (as opposed to officers or directors) are subject only to civil sanctions. . . .

[I]t is expected that the established principles of liability, including principles of vicarious liability, that apply under the current version of the FCPA shall apply to the liability of U.S. businesses for acts taken on their behalf by their officers, directors, employees, agents, or stockholders outside the United States, regardless of the nationality of the officer, director, employee, agent, or stockholder.

The new offense . . . provid[es] for criminal jurisdiction in this country over bribery by foreign nationals of foreign officials when the foreign national takes some act in furtherance of the bribery within the territory of the United States.

Id. at 4-5 (emphases added). Each mention of foreign nationals is carefully followed by clarifications—often highly repetitive ones—noting that foreign nationals are liable only because they fall within one of the three categories. There is no mention, in any of these seemingly exhaustive descriptions of how foreign nationals are covered under the statute, of liability based on conspiring or aiding and abetting an offense even though the foreign national is not an agent, employee, officer, director, or shareholder of the American company, and even though the foreign national is operating entirely outside the territory of the United States.

ii. Language of the OECD Convention

The government‘s argument—that Congress must have intended to cover foreign nationals acting abroad who are not employees or agents of an American company—focuses heavily on the OECD Convention with which Congress intended to make American law comply. The government first contends that the OECD Convention shows that the United States agreed to “take such measures as may be necessary to establish that it is a criminal offence under its law for any person intentionally to” engage in bribery of foreign officials. Convention on Combating Bribery and Foreign Public Officials in International Business Transactions art. 1.1, Dec. 17, 1997, S. Treaty Doc. No. 105-43, 37 I.L.M. 1 (1998) (hereinafter “OECD Convention“) (emphasis added). The government reads the words “any person” to apply expansively, including to nonresident foreign nationals who do not have direct connections to American businesses.

The government‘s argument falters for two reasons. First, the requirement that intentional bribery by “any person” is illegal is a highly general one; it does not require approval of the precise type of complicity or conspiracy theory involved in this case. Second, Congress carefully considered the “any person” language, and interpreted it in a way that does not involve the government‘s theory of liability here. The Senate‘s Committee Report first noted that the “any person” text was effectuated by expanding the FCPA to include conduct by foreign nationals within the United States:

[T]he OECD Convention calls on parties to cover “any person“; the current FCPA covers only issuers with securities registered under the 1934 Securities Exchange Act and “domestic concerns.” The Act, therefore, expands the FCPA‘s coverage to include all foreign persons who commit an act in furtherance of a foreign bribe while in the United States.

1998 Senate Report, at 2-3. Congress also associated the “any person” language with applying criminal, rather than civil, penalties to foreign nationals who violated the statute as employees or agents of issuers or domestic concerns. Id. at 4, 5. In short, Congress focused specifically on the text the government discusses, and employed it in a reasonable way that is not connected to complicity or conspiracy liability for foreign nationals.

The government next notes that the OECD Convention specifically mentions ancillary theories of liability such as conspiracy and complicity:

Each Party shall take any measures necessary to establish that complicity in, including incitement, aiding and abetting, or authorisation of an act of bribery of a foreign public official shall be a criminal offence. Attempt and conspiracy to bribe a foreign public official shall be criminal offences to the same extent as attempt and conspiracy to bribe a public official of that Party.

OECD Convention art. 1.2. The government contends that the text of this provision, requiring that conspiracy to bribe a foreign official “shall be [a] criminal offence[] to the same extent as . . . conspiracy to bribe a public official of” the United States, demands that a conspiring foreign national operating abroad be covered by the FCPA. The government‘s argument is that federal bribery statutes would indeed cover situations where overseas defendants conspire to bribe an American official.

The difficulty with the government‘s position, however, is that this provision covers the content of substantive law—the particular acts prohibited by it—not the law‘s jurisdictional aspects. A separate part of the Convention addresses jurisdictional questions. See OECD Convention art. 4. Moreover, adopting the government‘s view that the jurisdictional reach of the FCPA must be coterminous with that of bribery of American officials would transform the FCPA into a law that purports to rule the world. The defendant notes, for example, that bribery statutes covering American officials prohibit not only crimes with foreign national conspirators acting overseas, therefore, under the government‘s theory, these statutes likely cover situations in which the entire offense occurred overseas—that is, where there is no U.S. nexus at all except that the official to be bribed is stateside. The government does not dispute this point. Consequently, if read as the government proposes, the above-quoted provision of the Convention would cover conspiracies to bribe foreign officials consisting entirely of actions taken abroad. That is obviously not consistent with the legislation Congress wrote, and it cannot be what the OECD Convention requires.

The government also points to provisions about the territorial reach of the OECD Convention. In particular, the government emphasizes the following passage:

Each Party shall take such measures as may be necessary to establish its jurisdiction over the bribery of a foreign public official when the offence is committed in whole or in part in its territory.

OECD Convention art. 4.1. The government essentially believes this passage to establish that, where “any part” of the offense occurs within the United States, the country is required to exercise jurisdiction over someone whose conduct is related to the offense, no matter how attenuated the person‘s connection to the acts taken on American soil.

The government‘s reading is undercut by the commentaries to the OECD Convention, and by Congress‘s careful consideration of the provision‘s meaning. The accompanying commentary to the Convention states, regarding Article 4.1, that “[t]he territorial basis for jurisdiction should be interpreted broadly so that an extensive physical connection to the bribery act is not required.” OECD

Convention cmt. 4.1. This language suggests that the Convention contemplated jurisdiction over persons with some “physical connection to the bribery act,” even if not an “extensive” one, rather than persons with no physical connection to the actions at all. Congress plainly shared this view of the provision. As the Committee Report noted:

The OECD Convention requires each Party to “take such measures as may be necessary to establish its jurisdiction over the bribery of a foreign public official when the offense is committed in whole or in part in its territory.” OECD Convention, Art. 4, ¶ 1. The new offense complies with this section by providing for criminal jurisdiction in this country over bribery by foreign nationals of foreign officials when the foreign national takes some act in furtherance of the bribery within the territory of the United States.

1998 Senate Report, at 5. Congress‘s reading, and the view described in the commentaries both comport with the Convention provision‘s text. A requirement that a nation “establish its jurisdiction over the bribery of a foreign public official” does not say that it must create jurisdiction over persons in foreign lands with only distant connections to the offense. It is fairly read to mean that a nation that has agreed to the Convention must enact a law covering persons who commit acts within the nation‘s own borders.

c. The Legislative History‘s Demonstration of an Affirmative Legislative Policy

The strands of the legislative history demonstrate, in several ways, the affirmative policy described above: a desire to leave foreign nationals outside the FCPA when they do not act as agents, employees, directors, officers, or shareholders of an American issuer or domestic concern, and when they operate outside United States territory.

First, it is clear that the FCPA‘s enumeration of the particular individuals who may be held liable under the Act demonstrated a conscious choice by Congress to avoid creating individual liability through use of the conspiracy and complicity statutes. As discussed above, the statute‘s initial approach was to place liability for bribery largely upon companies, and then to allow prosecution of individuals for conspiring with companies or aiding and abetting their violations of the law. But the Carter Administration objected to that approach, voicing concerns for due process protections and clarity of rules for foreign persons. The statute was amended; the amended version narrowly tailored the liability for foreign individuals, and did not contemplate a reversal of that narrow tailoring by means of conspiracy and complicity theories. These changes were principally discussed in the Senate. But the House bill, and the final legislation, were structured similarly to the Senate‘s revised bill. At the same time that the Senate made these changes, the House was revising its own legislation to cut back on liability placed upon foreign agents, again because of specific concerns expressed by executive-branch officials regarding overreach.

The 1998 amendments surely extended the statute‘s jurisdictional reach. But in doing so, Congress delineated as specifically as possible the persons who would be liable, and under what circumstances liability would lie. None of the changes included liability for the class of individuals involved in this case. And despite the government‘s urging to the contrary, nothing in the OECD Convention required Congress to create such liability.

Congress also repeatedly emphasized that out-of-reach foreign entities should not create concern because American companies would be liable for violating the Act even if they did so indirectly through such persons. See 1998 Senate Report at 5 (“Although this section imposes liability only on U.S. persons, it is expected that the established principles of liability, including principles of vicarious liability, that apply under the current version of the FCPA shall apply to the liability of U.S. businesses for acts taken on their behalf by their officers, directors, employees, agents or stockholders outside the United States, regardless of the[ir] nationality . . . .“); 1977 Conference Report at 14 (noting, despite “inherent jurisdictional, enforcement, and diplomatic difficulties raised by the inclusion of foreign subsidiaries,” “the conferees intend to make clear that any issuer or domestic concern which engages in bribery of foreign officials indirectly through any other person or entity would itself be liable under the bill“).

Finally, limitations on liability for foreign nationals based on conspiracy and complicity theories were sensible given congressional concerns and aspirations in enacting the FCPA. In passing the statute, Congress was largely concerned with ensuring the SEC‘s ability to supervise and police companies, S. Rep. No. 95-114, at 2 (1977), reprinted in 1977 U.S.C.C.A.N. 4098, 4099, as well as the negative perception that bribery could create for American companies, its effect on the marketplace, and the foreign policy implications of the conduct, see H.R. Rep. No. 95-640, at 4-6 (1977). But Congress also desired that the statute not overreach in its prohibitions against foreign persons. Protection of foreign nationals who may not be learned in American law is consistent with the central motivations for passing the legislation, particularly foreign policy and the public perception of the United States. And the desire to protect such persons is pressing when considering the conspiracy and complicity statutes: these provisions are among the broadest and most shapeless of American law, and may ensnare persons with only a tenuous connection to a bribery scheme.

In short, the legislative history of the FCPA further demonstrates Congress‘s affirmative decision to exclude from liability the class of persons considered in this case and we thus hold that the government may not override that policy using the conspiracy and complicity rules.

D. Presumption Against Extraterritorial Application

Even if we were not persuaded that Congress had demonstrated an affirmative legislative policy in the FCPA to limit criminal liability to the enumerated categories of defendants, we would still rule for Hoskins because the government has not established a ““clearly expressed congressional intent to” allow conspiracy and complicity liability to broaden the extraterritorial reach of the statute. RJR Nabisco, 136 S. Ct. at 2100.

The Supreme Court‘s recent opinion in RJR Nabisco explained a “two-step framework for analyzing extraterritoriality issues“:

At the first step, we ask whether the presumption against extraterritoriality has been rebutted—that is, whether the statute gives a clear, affirmative indication that it applies extraterritorially. . . If the statute is not extraterritorial, then at the second step we determine whether the case involves a domestic application of the statute, and we do this by looking to the statute‘s “focus.” If the conduct relevant to the statute‘s focus occurred in the United States, then the case involves a permissible domestic application even if other conduct occurred abroad; but if the conduct relevant to the focus occurred in a foreign country, then the case involves an impermissible extraterritorial application regardless of any other conduct that occurred in U.S. territory.

What if we find at step one that a statute clearly does have extraterritorial effect? . . . . [W]e addressed this issue in Morrison [v. National Australia Bank Ltd., 561 U.S. 247 (2010)] explaining that it was necessary to consider § 10(b)‘s “focus” only because we found that the statute does not apply extraterritorially: “If § 10(b) did apply abroad, we would not need to determine which transnational frauds it applied to; it would apply to all of them (barring some other limitation).” The scope of an extraterritorial statute thus turns on the limits Congress has (or has not) imposed on the statute‘s foreign application, and not on the statute‘s “focus.”

Id. at 2101 (internal citation omitted).

In RJR Nabisco, the Court evaluated the extraterritorial application of the RICO Act. As the Court noted, the RICO Act does not describe distinctive conduct not punished by other laws. Id. at 2096-97. A given violation of the RICO Act is based on a pattern of violations of other criminal statutes—so-called “predicate offense[s]“—named within the RICO Act. Id. When a defendant commits several “predicate offenses” within a given period of time, the defendant may be sued for a separate racketeering offense under RICO. Applying the test quoted above, the Court determined that the RICO Act “applie[d] to foreign racketeering activity—but only to the extent that the predicates alleged in a particular case themselves apply extraterritorially.” Id. at 2102. The Court emphasized that “foreign conduct must violate a predicate statute that manifests an unmistakable congressional intent to apply extraterritorially.” Id (internal quotation marks omitted). That is because, “‘when a statute provides for some extraterritorial application, the presumption against extraterritoriality operates to limit that provision to its terms.‘” Id. (quoting Morrison, 561 U.S. at 265).

Even after determining that certain substantive provisions of the RICO Act applied extraterritorially, the Court ruled that the statute‘s provision allowing a private right of action—the basis of the legal claims in RJR Nabisco—did not apply extraterritorially. Id. at 2110. The Court‘s conclusion was based on its reading of the statute, as well as its observation that “providing a private civil remedy for foreign conduct creates a potential for international friction beyond that presented by merely applying U.S. substantive law to that foreign conduct.” Id. at 2106.

In brief, the Supreme Court‘s teachings in RJR Nabisco were that (1) when a statute includes some extraterritorial application, that application is limited by the statute‘s terms, and that (2) remedial provisions must be analyzed independently to discern whether they permit extraterritorial application.¹⁰ These principles, though articulated in the RICO context, are consistent with prior rulings by a number of courts regarding extraterritorial liability based on the conspiracy and complicity statutes. Those statutes, like RICO, do not create new substantive offenses, but merely allow liability for other legal violations. Accordingly, courts have repeatedly ruled that “[g]enerally, the extraterritorial reach of an ancillary offense like aiding and abetting or conspiracy is coterminous with that of the underlying criminal statute.” United States v. Ali, 718 F.3d 929, 939 (D.C. Cir. 2013); see also United States v. Yakou, 428 F.3d 241, 252 (D.C. Cir. 2005) (“The aiding and abetting statute, however, is not so broad as to expand the extraterritorial reach of the underlying statute.“); United States v. Hill, 279 F.3d 731, 739 (9th Cir. 2002) (“[A]iding and abetting, and conspiracy have been deemed to confer extraterritorial jurisdiction to the same extent as the offenses that underlie them.“); see also United States v. Bowman, 260 U.S. 94 (1922) (considering the extraterritorial application of the substantive offense and of a conspiracy charge without distinction).

These rules demonstrate that the conspiracy and complicity statutes may not be used to bring the charges involved in this appeal. Because some provisions of the FCPA have extraterritorial application, “‘the presumption against extraterritoriality operates to limit th[ose] provision[s] to [their] terms,‘” RJR Nabisco, 136 S. Ct. at 2102 (quoting Morrison, 561 U.S. at 265). And, as detailed at length above, the FCPA does not impose liability on a foreign national who is not an agent, employee, officer, director, or shareholder of an American issuer or domestic concern—unless that person commits a crime within the territory of the United States, see 15 U.S.C. § 78dd-3 (providing liability for persons “other than an issuer . . . or a domestic concern . . . or . . . any officer, director, employee, or agent of such person or any stockholder thereof” only if the person‘s conduct is undertaken “while in the territory of the United States“). In other words, the territorial limitations of the FCPA preclude liability for such a person. The government may not expand the extraterritorial reach of the FCPA by recourse to the conspiracy and complicity statutes.

The government cites numerous cases that it believes to stand for the proposition that the conspiracy and complicity statutes can cover extraterritorial conduct even when the underlying statute does not. See, e.g., Ford v. United States, 273 U.S. 593 (1927); United States v. Inco Bank & Tr. Corp., 845 F.2d 919 (11th Cir. 1988); United States v. Winter, 509 F.2d 975 (5th Cir. 1975); United States v. Lawson, 507 F.2d 433 (7th Cir. 1974); Rivard v. United States, 375 F.2d 882 (5th Cir. 1967). But these cases all considered statutes prohibiting illegal importation of various items—statutes that certainly contemplated the punishment of extraterritorial action of precisely the kind that the defendants in the cases were convicted. See Ford, 273 U.S. at 600-01 (conspiracy to import liquor into the United States); Inco Bank, 845 F.2d at 919-20 (conspiracy to defraud the United States by laundering money); Winter, 509 F.2d at 977 (importation of marijuana); Lawson, 507 F.2d at 435 (importation of cocaine); Rivard, 375 F.2d at 887 (importation of heroin). The government cites no case in which a statute drew specific lines as to its extraterritorial application, and those lines were exceeded using the conspiracy or complicity theories. The argument thus poses no difficulty for our understanding of RJR Nabisco and related principles of the extraterritorial application of conspiracy and complicity rules.

Consequently, the presumption against extraterritoriality bars the government from using the conspiracy and complicity statutes to charge Hoskins with any offense that is not punishable under the FCPA itself because of the statute‘s territorial limitations. That includes both charges that are the subject of this motion—conspiracy to violate Sections 78dd-2 and 78dd-3 of the FCPA, and liability as an accomplice for doing so—because the FCPA clearly dictates that foreign nationals may only violate the statute outside the United States if they are agents, employees, officers, directors, or shareholders of an American issuer or domestic concern. To hold Hoskins liable, the government must demonstrate that he falls within one of those categories or acted illegally on American soil.

III. The Second Object of the Conspiracy

Notwithstanding this Court‘s conclusion that Hoskins cannot be held liable under the FCPA if he is not in the categories of persons directly covered by the statute, the government argues that it was error for the district court to dismiss the second object of the conspiracy. We agree. The second object alleges that Hoskins willfully conspired with various codefendants to, “while in the territory of the United States,” commit acts in furtherance of bribing foreign officials in violation of Section 78dd-3. Indictment ¶ 26(b). The district court held that, because “it is undisputed that Mr. Hoskins never entered the territory of the United States and thus could not be prosecuted under this section,” Gebardi barred the government from charging Hoskins with the second object of the conspiracy. Hoskins, 123 F. Supp. 3d at 327 n.14. This Court agrees that Hoskins cannot be directly liable under Section 78dd-3. However, the government “maintains that it still intends to prove that [Hoskins] acted as an agent of a domestic concern liable as a principal for the substantive FCPA counts charged in the indictment” in violation of Section 78dd-2. Id. at 318-19 n.1. Provided that the government makes this showing, there is no affirmative legislative policy to leave his conduct unpunished, nor is there an extraterritorial application of the FCPA. Accordingly, the government should be allowed to argue that, as an agent, Hoskins committed the first object by conspiring with employees and other agents of Alstom U.S. and committed the second object by conspiring with foreign nationals who conducted relevant acts while in the United States. Accordingly, the district court erred in concluding that Hoskins was incapable of conspiring to violate Section 78dd-3 because he was never in the United States and we reverse this portion of the holding with instructions for the district court to reinstate the second object of the conspiracy.

CONCLUSION

We have considered the government‘s remaining arguments and find them to be without merit. Accordingly, the district court‘s orders are AFFIRMED in part and REVERSED in part.

GERARD E. LYNCH, Circuit Judge, concurring:

I join in Judge POOLER‘s thorough opinion for the Court, and write only to state why I regard this as a close and difficult case.

I

The operative indictment in this case charges, among other things, that Lawrence Hoskins, a senior vice president of a foreign corporation based abroad, (1) conspired with others, among them named employees of a Connecticut power company that is a subsidiary of the foreign corporation for which Hoskins worked, to commit offenses against the United States, including to violate the Foreign Corrupt Practices Act (“FCPA“), 15 U.S.C. § 78dd-2, by bribing Indonesian public officials, and (2) committed substantive violations of that statute in that he aided and abetted those American individuals who engaged in the charged bribery.

For purposes of this appeal, we assume that Hoskins was neither an employee nor an agent of a domestic concern.¹ He therefore does not fall within the terms of the statute, any more than the get-away driver for a bank robbery team has personally “by force and violence . . . take[n] . . . from the person or presence of another . . . any property . . . belonging to . . . any bank.” 18 U.S.C. § 2113(a). He did not, by his own actions, commit the acts prohibited by the FCPA, nor is he within the class of individuals who could commit the prohibited acts.

The problem for Hoskins, however, as for the get-away driver, is that 18 U.S.C. § 2 punishes “as a principal” anyone who, although he or she does not personally commit the acts constituting the offense, “aids, abets, counsels, commands, induces or procures” the commission of those acts by another; in addition, 18 U.S.C. § 371 punishes anyone who “conspire[s]” with another to commit the offense. The indictment in this case expressly charges that someone (specifically the alleged co-conspirator Pomponi, who was an employee or agent of a domestic concern) did engage in substantive violations of the FCPA, and that Hoskins conspired with, and aided and directed, that person in the commission of the offense.

As the opinion for the Court expressly recognizes, were we to rely solely on “the plain language of the general statutes regarding conspiracy and accessorial liability — which nothing in the text of the FCPA purports to override or limit — if Hoskins did what the indictment charges, he would appear to be guilty of conspiracy to violate the FCPA and (as an accomplice) of substantive violations of that statute,” Court Op. 19–20, just as the wheelman for the bank robbery team is guilty of conspiracy and of substantive violations of the bank robbery statute. That is precisely the result that the conspiracy and complicity statutes, and the common-law doctrines that long preceded such statutes and which the statutes codify, are designed to effect.

Moreover, as the Court also recognizes, the fact that the FCPA specifies that only particular classes of people can violate the law by bribing foreign officials does not in itself restrict the reach of conspiratorial or aiding and abetting liability to those same classes of people. Many offenses are defined such that they may be committed only by the actions of particular types of people. But as the Court‘s opinion ably documents, “the firm baseline rule with respect to both conspiracy and complicity is that where the crime is so defined that only certain categories of persons . . . may commit the crime through their own acts, persons not within those categories can be guilty of conspiring to commit the crime or of the substantive crime itself as accomplices.” Court Op. 21. That is not only the rule established by federal precedent; as the Court notes, that position is “universally held” in the states and in other Anglo-American jurisdictions. Id. n.4, quoting the American Law Institute, Model Penal Code and Commentaries, § 2.06 at 323 (1985).²

That baseline principle, like most principles, admits of exceptions, some of which have longstanding common-law roots. One such exception, as the Court‘s opinion notes, is exemplified by the Supreme Court‘s opinion in Gebardi v. United States, 287 U.S. 112 (1932). Sometimes we can infer from the apparent purpose of the statute that the legislature cannot have intended to extend accessorial liability to a class of persons who might better be thought of as victims of the crime (such as a willing underage participant in a sex act defined as rape because of the underage party‘s incapacity to consent, even where the minor intentionally facilitates the act and is old enough to have the capacity to commit a crime), or where a legislative sentencing scheme distinguishes levels of culpability among various participants, and treating the less culpable party as an accomplice of the more culpable one would undermine that scheme (as in the case of narcotics transactions, where possessors of illegal drugs for their own use are punished less severely than distributors; the purchaser is not treated as the accomplice of the seller even though he or she intentionally facilitates the sale).³

This exception, however, must be construed narrowly. Discerning when the legislature “must have” intended to exempt a particular class of persons from the plain text of its statutes is a tricky business. What, after all, in the language or structure of the statute distinguishes one statute that limits the category of principal offenders from another, such that some few should be singled out as clearly intending to preclude some persons outside that category from liability, and distinguished from the general run of statutes where no such intention can be discerned? Nothing, or at least not much, in the statutes at issue tells us to exclude some or all persons not within the designated category from accomplice liability.

In my view, it is helpful in most cases to look to the traditional principles derived from the common law, and embodied in the legislation and judicial decisions of the fifty states, which have the primary responsibility for enforcing criminal law in this country. That is not an idiosyncratic position of my own; it was the view expressed by Justice Jackson, writing for a unanimous Supreme Court in Morissette v. United States, 342 U.S. 246 (1952). The Morissette Court looked to the common law and the rulings of “[s]tate courts of last resort, on whom fall the heaviest burden of interpreting criminal law in this country,” to interpret a federal statute (the prohibition on theft of federal property, now codified at 18 U.S.C. § 641) that lacked “any express prescription of criminal intent.” Id. at 260–61. Justice Jackson concluded that, since Congress acted against the backdrop of “an unbroken course of judicial decision in all constituent states of the Union,” its silence about mens rea reflected the adoption of common-law principles, rather than their rejection. Id. at 261–62.

As our principal opinion today correctly points out, the Supreme Court in Gebardi seems to have considered its ruling, rejecting conspiratorial liability under the Mann Act for women who agreed to be transported across state lines, as a slight extension of the traditional common-law rules. Court Op. 26–27. But this fact alone would not persuade me that Gebardi opens a broad door to finding “legislative policy” exceptions to the general principle that persons outside defined legislative categories of principal liability may still be guilty of conspiracy and complicity. In codifying the general consensus of common-law complicity principles, the Model Penal Code confined the exceptions to two closely related situations, proposing to exclude from accomplice liability only “victim[s] of [the] offense,” MPC § 2.06(6)(a), and those whose “conduct is inevitably incident to its commission” in light of the definition of the offense, id. § 2.06(6)(b). The Commentary to these proposals, which extensively discusses the many decisions classifying persons as accomplices or not across a variety of crimes, treats Gebardi as an example of the latter exception. See American Law Institute, Model Penal Code and Commentaries, § 2.06 at 323–25 & nn.74, 82.⁴

The Fifth Circuit‘s decision in United States v. Castle, 925 F.2d 831 (5th Cir. 1991), illustrates a classic application of the Gebardi principle to the FCPA. There, the government attempted to prosecute the foreign official who received a bribe from representatives of a domestic concern for conspiracy to violate the FCPA. Such a prosecution runs directly in the face of Gebardi and its common-law predecessors. As the Fifth Circuit ruled, “Congress intended in both the FCPA and the Mann Act to deter and punish certain activities which necessarily involved the agreement of at least two people, but Congress chose in both statutes to punish only one party to the agreement.” Id. at 833. Thus, “the very individuals whose participation was required in every case — the foreign officials accepting the bribe — were excluded from prosecution for the substantive offense.” Id. at 835. In such a case, it makes sense to conclude that Congress must have intended to impose liability only on the American bribe giver, and not on the foreign official who receives the bribe, even though the latter plainly facilitates the criminal conduct of the bribe giver. The legislative history cited in the Court‘s opinion corroborates that intuition. If anything is clear from that history, it is that Congress was concerned about intruding into foreign sovereignty by attempting to punish under American law foreign officials taking bribes from Americans on their own soil.

It is not at all clear to me, however, that Hoskins benefits from the Gebardi principle. Although every bribe given by an American company to a foreign government official in return for favorable official treatment inherently involves such a foreign official (the foreign official‘s receipt of the bribe is an “inseparable incident,” in Justice Stone‘s phrase, of the American bribe-giver‘s violation of the FCPA), not every such bribe involves the direction of the American company‘s action, or the provision of expert advice on how to execute the scheme, by an executive of a foreign parent of the American company. Thus, while Congress could not help but be aware that every bribe given abroad by an affiliate of an American entity involved by definition a foreign official who received the bribe, there is no particular reason to think that the conceivable (but somewhat arcane) case of a foreign parent company‘s executives who directed, supervised, or assisted an American company to pay a bribe was necessarily present to the minds of the members of Congress who voted to adopt the FCPA.

II

If this case involved an ordinary criminal statute with purely domestic applications, then, I would likely reach a different conclusion here. But as Justice Jackson also explained in Morissette, while in many cases Congress can best be understood to have created federal criminal law against the backdrop of traditional common-law principles of criminal liability, in some instances Congress faces novel situations not anticipated by the common law of crimes, and legislates “to call into existence new duties and crimes.” 342 U.S. at 253. In doing so, Congress may have to deal with issues that are unique to federal law.

The FCPA, for example, is not an ordinary domestic criminal law, but a novel expansion of criminal liability to impose duties on American businesses to conform to domestic ethical standards even when they operate beyond our borders, in lands with different cultures, laws, and traditions. I agree with my colleagues that the extraterritorial effects of the FCPA require us to exercise particular caution before extending its reach even farther than that expressly declared by the statutory text. Although the FCPA explicitly and by design applies extraterritorially, the same concerns that generate a presumption against implying extraterritorial application of United States law, see Morrison v. National Australia Bank, Ltd., 561 U.S. 247 (2010) warrant special caution in applying normal principles of accessorial liability when Congress has delineated the particular circumstances in which the statute applies abroad.

In adopting the FCPA, Congress sought to criminalize wrongful conduct by Americans and those who in various ways work with Americans, while avoiding unnecessary imposition on the sovereignty of other countries whose traditions and laws may differ from our own. The legislative history described in the Court‘s opinion demonstrates that, in confronting the delicate line-drawing exercises involved in balancing these concerns, Congress intended to limit the overseas applications of the statute to those that it explicitly defined. The particular situation of someone like Hoskins, who is alleged to have participated in directing or assisting Americans to violate the FCPA, was not explicitly discussed by Congress, and such a case might well never have occurred to those drafting the law. The extraterritorial application that primarily concerned Congress was the potential intrusion into the sovereignty of the nations whose officials were bribed. But we do not sit to decide how Congress might have written the law if it had specifically considered this case. We can only apply the law that Congress did write, which limits the extraterritorial application of the FCPA to specific cases that do not include Hoskins‘s situation.

I therefore think that a combination of the presumption against extraterritorial application of United States law generally discussed in Part II D of the majority opinion, and the legislative history, discussed in Part II C 3 demonstrating that Congress drew lines in the FCPA out of specific concern about the scope of extraterritorial application of the statute, warrant affirmance of the district court‘s partial dismissal of the indictment in this case.

III

I would be remiss, finally, not to note that Congress might want to revisit the statute with this case in mind, as the result we reach today seems to me questionable as a matter of policy. The FCPA represents an effort by the United States to keep its own nationals free of corruption when dealing in foreign countries where corruption is endemic. Such corruption undermines the ethical foundations of American businesses, and risks accustoming American businesspeople and corporations to corrupt practices that they encounter abroad, with the attendant possibility of importing back to the United States practices they become familiar with in countries with less developed principles of the rule of law and the transparency and impartiality of government regulation.

Moreover, by embroiling American companies in the corrupt activities of foreign officials, such bribery tends to perpetuate the corruption of developing nations, to the long-run disadvantage of the United States both in foreign policy (by associating the United States and its citizens and businesses with unpopular corrupt regimes) and in commerce (by perpetuating the corruption “tax” levied on all those who do business with such regimes). See Court Op. 50–51 & n.7 (discussing purposes of the FCPA).

As noted above, these important purposes must be balanced against a concern about intruding into foreign sovereignty. It is thus not surprising that Congress chose to stop American businesses from engaging in the corrupt cultures of countries where they may do business, but did not attempt to reform those countries themselves by punishing their public officers.

If the charges in the indictment are proved true, however, the prosecution of Hoskins does not implicate that concern. Hoskins was not an official of a corrupt foreign government, operating in a legal and business culture distinct from that of the United States and other Western democracies. He was, rather, a citizen of the United Kingdom, employed by a British subsidiary of the French parent company of the American entity that allegedly paid bribes to Indonesian legislators to secure business for the American company, working in France from the offices of a French subsidiary of the same French parent. Both his country of citizenship (the United Kingdom) and the country where he worked and where the company whose interests he was ultimately advancing was incorporated (France) are signatories of the Organisation for Economic Co-operation and Development (“OECD“) Convention on Combating Bribery of Foreign Public Officials in International Business Transactions, and are thus committed, as is the United States, to enacting legislation along the lines of the FCPA.⁵ Although any prosecution of a foreign national for actions taken on foreign soil raises some concern about possible friction with other countries, the prosecution of someone in Hoskins‘s position does not threaten a foreign country‘s sovereign power to select, retain, and police the officials of its own government, nor does it conflict with the policies of the countries involved.

Nor were the effects of Hoskins‘s alleged actions felt only in his own countries of citizenship and employment. Hoskins is alleged to have been part of the team that reached into the United States to counsel and procure the commission of an American crime by an American company, and to assist that company in executing bribes in violation of American law. A prosecution on these facts does not evince an effort by the United States to “rule the world,” RJR Nabisco, 136 S. Ct. at 2100, but rather an effort to enforce American law against those who deliberately seek to undermine it.

Moreover, the FCPA explicitly contemplates the prosecution of at least some foreign nationals who operate entirely abroad, in that it penalizes foreign nationals who act as the agents of American companies in paying bribes abroad. Thus, for example, if Alstom U.S. had channeled its bribes to Indonesian officials through Indonesian citizens who were low-level Alstom employees in Indonesia, the FCPA would appear to penalize those employees. Indeed, our decision today leaves intact the possibility that Hoskins himself may be convicted under this indictment for violating the FCPA, if the government establishes that he functioned as the agent of the American company, rather than as one who directed the actions of the American company in the interests of its French parent company. That seems to me a perverse result, and one that is unlikely to have been specifically anticipated or intended by Congress. It makes little sense to permit the prosecution of foreign affiliates of United States entities who are minor cogs in the crime, while immunizing foreign affiliates who control or induce such violations from a high perch in a foreign parent company. That is the equivalent of punishing the get-away driver who is paid a small sum to facilitate the bank robber‘s escape, but exempting the mastermind who plans the heist.

It is for Congress to decide whether there are sound policy reasons for limiting the punishment of foreign nationals abroad to those who are agents of American companies, rather than to those who make American companies their agents. Our only task is to enforce the laws as Congress has written them. While I see no reason to believe that Congress specifically contemplated a case quite like this one, I agree that the lines drawn in the FCPA, which were crafted specifically to define when extraterritorial prosecutions would be permitted, do not encompass the present case. I therefore concur in the opinion and judgment of the Court.

Notes

1

Because the question before us is whether conspiracy and complicity charges can be used to extend liability beyond the categories delineated in the statute, we assume that Hoskins is not an agent of Alstom U.S. only for the sake of arguments advanced on appeal and express no views on the scope of agency under the FCPA. See Court Op. 4 n.1. The district court dismissed the indictment only insofar as it relies on the conspiracy and aiding and abetting theories, and left the charges against Hoskins intact insofar as he is charged as an “agent” of an American entity.

2

Although not alleged in the indictment, the government also represents that one consultant was “based” in Maryland. For example, 18 U.S.C. § 656 makes it a crime for an “officer, director, agent or employee of” certain banking institutions to “embezzle[ ] . . . or misappl[y]” the bank‘s funds. But it would be quite peculiar if the language of the law defining the kinds of persons who could be the principals of an embezzlement scheme precluded from liability, by that very definition, those who aid or induce a bank officer to engage in acts of embezzlement. Thus, if a bank officer persuades a friend with computer expertise to write code that could be insinuated into the bank‘s computers to divert funds into the officer‘s account, in exchange for a share of the proceeds of the crime, the computer-savvy friend would of course be punishable for knowingly joining and assisting the criminal plan.

3

The remainder of the indictment, which is not at issue in this appeal, charges Hoskins with one count of conspiracy to commit money, and four counts of money laundering. The latter aspect of the exception largely accounts for United States v. Amen, 831 F.2d 373 (2d Cir. 1987). The “continuing criminal enterprise” or “kingpin” statute, 21 U.S.C. § 848, is essentially a sentence-enhancing statute that identifies some drug dealers as worthy of increased punishment because of the extent and magnitude of their criminal acts. To apply accessorial liability as a way of bumping the kingpin‘s henchmen up to the same level of punishment because they help him in his enterprise would disrupt the legislature‘s intention to impose graduated punishments according to various conspirators’ different levels of involvement in the narcotics trade.

4

See American Law Institute, Model Penal Code and Commentaries, § 2.06 at 323 (1985) (“Many crimes are designed to control the conduct of persons who occupy special positions, and thus can only be committed by those who actually occupy the designated position. It is universally held, on the other hand, that one who assists an occupant of the designated position in the commission of the offense Indeed, the language of the second exception is closely derived from that of Justice Stone in Gebardi, who wrote:

We think it a necessary implication of that policy that when the Mann Act and the conspiracy statute came to be construed together, as they necessarily would be, the same participation which the former contemplates as an inseparable incident of all cases in which the woman is a voluntary agent at all, but does not punish, was not automatically to be made punishable under the latter. It would contravene that policy to hold that the very passage of the Mann Act effected a withdrawal by the conspiracy statute of that immunity which the Mann Act itself confers.

287 U.S. at 123 (emphasis added).

5

“Logic and precedent dictate that the starting point in every case involving construction of a statute is the language itself.” Greyhound Corp. v. Mt. Hood Stages, Inc., 437 U.S. 322, 330 (1978) (brackets and internal quotation marks omitted). “In evaluating ambiguity we look to the statutory scheme as a whole and place the particular provision within the context of that statute.” Raila v. United States, 355 F.3d 118, 120 (2d Cir. 2004). “As a general matter, we may consider reliable legislative history where, as here, the statute is susceptible to divergent understandings and, equally important, where there exists authoritative legislative history that assists in discerning what Congress actually meant.” United States v. Gayle, 342 F.3d 89, 94 (2d Cir. 2003). As the Court‘s opinion describes, the OECD Convention was a diplomatic initiative of the United States to encourage its principal trading partners and competitors in international commerce to enact anti-corruption laws similar to those of the FCPA. The 1998 FCPA amendments were adopted in part to conform American law to the Convention‘s requirements. See Court Op. 55–63.

6

Wharton‘s Rule applies only to conspiracy, which means that there could be daylight between it and the government‘s proposed rule that neither conspiracy nor complicity liability will lie where “the defendant‘s consent or acquiescence is inherent in the [substantive] offense.” Appellant‘s Opening Br. at 24 (internal quotation marks omitted). But within Gebardi itself, the government‘s proposed rule would have operated identically to Wharton‘s Rule, since Gebardi dealt only with a conspiracy charge. And since the Supreme Court said in Gebardi that it was not relying on Wharton‘s Rule, the government‘s rule cannot be defended as the rule the Supreme Court meant to adopt.

7

It is noteworthy that the Committee Report accompanying the amended House bill in September of 1977 discussed the legislation‘s purposes, and the need for the legislation, significantly more broadly than did the Senate‘s comparable report:

The payment of bribes to influence the acts or decisions of foreign officials, foreign political parties or candidates for foreign political office is unethical. It is counter to the moral expectations and values of the American public. But not only is it unethical, it is bad business as well. It erodes public confidence in the integrity of the free market system. It short-circuits the marketplace by directing business to those companies too inefficient to compete in terms of price, quality or service, or too lazy to engage in honest salesmanship, or too intent upon unloading marginal products. . . . Bribery of foreign

officials by some American companies casts a shadow on all U.S. companies. The exposure of such activity can damage a company‘s image, lead to costly lawsuits, cause the cancellation of contracts, and result in the appropriation of valuable assets overseas. Corporate bribery is also unnecessary. . . . Corporate bribery also creates severe foreign policy problems for the United States. The revelation of improper payments invariably tends to embarrass friendly governments, lower the esteem for the United States among the citizens of foreign nations, and lend credence to the suspicions sown by foreign opponents of the United States that American enterprises exert a corrupting influence on the political processes of their nations. . . . Finally, a strong antibribery statute would actually help U.S. corporations resist corrupt demands.

8

In United States v. Bodmer, 342 F. Supp. 2d 176, 188 (S.D.N.Y. 2004), the district court ruled that the phrase “otherwise subject to the jurisdiction of the United States” did not expand the persons subject to liability under this section, because the phrase is superfluous. In any case, the government does not here suggest that it creates liability over foreign nationals.

9

Neither party relies significantly on changes to the statute that occurred in 1988, so the Court need not analyze that set of amendments.

10

The government cites European Cmty. v. RJR Nabisco, Inc., 764 F.3d 129, 142 (2d Cir. 2014), rev‘d, 136 S. Ct. 2090 (2016) for the proposition that if all elements of a statute are satisfied, “that statute is violated even if some further conduct contributing to the violation occurred outside the United States.” [Gray 14-15.] But the government neglects to mention that the text directly preceding this statement instructs that the proposition only applies to a statute “that does not apply extraterritorially.” Id. Clearly, the FCPA applies extraterritorially, so the proposition the government cites is inapplicable here.

Case Details

Case Name: United States v. Hoskins

Court Name: Court of Appeals for the Second Circuit

Date Published: Aug 24, 2018

Citation: 902 F.3d 69

Docket Number: Docket 16-1010-cr; August Term, 2016

Court Abbreviation: 2d Cir.