Harvard Law Review Harvard Law Review Harvard Law Review


United States v. Esquenazi

Eleventh Circuit Defines "Government Instrumentality" Under the FCPA.

Enacted in 1977, the Foreign Corrupt Practices Act (FCPA) prohibits companies from corruptly paying foreign officials in order to obtain or retain business.1×1. See 15 U.S.C. § 78dd-2(a)(1) (2012). While the FCPA’s antibribery provision lay practically dormant through the turn of the millennium, the past decade has seen “skyrocketing number[s] of foreign bribery investigations and prosecutions” by the DOJ and SEC.2×2. Joel M. Cohen et al., Under the FCPA, Who Is a Foreign Official Anyway?, 63 Bus. Law. 1243, 1247 (2008). This sea change in enforcement has taken place against the backdrop of persistent confusion regarding the scope of one of the statute’s key terms: “foreign official.”3×3. See, e.g., Mike Koehler, The Façade of FCPA Enforcement, 41 Geo. J. Int’l L. 907, 916 (2010) (“[N]o element is more urgently in need of judicial scrutiny than . . . ‘foreign official’ . . . .”). While the FCPA defines “foreign official” as “any officer or employee of a foreign government or any department, agency, or instrumentality thereof,”4×4. 15 U.S.C. § 78dd-2(h)(2)(A). the generic concept of “instrumentality” remained undefined. In the first thirty-six years of the FCPA’s history, not a single court of appeals was called upon to clarify the meaning of this elusive, yet fundamental, term.5×5. Petition for Writ of Certiorari at 11, Esquenazi v. United States, 135 S. Ct. 293 (2014) (No. 14-189). Amidst such uncertainty, some district courts have offered nonexhaustive lists of pertinent considerations to help juries resolve the “question of fact” of whether a given entity is an instrumentality. E.g., United States v. Carson, No. SACR 09-00077-JVS, 2011 WL 5101701, at *3–4 (C.D. Cal. May 18, 2011). Recently, in United States v. Esquenazi,6×6. 752 F.3d 912 (11th Cir. 2014), cert. denied, 135 S. Ct. 293 (2014). the Eleventh Circuit received such an opportunity and defined “instrumentality” under the FCPA as “an entity controlled by the government of a foreign country that performs a function the controlling government treats as its own.”7×7. Id. at 925. Unfortunately, the court’s indeterminate and unwieldy two-pronged instrumentality test fails to provide much-needed judicial guidance to businesses and regulators interpreting the Act.

Joel Esquenazi and Carlos Rodriguez co-owned and directed Terra Telecommunications Corp. (Terra), a Florida corporation that purchased phone time from foreign vendors and resold the minutes to U.S. customers.8×8. Id. at 917. One of Terra’s main vendors was Telecommunications D’Haiti, S.A.M. (Teleco), a company that initially received a national monopoly on telecommunication services from the Haitian government.9×9. Id. In the 1970s, the National Bank of Haiti gained ninety-seven percent ownership of Teleco. Id. at 918. Between 2009 and 2010, Teleco was privatized. Id. By 2001, Terra owed Teleco over $400,000.10×10. Id. Upon Esquenazi’s offer, Teleco officials agreed to ease Terra’s debt in exchange for receiving side payments: between 2001 and 2005, Terra paid several Teleco officials over $800,000 in exchange for continued business and a reduction of its bills by over $2 million.11×11. Id. at 918–19. Terra funneled the payments through sham consulting companies. Id. at 918.

In 2009, a grand jury indicted Esquenazi and Rodriguez on charges of conspiracy, substantive FCPA violations, and concealment money laundering.12×12. Id. at 917. Esquenazi and Rodriguez proceeded to trial in the Southern District of Florida and were found guilty on all counts.13×13. Id. at 919. Shortly after the convictions, the court received two declarations from the Haitian Prime Minister clarifying Teleco’s status under Haitian law. Id. at 919–20. The first stated, “Teleco has never been . . . a State enterprise”; the second noted that while “there exists no law specifically designating Teleco as a public institution . . . this does not mean that Haiti’s public laws do not apply to Teleco.” Id. On the basis of these declarations, Esquenazi and Rodriguez moved for acquittal and a new trial, in vain. Id. at 920. In 2011, the court sentenced Rodriguez to a seven-year term, and Esquenazi to fifteen years.14×14. Id. Both defendants appealed, arguing inter alia that the FCPA does not cover corrupt payments to state-owned enterprises (SOEs) that provide commercial services, such as Teleco.15×15. Id.

The Eleventh Circuit affirmed the convictions.16×16. Id. at 917. Writing for the panel, Judge Martin17×17. Judge Martin was joined by Judges Jordan and Suhrheinrich (sitting by designation). began by addressing the “central question”: what constitutes a government “instrumentality” for purposes of defining “foreign official”?18×18. Esquenazi, 752 F.3d at 920. “[M]indful of the needs of both corporations and the government for ex ante direction,” Judge Martin set out to fill the statute’s definitional gap.19×19. Id. at 925. She first considered the plain meaning of “instrumentality.”20×20. Id. at 920. While the term is susceptible to more than one meaning, she gleaned from dictionary definitions that an instrumentality need not be an actual part of the government but must nonetheless perform “a government function at the government’s behest.”21×21. Id. at 921. Applying the canon of noscitur a sociis — a word is known by the company it keeps — Judge Martin derived from the juxtaposition of “instrumentality” with “agency” and “department” that an entity will qualify as an instrumentality only if it is both (i) “under the control” of and (ii) “doing the business of” the government.22×22. Id. at 922.

To determine whether commercial services may ever constitute “government business,” Judge Martin turned to the FCPA’s broader statutory context. Because an FCPA exception excuses payments offered “to secure the performance of a routine governmental action”23×23. 15 U.S.C. § 78dd-2(b) (2012). and lists “provi[sion of] phone service” as such an action,24×24. Id. § 78dd-2(h)(4)(A)(iv). to interpret “instrumentality” as categorically excluding phone service providers would render meaningless a portion of the statute.25×25. Esquenazi, 752 F.3d at 922. Judge Martin concluded that the FCPA contemplates that “in some instances,” SOEs providing commercial services count as “instrumentalities.”26×26. Id. at 922–23.

Judge Martin found further support for construing “instrumentality” to include SOEs in the 1998 FCPA amendments,27×27. See id. at 923–24. intended to implement a recently ratified Organisation for Economic Cooperation and Development (OECD) convention that required state parties to criminalize bribery of foreign officials under certain circumstances.28×28. Organisation for Economic Co-operation and Development, Convention on Combating Bribery of Foreign Public Officials in International Business Transactions, Dec. 17, 1997, S. Treaty Doc. No. 105-43 (1998) [hereinafter OECD Convention]. The Convention’s commentary defines “foreign public official”29×29. Id. art. 1, para. 4. to include employees of any public enterprise that does not “operate[] . . . substantially [like] a private enterprise,” and defines “public enterprise” as “any enterprise . . . over which a government . . . exercise[s] a dominant influence.”30×30. Id. cmts. 14 & 15. Because Congress responded by leaving its “foreign official” definition unchanged save for one unrelated amendment,31×31. The sole alteration was the addition of “officer or employee of . . . a public international organization.” 15 U.S.C. 78dd-2(h)(2)(A) (2012) (emphasis added); Esquenazi, 752 F.3d at 923. Judge Martin concluded that Congress considered the FCPA’s 1977 version to already match the Convention.32×32. Esquenazi, 752 F.3d at 923. With this equivalency established, Judge Martin could not limit “instrumentality” to “entities that perform . . . core government functions” because such an interpretation would categorically exclude government-controlled commercial services and thereby put the United States “out of compliance with its international obligations.”33×33. Id. at 924. To justify her reliance on the OECD Convention, Judge Martin pointed to the Fifth Circuit’s assertion that the FCPA’s preexisting language “should be construed to cover the Convention’s mandate.” Id. (citing United States v. Kay, 359 F.3d 738, 754 (5th Cir. 2004)). Instead, the court had to heed the Supreme Court’s instruction that “the concept of a ‘usual’ or ‘proper’ governmental function changes over time and varies from nation to nation.”34×34. Id. (quoting First Nat’l City Bank v. Banco Para El Comercio Exterior de Cuba, 462 U.S. 611, 633 n.27 (1983)) (internal quotation marks omitted).

With such groundwork laid, Judge Martin defined “instrumentality” under the FCPA as “an entity controlled by the government of a foreign country that performs a function the controlling government treats as its own.”35×35. Id. at 925. Recognizing that this definition’s two elements — control and function — require fact-specific inquiries, Judge Martin drew upon the OECD Convention and U.S. Supreme Court case law to generate nonexhaustive lists of factors relevant to each determination.36×36. See id. at 925–27. Regarding control, courts and juries may consider: (i) the entity’s formal designation; (ii) whether the government has a majority stake and/or (iii) authority to hire and fire; (iv) the extent to which the government treats the entity’s finances as its own and funds the entity when needed; and (v) the length of time these indicia have existed.37×37. Id. at 925. Regarding function, courts and juries may consider: (i) whether the entity has a monopoly and (ii) provides services to the public at large; (iii) whether the government subsidizes the entity; and (iv) whether the “public and the government . . . perceive the entity to be performing a governmental function.”38×38. Id. at 926.

Armed with this two-pronged test, Judge Martin succinctly dispensed with the appellants’ challenges to their FCPA convictions. First, she found that the jury instructions “substantially cover[ed]” the factors outlined in the court’s instrumentality definition.39×39. Id. at 928. Second, she concluded that the evidence was sufficient to support the jury’s finding that Teleco was an instrumentality of the Haitian government.40×40. Id. at 928–29. Judge Martin stressed the following: Haiti’s grant to Teleco of a monopoly over telecommunications service; the National Bank of Haiti’s ninety-seven percent ownership of Teleco; the Haitian President’s power to appoint all of Teleco’s board members; and expert testimony that Teleco belonged “totally to the state” and “everyone consider[ed] Teleco as a public administration.” Id. (alteration in original) (internal quotation marks omitted). Third, she repudiated Esquenazi’s assertion that the FCPA was unconstitutionally vague.41×41. See id. at 929. “Although . . . there may be entities near the definitional line for ‘instrumentality,’” the FCPA was not vague as applied to Esquenazi’s conduct. Id. Fourth, she found the evidence sufficient to conclude that Rodriguez knew Teleco was a government instrumentality.42×42. See id. at 930–32. Finally, Judge Martin proceeded to dispense with the appellants’ remaining challenges to their conspiracy and money laundering convictions, as well as various aspects of their sentences. See id. at 932–39. Accordingly, Judge Martin affirmed the convictions.43×43. Id. at 939.

In Esquenazi, the Eleventh Circuit failed to take advantage of an unprecedented opportunity to clarify a critical term of the FCPA. The court’s malleable, fact-intensive “instrumentality” test is too impracticable to provide real guidance to the business community and FCPA regulators. Moreover, the current trend in FCPA enforcement — out-of-court settlements rather than adjudication — forecloses the Eleventh Circuit’s vision of future courts gradually polishing the tentative test it has just announced. The court should have staked out clearer boundaries for the “instrumentality” category. For instance, it could have imported a more administrable test from the OECD Convention, upon which it already extensively relied.

The Esquenazi court faced a status quo of widespread uncertainty concerning the meaning of the term “instrumentality” within the FCPA’s definition of “foreign official.”44×44. Cohen et al., supra note 2, at 1243, 1245, 1250; Alexander L. Harisiadis, Comment, Foreign Official, Define Thyself: How to Define Foreign Officials and Instrumentalities in the Face of Aggressive Enforcement of the Foreign Corrupt Practices Act, 62 Cath. U. L. Rev. 507, 516–17 (2013). The recently published FCPA Resource Guide45×45. U.S. DOJ & U.S. SEC, A Resource Guide to the U.S. Foreign Corrupt Practices Act (2012). and DOJ advisory opinions have generated little guidance.46×46. See, e.g., Harisiadis, supra note 44, at 520 (“To date, no FCPA Opinion Procedure Release clearly defines an instrumentality.”); Note, Corporate Victims of “Victimless Crime”: How the FCPA’s Statutory Ambiguity, Coupled with Strict Liability, Hurts Businesses and Discourages Compliance, 54 B.C. L. Rev. 393, 408 (2013) (observing that the Resource Guide “provides no hypothetical examples or detailed guidance”). On one hand, critics roundly accuse the SEC and DOJ of exploiting statutory ambiguity to pursue overly aggressive, far-reaching FCPA enforcement that hurts U.S. businesses, spawns excessive compliance programs, and undermines U.S. influence overseas.47×47. See, e.g., Koehler, supra note 3, at 1002–04; Matthew W. Muma, Note, Toward Greater Guidance: Reforming the Definitions of the Foreign Corrupt Practices Act, 112 Mich. L. Rev. 1337, 1347–51 (2014). But see Philip Urofsky et al., How Should We Measure the Effectiveness of the Foreign Corrupt Practices Act? Don’t Break What Isn’t Broken — The Fallacies of Reform, 73 Ohio St. L.J. 1145 (2012) (arguing that “the general scope of the FCPA is fairly well-delineated and understood,” id. at 1147, and that the corporate community’s commitment to internal compliance programs demonstrates that the FCPA has actually been effective, see id. at 1153, 1179). On the other hand, one may discount the damage caused by such alleged over-deterrence, insofar as the underlying prohibited conduct — bribery — is widely deemed a culpable act in the United States. Ultimately, these lines of argument conflate overbroad enforcement with statutory vagueness.48×48. See Matthew Stephenson, U.S. v. Esquenazi: Some Knee-Jerk Reactions, Global Anticorruption Blog (May 20, 2014), http://globalanticorruptionblog.com/2014/05/20/u-s-v-esquenazi-some-knee-jerk-reactions [http://perma.cc/S22R-NQK6]. Even proponents of higher levels of FCPA enforcement may deem greater statutory clarity desirable. See Lena E. Smith, Note, Is Strict Liability the Answer in the Battle Against Foreign Corporate Bribery?, 79 Brook. L. Rev. 1801, 1804 (2014) (“[A]dvocates for reform should not focus on scaling back FCPA enforcement, but rather on removing part of the FCPA’s ambiguity . . . .”). Whether broad or narrow, FCPA boundaries should be clearly established. Predictability and fair notice are particularly valuable for a criminal statute that threatens individuals with incarceration for conduct that may constitute illegal corruption in some countries but standard business practice in others.49×49. See Steven R. Salbu, Extraterritorial Restriction of Bribery: A Premature Evocation of the Normative Global Village, 24 Yale J. Int’l L. 223, 234 (1999) (“In [some] countries and cultural contexts . . . gifts that would be presumed illegal or corrupt under American norms may be considered merely a form of common courtesy or . . . relational etiquette.”); id. at 234–40. Independently of the contentious issue of statutory breadth, the FCPA regime will remain flawed as long as its ill-defined terms prevent corporate officers from ascertaining their exposure to criminal liability.

While professing to address such statutory indeterminacy,50×50. See Esquenazi, 752 F.3d at 925. the Esquenazi court ultimately espoused a multifaceted, malleable framework that privileges accuracy over predictability.51×51. The court noted: “It would be unwise and likely impossible to exhaustively answer” “what constitutes control and what constitutes a function the government treats as its own.” Id. Functionally, it adopted a totality-of-the-circumstances approach by inviting future courts to tweak its “instrumentality” test’s nonexhaustive sets of nondispositive factors. Insofar as this open-ended definitional method helps avoid the over- and under-inclusiveness that inhere in bright-line rules,52×52. See, e.g., Kathleen M. Sullivan, The Supreme Court, 1991 Term — Foreword: The Justices of Rules and Standards, 106 Harv. L. Rev. 22, 58–59 (1992). For an argument that totality-of-the-circumstances tests do not actually achieve fairer results, see Antonin Scalia, Essay, The Rule of Law as a Law of Rules, 56 U. Chi. L. Rev. 1175, 1182 (1989) (“[E]quality of treatment is . . . impossible to achieve; predictability is destroyed; judicial arbitrariness is facilitated . . . .”). the court purports to achieve fairer rulings by fostering sensitivity to the unique circumstances of each case.

However, even if flexible standards have strong merit, they are unsuitable for the realm of FCPA enforcement. First, the factfinding requirements in the control and function prongs of the court’s test will prove particularly unwieldy for courts and businesses alike. Whether a government “controls” an SOE is a complex question susceptible to highly nuanced answers. By treating majority-shareholder status and authority to hire and fire directors as merely two among a host of indicators, the court is forcing businesses and prosecutors to conduct extensive and at times indeterminate inquiries into the levels of de facto control governments directly and indirectly exert over SOEs.53×53. See, e.g., Muma, supra note 47, at 1346 (“Given how entwined the state is with private enterprise in many countries, it may prove virtually impossible for a foreign company to ascertain exactly how much control the local government has over any given entity.”). In some parts of the world, “ownership interest” is ascertainable but “degree of control” is not: “Because of the broad movement toward privatization in many developing countries, it remains unclear whether certain businesses are fully private or are still (at least to some extent) under the control of their respective governments.”54×54. Cohen et al., supra note 2, at 1267. In Russia, for example, the “revolving door” of personnel between the Kremlin and Gazprom, Russia’s largest energy company, reveals how “the line separating big business and the state is becoming so fine that it’s almost nonexistent.” Id. at 1269 n.156 (quoting Andrew E. Kramer, As Gazprom Goes, So Goes Russia, N.Y. Times, May 11, 2008, at BU1) (internal quotation mark omitted). Although juries are frequently expected to make judgments on fact-heavy issues, the analysis of overseas entities’ ownership structures may prove excessively difficult for laypersons.55×55. See Muma, supra note 47, at 1346. Regarding the function prong, the court’s confidence that it will be “relatively easy”56×56. Esquenazi, 752 F.3d at 925 n.8. to determine whether the foreign government and public generally view the entity as performing a governmental function seems misguided.57×57. In passing, the court suggested relying on “objective factors, like control, exclusivity, governmental authority to hire and fire, subsidization, and whether an entity’s finances are treated as part of the public fisc.” Id. However, such a practice would effectively gut the governmental function inquiry by tethering it to the control prong factors. Overall, the court’s test requires businesses, regulators, and juries to engage in indeterminate inquiries that, at least for some countries, are unlikely to yield conclusive results.

Second, the prevalence of out-of-court settlements of FCPA actions undermines the possibility of gradual judicial refinement and clarification of the Eleventh Circuit’s open-ended “instrumentality” definition. Under the FCPA regime’s climate of legal uncertainty, risk-averse corporate officers faced with the prospect of heavy fines and incarceration refrain categorically from going to trial.58×58. See Brief of Washington Legal Foundation and the Independence Institute as Amici Curiae in Support of Petitioners at 7, Esquenazi v. United States, 135 S. Ct. 293 (2014) (No. 14-189). As a result of the prevalence of out-of-court settlements of FCPA actions,59×59. See Koehler, supra note 3, at 932 (“[E]very corporate FCPA enforcement action over the last two decades has been resolved through a DOJ NPA [non-prosecution agreement], DPA [deferred prosecution agreement], plea (or combination thereof) or SEC settlement, and nearly every individual FCPA enforcement action has been resolved through a plea or SEC settlement.”). courts have few opportunities to develop case law that will clarify the contours of the FCPA’s reach. As Justice Scalia has argued, “[t]he idyllic notion of ‘the court’ gradually closing in on a fully articulated rule of law by deciding one discrete fact situation after another . . . cannot be applied to a court that will revisit the area in question with great infrequency.”60×60. Scalia, supra note 52, at 1178 (emphasis added). In the realm of FCPA enforcement, the lack of judicial scrutiny strongly favors incorporating at least some bright-line rules into the court’s factfinding test. Indeed, if it took thirty-seven years for a single appellate court to broach the meaning of “foreign official,” the Eleventh Circuit’s invitation for courts to iteratively polish its test appears to be little more than wishful thinking.

Moreover, the Eleventh Circuit had clearer alternative solutions available. Notably, it could have tailored its “instrumentality” test to mirror the OECD Convention’s definition of “public enterprise.”61×61. See Cohen et al., supra note 2, at 1270 (recommending that “the term ‘instrumentality’ in the FCPA’s definition of ‘foreign official’ [be defined] synonymously with the term ‘public enterprise’ in the OECD Anti-Bribery Convention”). Under the Convention, a “public enterprise” is any enterprise over which the government exercises a “dominant influence,” which is deemed to be the case whenever it “hold[s] the majority of the enterprise’s subscribed capital, control[s] the majority of [shareholder] votes . . . or can appoint a majority of the members of the . . . board.”62×62. OECD Convention, supra note 28, cmt. 14. By incorporating these three dispositive factors into its control inquiry, the court could have crafted a clearer test.63×63. Admittedly, the Convention’s bright-line factors are sufficient but not necessary indicators of government control. See id. The Convention leaves open the possibility that governments, as minority shareholders, may still exert sufficient influence for SOEs to qualify as “instrumentalities.” Nevertheless, the Convention establishes a clear baseline. Furthermore, incorporation of the OECD definition would have accorded with the court’s emphatic treatment of the Convention as a binding model for the FCPA.64×64. The court asserted that the purpose of the 1998 FCPA amendments was to ensure full compliance with international obligations under the Convention. Esquenazi, 752 F.3d at 924. Even if the court were to decline to match the OECD,65×65. The fact that other signatories to the Convention have interpreted its terms in inconsistent ways weakens the contention that the United States must closely adhere to its commentary. See Justin Epner, Note, Settling on an Interpretation of “Instrumentality” in the FCPA, 2013 Colum. Bus. L. Rev. 854, 902–04. scholars have proposed other solutions that would achieve greater determinacy than the court’s totality-of-the-circumstances approach.66×66. See id. at 894–99 (proposing a bright-line state ownership percentage threshold); Kayla Feld, Comment, Controlling the Prosecution of Bribery: Applying Corporate Law Principles to Define a “Foreign Official” in the Foreign Corrupt Practices Act, 88 Wash. L. Rev. 245, 275–79 (2013) (proposing to define control through U.S. corporate law principles and case law). Others have suggested that clarity should come from the executive branch rather than the courts. See Muma, supra note 47, at 1353–59 (proposing that the State Department provide country-specific guidance on which types of entities should qualify as “instrumentalities”).

In Esquenazi, the Eleventh Circuit received a rare opportunity to clarify the scope of a term that lies at the heart of a contentious criminal statute. Although the court purported to define “instrumentality” with an eye toward helping companies and regulators determine which SOEs fall within the FCPA’s reach, it ultimately provided unwieldy guidelines that lower courts are unlikely to refine. Operating against a backdrop void of judicial exegesis, risk-averse businesses will continue to settle enforcement actions exclusively out of court, leaving unabated the legal haze that currently envelops the FCPA.