Adam Davidson’s New Yorker piece from earlier this month, “Donald Trump’s Worst Deal,” has been getting a lot of attention, and deservedly so. The article, which focuses on the Trump Organization’s involvement in a hotel deal in Baku, Azerbaijan, does a very nice job highlighting the troubling background of the Trump Organization’s Azeri business partners and the Trump Organization’s casual approach (to put it charitably) to due diligence. However, the piece also suggests that the Trump Organization’s involvement with the Baku hotel deal may have violated the Foreign Corrupt Practices Act (FCPA), and many of the follow-up discussions of Mr. Davidson’s piece have repeated this claim (see, for example, here and here). On this point, not everyone agrees. Professor Mike Koehler, for example, wrote a lengthy critique of Mr. Davidson’s discussion of the FCPA issues, concluding that nothing in the facts as reported in the article suggests that the Trump Organization violated the FCPA – and that many of the article’s assertions to the contrary are based on incomplete and misleading representations of the statute and prior case law.

After having finally had a chance to read Mr. Davidson’s article carefully, it seems to me that Professor Koehler has the better of the argument—mostly. Much of the discussion of potential FCPA violations in Mr. Davidson’s article is confused and potentially misleading. That said, I do think there’s at least one plausible basis for the claim that the Trump Organization may have violated the FCPA in this case.

Here’s my take:

First, a quick-and-dirty summary of what the FCPA does and doesn’t prohibit:

• The FCPA prohibits any entity subject to its jurisdiction (typically either a US firm or a foreign firm that lists on a US exchange) from giving or offering anything of value to a foreign public official for purposes of obtaining or retaining business.
• The FCPA also prohibits giving anything of value to a third party, while “knowing” that this valuable asset will in turn be offered or given to a foreign public official in order to help the original giver obtain or retain business. “Knowledge” here is construed broadly to preclude a defense of willful ignorance (the “head in the sand” approach).
• Standard principles of U.S. criminal liability mean that corporations can be liable for FCPA violations by their agents, including their employees and potentially their subsidiaries and partners as well, if the corporation exercised sufficient direction control such that the partner or subsidiary can be considered an agent of the corporation.
• The FCPA does not, however, prohibit a covered corporation from doing business with—or acquiring—a firm that has engaged in conduct that would have been considered an FCPA violation if performed by an entity covered by the FCPA.

With that as background, here’s my understanding of the rough chronology of events surrounding the Baku tower deal:

• The tower project was announced, and construction began, in 2008. From Mr. Davidson’s reporting, which I’ll assume to be accurate, the whole process of approving and building the tower was a veritable orgy of corruption. The firms that owned the tower were controlled by close family members of government officials—most notably the shady transportation minister. The developers acquired the land at cut-rate prices thanks to the government’s intervention. The construction process was, as Mr. Davidson’s sources put it, “nakedly corrupt,” with “giant pile[s] of cash” changing hands, and generally benefiting the business interests connected to Transportation Minister’s family. In many cases, bribes were paid directly to government officials, such as tax officials, inspectors, and customs officers, who would often simply come by “to pick up envelopes of cash.”
• The Trump Organization got involved in the project in 2012. In May of that year, the Trump Organization signed a licensing deal that provided that the tower would be converted into a luxury hotel, allowed the Azeri company that owns the tower to use the Trump name, gave the Trump Organization the right to manage the hotel once it opened (for a fee tied to the hotel’s performance), and also involved the Trump Organization in design issues. Ultimately, though, the Trump Organization severed its ties with the hotel project in December 2016, after Trump was elected president.

So, we can divide the analysis of potential FCPA violations into two periods: from 2008 to May 2012, and from May 2012 to December 2016.

With respect to the first period—the period for which Mr. Davidson’s piece provides the most damning evidence of outright corruption—the Trump Organization cannot be liable under the FCPA, for the simple reason that the Trump Organization was not involved in the project at that time. The FCPA does not prohibit corporations or individuals from doing business with partners who in the past engaged in conduct that may have violated the FCPA. Indeed, as the U.S. government’s FCPA Resource Guide makes clear, even if a U.S. company acquires a foreign company not previously subject to the FCPA that had in the past paid bribes to government officials, the parent company is not liable under the FCPA for these past bribes (though after the acquisition the parent might well be liable for future bribes paid by the newly-acquired subsidiary). Now, the fact that a prospective business partner has engaged in corrupt activity in the past should be a red flag for any company subject to the FCPA, because going forward, the latter company may be exposed to FCPA risk for its partner’s conduct. Here it seems that the Trump Organization was troublingly lax. But it’s important to sharply distinguish (as Mr. Davidson’s article unfortunately does not) between the post-May 2012 period, when the Trump Organization could at least potentially be on the hook for the conduct of its Azeri partner, and the pre-May 2012 period, where the conduct of the Azeri partners cannot possibly subject the Trump Organization to FCPA liability.

Now, let’s turn to the period between May 2012 and December 2016, when the Trump Organization was involved in the Baku tower project. Could anything that happened during this period, including conduct by the Azeri partners, give rise to FCPA liability for the Trump Organization? Yes, possibly. But it’s not at all clear, and unfortunately Mr. Davidson’s treatment of the legal issue seems to confuse or conflate a few distinct potential bases for liability. As I understand it, in a situation like this, there are two main ways that the US partner could get in FCPA trouble for bribes paid by the foreign partner:

• First, as noted above, if the US partner provided “anything of value” to the foreign partner, while “knowing” that some or all of this would be passed onto a government official as a bribe, then the US partner is liable under the FCPA. However, that is not what happened here. It is certainly true, as Mr. Davidson’s article points out, that the Trump Organization provided “things of value” to its Azeri business partner (“its famous brand, its command of the luxury market, its technical advice”). But that’s irrelevant: The question is whether the Trump Organization provided something of value to a foreign government official—whether directly or funneled through a third-party intermediary. There’s no evidence or assertion in Mr. Davidson’s article that it did so.
• Second, the US partner may be liable under the FCPA for bribes paid by the foreign partner if the latter can be considered the former’s agent, or if the two partners can be deemed co-conspirators in the bribery scheme. This is the theory of FCPA liability that seems most viable in the case of the Trump Organization’s dealings in Baku. For this theory to apply, two conditions must hold: (1) bribery of Azeri government officials—the handing out of “envelopes of cash” to tax collectors and inspectors and the like, for example—continued between May 2012 and December 2016, and (2) the Trump Organization’s involvement in the project needs to have been sufficiently extensive that the Trump Organization could be considered responsible for the Azeri partner’s conduct. This is the one possible theory of liability that Professor Koehler’s otherwise thorough critique omits, and the reason that I’m at least open to the possibility that there was indeed an FCPA violation here. On the first condition, while all of Mr. Davidson’s discussion focuses on the first period, it strikes me as highly likely that bribery of Azeri government officials continued after May 2012, especially since construction on the hotel was not yet complete. With respect to the second condition, the issue is much harder. Professor Koehler correctly notes that we don’t have (to his knowledge or mine) any FCPA cases involving a licensor being responsible for the conduct of a licensee, nor do we have any cases involving the similar situation of a franchisor being liable for the conduct of a franchisee. At the same time, as Mr. Davidson’s article notes, the Trump Organization’s degree of direct involvement in the Baku project was unusually extensive for a licensor. So at the very least, it’s a hard legal question. Of course, it’s all hypothetical, as the odds that the U.S. government will prosecute the Trump Organization strike me as vanishingly small.