Is it possible for violations of the Foreign Corrupt Practices Act (FCPA) and other domestic anti-corruption regimes to influence the outcomes of international investment arbitrations? In my last post, I showed how allegations of corruption could be relevant in investment treaty arbitration, both as a “sword” wielded by investors, and as a “shield” used by sovereign states. In this post, I consider how evidence from domestic anti-corruption proceedings could be used in arbitration, and what effects its use might have on the international system.
No arbitral tribunal has directly considered evidence from an FCPA investigation, but they have come very close. Consider the case of Siemens A.G. and Argentina. In 2007, Siemens won a large arbitral award from Argentina. However, about a year later, Siemens became the subject of a massive FCPA investigation, and U.S court filings revealed that the Siemens-Argentina deal on which the arbitration had centered had been obtained through bribery. In response, Argentina filed for an annulment of the arbitral award, and Siemens eventually agreed to cease attempts at recovering. Had settlement not occurred, the arbitral tribunal would have been forced to assess the legal weight of evidence from the FCPA proceeding.
For parties in arbitration seeking to prove corruption, piggy-backing off of domestic anti-corruption investigations could be very attractive. Proving corruption is notoriously difficult. Direct evidence of corruption is hard to come by, and states may have external incentives against revealing the full extent to which their internal workings are tainted by corruption. In World Duty Free v. Kenya — the canonical case on the effect of corruption in international arbitration — the corruption issue was essentially forced on the tribunal because the claimant admitted that it had engaged in bribery. This kind of situation seems unlikely to repeat itself, and tribunals’ own investigative powers are very limited, so evidence gleaned by domestic investigative authorities could be extremely valuable to claimants and states parties alike.
If FCPA violations do become relevant to the work of arbitral tribunals, this could have some interesting and potentially troubling implications.
First, although domestic anti-bribery statutes are becoming more common, most of the large-scale investigations are still undertaken by the U.S. As previous scholars have noted, FCPA enforcement decisions are not free of political influence and may in fact reflect U.S. foreign policy priorities. As a result, reliance on FCPA-based evidence (or evidence from any domestic anti-corruption proceeding, for that matter) could skew corruption allegations in investment treaty arbitration in a direction influenced by the policy priorities of the states that conduct their own corruption investigations. Thus, in the future, corruption allegations in arbitration could pose a challenge to ICSID’s legitimacy as a relatively unbiased forum for resolving investor-state disputes.
Second, and perhaps more interesting, the potential for secondary liability in international arbitration could affect the character of FCPA investigations and negotiations. Most of the deferred prosecution and non-prosecution agreements that resolve FCPA cases are products of extended negotiations between the DOJ/SEC and offending companies. The specter of secondary liability through arbitration could make certain companies less likely to admit to wrongdoing, or to sign agreements detailing their wrongdoing. Thus, if domestic anti-corruption violations do become relevant to ICSID arbitrations, the long-term effect could be more contentious and volatile FCPA settlement processes.
GAB | The Global Anticorruption Blog 
How much weight a deferred prosecution or non-prosecution agreement would be given in an arbitration proceeding – would such an agreement be admitted as evidence of corruption? Are there any sort of due process/rules of evidence protections in international arbitration that might mitigate the concerns you raise about the use of FCPA investigations?
Also, I’m wondering if the threat of secondary liability might have a beneficial effect in the long run, especially once countries aside from the US start investigating and prosecuting corruption more often – maybe the threat of more severe repercussions for corrupt activities would deter companies from engaging in corrupt activities in the first place, rather than incentivizing companies not to admit wrongdoing once it has already happened during negotiations with DOJ/SEC.
Would this secondary liability issue encourage corporations to take FCPA cases to trial in some cases?
It seems like that would be a high risk high reward situation for the company, although I wonder if the reward would be there: would an arbitral forum feel itself found by an acquittal? I was thinking the issues in the post might incentivize paying large settlements without admitting wrongdoing: I’m imagining a two variable axis, with wrongdoing and monetary payments as the axes, but I don’t know the extent to which that’s accurate.
So you would expect firms to pay more to avoid admitting wrongdoing. But wouldn’t a settlement alone draw a tribunal’s attention, especially if there is some other compelling evidence suggesting bribery?
I can see why taking a case to trial would make sense in this regime — particularly in situations where a state is already involved in or is preparing to enter arbitration, and thinks it has a very strong trial case. However, my intuition is that for most companies, the lingering specter of arbitral consequences will mostly affect the way the company negotiates with the DOJ/SEC and how hard it pushes for settlements that limit the amount of wrongdoing they are forced to admit. My suspicion is that, except for the companies that are already caught up in mega-arbitrations, the threat of criminal liability in the U.S. is still a more serious concern, and taking the FCPA case to trial still wouldn’t be worth it on a risk/reward basis.
Sam, I think your first implication rests an assumption that is not often properly teased out in the literature. Let me explain: Yes, the FCPA was created in large part to ameliorate (or at least control) the foreign policy ramifications of bribery scandals involving U.S. companies. Presumably many of the FCPA’s foreign counterparts are, to some degree, also born of foreign policy considerations (e.g., consider those states that adopt a statute as a step toward developing closer relations with the OECD and its member countries because it serves some political or economic interest of the state). But there is a significant leap from “born of foreign policy considerations” to “foreign policy instruments”; to equate the two requires an identity between antibribery prosecutors and policymakers. In the United States, most FCPA prosecutions go forward with little or no “political” interference from the policymaking elements at the top of the executive branch; in a way, FCPA prosecution decisions are driven by civil service employees far removed from any meaningful oversight by foreign policymaking authorities. Of course, this may not be true of other nations — it isn’t hard to imagine, for example, that Chinese policymaking authorities might employ their new foreign bribery law to target specific companies or countries in furtherance of particular policy goals. This isn’t to say that reliance on FCPA-type enforcement actions couldn’t skew international arbitration decisions; I just think its helpful to realize that some states (like the U.S.) may not employ antibribery statutes to accomplish “policy priorities” in the type of systematic manner that might be pertinent to skewing such decisions in the manner you suggest.
Fair point, and perhaps I overstated insofar as I suggested I thought FCPA prosecutions were directly influenced by foreign policy consideratons. Here is what the Choi & Davis paper shows: (1) the FCPA is enforced more aggressively in countries that have bilateral legal assistance agreements with the United States, and (2) sanctions tend to be lower in any individual FCPA action involving a U.S. firm. Is this conclusive evidence that foreign policy considerations are inflecting FCPA enforcement? No. But I would argue it at least suggests that the way in which the FCPA is being enforced within the United States is reflective of underlying normative ideas that exist among U.S. prosecutors about what kinds of crimes are most worth prosecuting. So in that sense, FCPA prosecutions do reflect a U.S. “sensibility” about who should pay and that, I would argue, could indeed have negative effects if imported into the arbitration system.
I had a different concern about the argument (which certainly gets repeated throughout the literature) that the dominance of the FCPA system could skew matters in favor of US interests: Why don’t we hear that argument being made so loudly in other contexts in which a sovereign can use prosecution to promote its policy goals in ways that have collateral consequences? It seems that fact or issue preclusion operates similarly in the domestic US system, but I haven’t heard similar complaints even though I’m sure arguments could be made about the effects of government policy on prosecutorial decisions.
Also, perhaps if corruption prosecutions begin having an effect in areas like arbitration, that’ll create an incentive for other countries to develop more robust systems for investigating and prosecuting corruption so as not to de facto outsource the entire system to the US government. According to your argument, could attaching greater significance to corruption investigations through their collateral effects in settings such as arbitration enhance the incentive for nations to participate in the global corruption policing system?
Eden, as to your second point, I agree it logically follows from the argument I’ve made. I’m somewhat skeptical it would actually work this way, though, because in all likelihood the absolute number of arbitrations in which corruption will actually be a factor will never be that high, and any “skewing” of the system toward U.S. policy preferences would be fairly subtle. That being the case, it’s a bit difficult to image that the prospective of FCPA-centric arbitrations would be enough by itself to impel states to increase enforcement.
With respect to the findings from Choi & Davis, I’m not convinced that these do reflect a “sensibility” about who should pay. The fact that the FCPA is enforced more aggressively in countries that have bilateral legal assistance agreements is easy to explain — if we have a MLAT agreement with a country, it is easier to get evidence from that country, and as a result, it is easier to prosecute bribery taken part in that country. The existence of an MLAT is suggestive of a better and more efficient working relationship between U.S. and foreign regulators; and given that they are human, prosecutors will often chase the easiest to prove cases.
As for the fact that sanctions tend to be lower in FCPA actions involving U.S. firms, I know this holds true as an absolute matter. (Many of the largest settlements in the past 10 years have targeted foreign firms.) But this could be a product of many things: perhaps historical U.S. enforcement has increased compliance among U.S. companies such that their conduct is less culpable on average. I don’t buy this argument as much — I know there is at least some pressure to target foreign companies to level the playing field at DOJ — but I wanted to throw it out there anyway.
I also wonder if a legitimate concern is that FCPA investigations can counter the productivity of international arbitration especially if an investigation occurs after an arbitral award is assigned. Your example of Siemens and Argentina shows the risk of arbitration parties relying on FCPA investigations to avoid paying arbitral awards in the future. While I am inclined to think that a Siemens scenario is uncommon, will parties ordered to pay arbitral awards try to stall payments as much as possible with the hope that the other party could be investigated for an FCPA violation. Sam, do you know if arbitral awards are supposed to be paid within a fairly short amount of time?