Academics like to measure things.
However, just because something can be measured, doesn’t necessarily mean that it should be measured or that the measurement has any meaningful significance.
There are lots of things in the Foreign Corrupt Practices Act space that can and should be measured.
However, this recent article titled “In-Group Favoritism as Legal Strategy: Evidence from FCPA Settlements” once again demonstrates the silliness of measuring certain things. (See here for a prior post).
The article abstract states:
“Anti-corruption laws aim to bolster public integrity by punishing attempts to illegitimately curry favor with government decision-makers. These laws, however, can generate integrity risks of their own. This Article examines one such risk: that firms subject to scrutiny under the Foreign Corrupt Practices Act (FCPA) may attempt to influence prosecutors by exploiting shared political leanings or related socio-cultural ties. Drawing on social psychology, we theorize that FCPA defendants retain defense attorneys that are ideologically aligned with Justice Department officials. This behavior is consistent with a strategy of marshaling affective polarization—i.e., the psychological tendency for individuals to view more favorably those that share their political beliefs—to defendants’ advantage. Alternatively, it may reflect defendants’ related belief that they benefit from retaining counsel that share social or cultural ties with prosecutors, where these ties happen to aligned with political orientation. Under either explanation, the strategy of hiring aligned counsel may be particularly auspicious in FCPA matters, in which prosecutors engage in subjective, trust-based assessments of defendants’ self-investigatory efforts, typically with minimal judicial oversight.
We test this theory by matching attorneys listed on court filings for all FCPA matters over eighteen years with a database of individuals’ political views based on their patterns of political donations. This analysis reveals that defendants tend to hire more liberal attorneys during Democratic administrations and more conservative attorneys during Republican presidencies. They also are more likely to hire liberal attorneys when Justice Department prosecutors lean left and conservative ones when prosecutors lean right.
These findings are consistent with our theory that FCPA defendants select counsel based on perceived benefits of their alignment with government officials. That possibility is noteworthy given the importance of shielding anti-corruption enforcement from even the perception of improper influence. In light of these findings, we offer policy prescriptions aimed at increasing transparency and judicial oversight of FCPA matters to mitigate integrity risks.”
In terms of methodology, the article states:
“To assess this theory, we assemble and analyze an original dataset of ideological preference estimates for DOJ Fraud Section leaders, U.S. Attorneys, and defense counsel involved in 138 FCPA prosecutions between 2001 and mid-2019 [according to the authors, the dataset is comprised of 57 deferred-prosecution agreements, 39 non-prosecution agreements, 30 plea agreements, and 13 declinations].
Next, we identified the lead defense attorneys and prosecuting attorneys with supervisory authority from documents filed with the courts (mainly deferred-prosecution agreements, nonprosecution agreements, and the like), letters from Justice Department attorneys to the defendant outlining the terms of plea agreements and other measures, and press releases. […] For each enforcement action, we note the following information from the face of these documents: (1) the highest-ranking prosecutor at Justice Department’s Fraud Section (typically the head or deputy chief of the Section); (2) the U.S. Attorney or highest-ranking prosecutor at the relevant U.S. Attorney’s Office, if applicable; and (3) the first-listed defense counsel for each law firm included on the documents. Examining a subset of these documents reveals that the first-listed defense attorney is always the most senior. These lead counsel are likely to be the final negotiators during discussions with prosecutors and ultimate point-of-contact with their clients.
Finally, we leveraged the Database on Ideology, Money in Politics, and Elections (DIME) to determine the political preferences of each identified individual. DIME collects 130 million contributions by 14.7 million people over a 36-year span. It then uses an algorithm to place donors and recipients along a left-right scale such that the distance between donors with similar patterns of political giving, and recipients with similar donor pools, is minimized. Negative values on this scale correspond to more liberal donors, whereas positive values refer to more conservative ones. The pool of CF scores is normalized, with a mean score for donors of zero and a standard deviation of 1. For each individual that we identified from the Garrett & Ashley Registry, we searched DIME to identify that individual’s CF score. To ensure that we identified the correct donor—and not a donor with the same name—we cross-checked the profession, employer, and location fields in DIME with the correct individual’s biographical information available via professional websites and on the face of Registry documents. This method was generally successful. For 127 of the 138 enforcement actions in our dataset, we located the CF score for a relevant government attorney: either the senior prosecutor at Main Justice or, if applicable, the relevant U.S. Attorney (e.g., Sandra Moser and Richard Donoghue in our Legg Mason example). Further, we identified a first listed defense attorney for 119 enforcement actions.”
What to make of this article?
For starters, there is lots in the article that is rather common knowledge or has been extensively written about (indeed the article contains numerous citations to my prior work). Consider the following excerpt.
“With little judicial involvement, prosecutors are the key decision-makers in FCPA matters. Resolutions, usually in the form of deferred-prosecution or non-prosecution agreements, turn on prosecutors’ assessments of defendants’ difficult-to-observe conduct: namely, the quality of their internal investigations and the rigor of their forward-looking compliance efforts. Given the subjective, trust-based nature of these assessments, defendants may find it particularly beneficial for the prosecutor to have a favorable impression of defense counsel.”
I certainly agree with most of the above assertions.
However, I find several flaws in the methodology of the article and its conclusions.
First, while the authors purport to be analyzing FCPA enforcement actions, that does not appear to be the case in several instances. The article states:
“The Registry contains information concerning 138 FCPA enforcement actions over our study period. Defendants in these actions include several of the largest American and foreign corporations, such as Walmart, Chevron, and Siemens, as well as their subsidiaries, e.g., DaimlerChrysler China and Shell Nigeria. It also includes more obscure entities, such as U.S. Imagina and Latin Node.”
The problem with the above is that Chevron and U.S. Imagina have never resolved an FCPA enforcement action. Because the authors do not list the “138 FCPA enforcement actions” it is difficult to assess whether these are the only two non-FCPA enforcement actions counted as FCPA enforcement actions or if there are others. All in all though, probably a minor point.
Second, and much more importantly, while certain companies under FCPA scrutiny do engage counsel specifically for an FCPA matter, other companies under FCPA scrutiny utilize FCPA counsel from a law firm that services the company in many other areas of law. In such a situation, the company did not actively seek out FCPA counsel, but rather the company went with the same law firm that it uses for many other legal matters. Thus, drawing a linkage in such situations between the political donations of FCPA counsel and DOJ attorneys is tenuous at best.
Third, and also very important, the average length of time a company is under FCPA scrutiny is approximately four years (see here) with several instances of FCPA scrutiny lasting more than six years. If the scrutiny arose because of a voluntary disclosure, it is likely that FCPA counsel was engaged anywhere from one month to perhaps one year in advance of the voluntary disclosure. In other words, it is a near certainty (given the frequent turnover in various DOJ positions) that when FCPA counsel was engaged the officials at the DOJ are going to be long gone by the time the resolution occurs. On the flip side, the DOJ officials whose names appear on the resolution documents were most likely not the same DOJ officials at the time FCPA counsel was engaged.
Consider, Walmart’s FCPA scrutiny. The company disclosed its scrutiny in December 2011 and FCPA counsel was reportedly engaged by Spring 2011 at the latest. Walmart finally resolved its FCPA scrutiny in June 2019. In other words, Walmart was under FCPA scrutiny for approximately 8 years and several DOJ officials whose names appear on the Walmart resolution documents were not even at the DOJ when Walmart engaged FCPA counsel – or if they were at the DOJ at this time – the individuals were in significantly different positions.
Fourth, when an FCPA enforcement action is resolved through a non-prosecution agreement or declination with disgorgement (as a meaningful percentage of corporate enforcement actions are), there is often little involvement (or at least apparent involvement from the resolution documents) of a local U.S. attorneys office. However, when an actual court filing is made – as in the case of deferred prosecution agreements, criminal informations, plea agreements, etc. the U.S. attorneys office is often involved (at least from a filing perspective), but may not play a meaningful substantive role in the enforcement action or the years in which the company was under FCPA scrutiny. In such a case, trying to make a linkage between the political affiliation of a U.S. Attorney and FCPA counsel is tenuous.