Today’s post is from Peter Leasure J.D. (current Ph.D. candidate, University of South Carolina, Department of Criminology and Criminal Justice).
*****
When asked about all sorts of automotive and household repairs, a dear friend used to reply, “I know just enough to get into trouble.”
Such can be the case with statistics and Foreign Corrupt Practices Act (FCPA) research. While statistical research within the FCPA is sparse[1], some efforts have been made. However, some of these efforts fell short and the entire legal field, especially those publishing legal journals, could greatly benefit from a better understanding of statistics.
My new article forthcoming in the Journal of Financial Crime titled “Embracing Fragility in Our Data: A Cautionary Example from Research on the FCPA and Voluntary Disclosure” (click here to download) provides a clear example of the need for a greater understanding of statistics in FCPA research.
“Embracing Fragility in Our Data” critiques the methodology of a previous study [2] which looked at whether there are benefits to voluntary disclosure of FCPA misconduct. In the study being critiqued, results showed that voluntarily disclosing companies faced nearly two and a half times higher total fines than companies that did not voluntarily disclose. A simple Google Scholar search yields twenty eight manuscripts that have relied upon and cited the findings noted by that previous study. It is no doubt likely that several other academics and practitioners have also relied upon and communicated those findings.
The problem is that appropriate statistical analysis of the data severely weakens the previous study’s findings. In fact, “Embracing Fragility in Our Data” shows that the findings of the previous study are rendered completely unreliable once responsible statistical analysis is applied. Ignored statistical assumptions, sample selection bias, and other methodological issues are pointed to as being responsible for the erroneous findings.
So what do we know about the benefits to voluntary disclosure from the previous study being critiqued? The responsible answer is very little, if anything. However, recent work has provided very interesting developments on this topic.
My second new article, also forthcoming in the Journal of Financial Crime, delivers a more reliable answer to the question of whether there is a benefit to voluntary disclosure by adding several key variables, increasing observations, and utilizing conservative estimates. The article entitled “To Disclose or Not to Disclose? Revisiting the FCPA with Improved Statistical Analysis Provides a Better Answer to this Illusive Question” (click here to download) is a must read for both practitioners and academics alike as it shows that companies do actually receive a benefit to voluntarily disclosing FCPA infractions.
Though this post is certainly a critique, I applaud the previous study’s author for attempting to use statistical analysis in FCPA research. It cannot be stressed enough that even though there are statistical errors in the previous study, I do not argue for the abandonment of statistical research in FCPA research or in the legal field. On the contrary, this post shows that a better understanding of statistics is indeed vital for FCPA research and those who publish and profile FCPA research on public websites.
____________________________
[1] In fact, the Organization for Economic Cooperation and Development (OECD) has noted this sparseness. In a Phase 2 Report, the OECD stated that “there are no clear, documented, formal processes between agencies to underpin the vital exchange of information and reporting of suspected violations, and a corresponding absence of statistics. This results in a lack of transparency and of data, which, if captured, could serve useful analytical purposes in reviewing the workings of [FCPA type legislation]”
[2] Hinchey, B. (2011). Punishing the penitent: Disproportionate fines in recent FCPA enforcements and suggested improvements. 40 Public Contract Law Journal 393 (2011).
