The failure of many signatories to the OECD Anti-Bribery Convention to enforce their new laws against the bribery of foreign public officials has been widely noted, including on this blog. There is no single factor that explains this lack of enforcement across the 30 or so countries (out of 41 total signatories) that have not yet seriously begun enforcing their anti-bribery laws. However, there is a fair amount of descriptive evidence about the extent to which signatories actually do so: Transparency International (TI) has, for the last nine years, released annual reports on progress, which provide a good deal of information on this level.

I compared this dataset to two others, also created by TI: the Bribe Payers Index (BPI) Report, and the Corruption Perceptions Index (CPI). My goal is to look at the relationship between corruption at home and enforcement abroad, as well as enforcement actions and the payment of bribes.

The BPI was last released in 2011; it measures perceived frequency with which bribes are paid by companies based in 28 of the world’s largest economies, on a scale of 0-10, with ten being the least corrupt. The scores range from 6.1 (Russia) to 8.8 (the Netherlands). The CPI had been widely discussed (including on this blog), so I won’t go into much depth on it; the index ranks countries by perceived domestic corruption, again on a 0-10 scale, with 10 being the least corrupt. I am also using the 2011 version of the CPI.

It is worth noting that the BPI and CPI “correlate strongly with each other,” with an r value of 0.904 for the countries in my sample. I compared the BPI and CPI scores to the 2013 TI OECD progress report (the first version that provides a score for each country), which ranks countries based on their enforcement activity.  (The OECD progress report gives each country a numerical score, which, adjusted for the amount of the country’s exports, places it into one of four categories, from “Little or No Enforcement” to “Active Enforcement.”) There are 19 countries that have data for all of these variables (CPI, BPI and enforcement class). The numbers are provided here.

There is only a weak (although positive) relationship between the level of OECD enforcement and BPI (r value 0.291) and CPI values (r value 0.386). In other words, countries that are perceived as less corrupt tend to enforce foreign anti-bribery laws a little bit more–but not a lot more. This trend is not consistent, though: while only relatively non-corrupt countries are strong enforcers, a number of non-corrupt countries are also non-enforcers. These countries (including the Netherlands, Japan, and Belgium) are probably the low-hanging fruit for those who are attempting to improve the coverage of the OECD Convention. Additionally, Italy, which has high domestic corruption, but strong foreign enforcement and low foreign bribe paying, is an interesting outlier. The data are subject to errors based on small sample sizes and time discontinuities between the different sets, but I believe it is valuable to see the relationships between these different factors. Charts of the relationship (with countries ranked by enforcement efforts on a sliding scale and in groups) are here.