Skip to content


Harry Cassin
Publisher and Editor

Andy Spalding
Senior Editor

Jessica Tillipman
Senior Editor

Bill Steinman
Senior Editor

Richard L. Cassin
Editor at Large

Elizabeth K. Spahn
Editor Emeritus

Cody Worthington
Contributing Editor

Julie DiMauro
Contributing Editor

Thomas Fox
Contributing Editor

Marc Alain Bohn
Contributing Editor

Bill Waite
Contributing Editor

Shruti J. Shah
Contributing Editor

Russell A. Stamets
Contributing Editor

Richard Bistrong
Contributing Editor

Eric Carlson
Contributing Editor

Arturo del Castillo: Revisiting World Cup fouls and corruption indexes

David Bligh’s recent post for the FCPA Blog asked a provocative question: Did the Corruption Perceptions Index correlate with World Cup fouls?

To answer this question, he did a linear regression analysis with the average number of fouls committed by each team that participated in the last World Cup versus the ranking obtained by those countries in Transparency International’s 2017 Corruption Perceptions Index.

In my opinion, David Bligh did not find significant results because the data used were not entirely correct.

Taking the average number of fouls committed in each game may have an important bias. Not all teams played the same number of games and, in addition, after the first round of games, matches could be played in extra time. This situation can increase the probability of committing a foul.

Therefore, it is better to take the average foul number of only the first three matches, because these games are only played in 90 minutes.

This table shows, in the second column, the average number of fouls committed by each team in the first three matches. It can be seen that the team that made the least foul was Spain with 8.5 per game and the team that made the most fouls was South Korea with 21 fouls per game.

But the most controversial data that may not allow to reach solid conclusions is in the CPI. The CPI measures the perception of how corrupt government institutions can be in a given country. It is an index of indices that is constructed from various surveys that year after year measure the perception of experts regarding events associated with corruption-related issues.

There is a wide debate on the methodological soundness of this index, but in general it is known that it is a difficult metric to use to measure the real impact of corruption in relations with private companies and, even less, to evaluate anti-corruption policies.

However, the exercise proposed by David Bligh is interesting if one takes into account not so much the ranking assigned in the CPI as the score obtained in this metric. In the CPI, the countries are qualified in a score that goes from 100 points (condition of total absence of perceived corruption) to 0 points (condition of total perception of corruption). Understood in this way the results are interesting.

My table shows, in the third column, the score obtained in the CPI 2017 for each country that participated in the World Cup. To make the data interpretation easier, the results are shown in three colors: Green (countries that had a score of 100 to 75 points in the CPI), Yellow (countries that obtained a score of 75 to 50 points) and Red (countries that obtained a score of 50 to 0 points).

Arranged in this way, it can be observed that surprisingly, the countries that committed from 14.3 fouls onwards are countries that, according to the CPI, are perceived as having a moderate to severe level of corruption in their public institutions.

But how to explain then the location in this table of other countries that committed fewer fouls, but that clearly have more serious levels of corruption like Mexico, Egypt, Peru, Brazil and Saudi Arabia.

From my point of view, the problem is in the type of data that we are trying to correlate. The CPI tells us how corrupt the institutions of government of a country are perceived and not how much citizens respect the rule of law.

For this reason, its necessary to use another type of data that better describes the behavior of individuals or private entities with respect to their relationship with the government. Assuming that type of data reflects the kind of hypothesis that we would like to prove.

In this regard, TRACE’s Bribery Risk Matrix (BRM) of 2017 could be a better data alternative. This metric is constructed from four dimensions. Each dimension measures different aspects that affect the relationship of privates with government authorities.

My table shows, in the fourth column, the ratings obtained by each country in the TRACE – Total Risk Score. This index classifies countries according to a rating that ranges from 1 point (condition of absolute integrity) to 100 points (condition of total corruption).

To make an easier comparison, the results are also shown in three colors: Green (countries that had a score of 1 to 25 points in the TRACE – Total Risk Score), Yellow (countries that obtained a rating of 25 to 50 points) and Red (countries that obtained a rating of 50 to 100 points).

At this point, the results of the CPI and the TRACE – Total Risk Score are more similar than different. This is due to the weighting that the TRACE index itself performs among the different dimensions that make them up. But if we focus our attention only on the dimension that measures the relationship of individuals with government authorities, we can obtain more robust results.

According to TRACE, the four dimensions of its index are defined as follows:

The first domain (Opportunity) focuses on a company’s (or its agents’) direct contact with foreign public officials, where the risk increases with the frequency of interaction (how many occasions there are for a bribe solicitation), along with the expectations surrounding bribery (how common it is and how thoroughly normalized) and the leverage an official might have at his disposal (how costly it could be made for the company to refuse a bribe demand).

The second domain (Deterrence) addresses the intensity of the government’s efforts to discourage bribery by enacting the necessary laws and by enforcing those laws effectively.

The third domain (Transparency), we examine how the government indirectly facilitates the detection of bribery by opening its books to inspection and maintaining them reliably.

The fourth domain (Oversight) looks outside the government to consider whether the press and civil society are free enough and strong enough to provide a check on public corruption.

My table shows, in the fifth column, that the Opportunity dimension of TRACE correlates higher with countries that foul the most in the World Cup. The average foul number goes up to 13.5 and the number of countries that we can consider as “zero positive” is reduced.

For some, this result remains questionable because even if there is a greater correlation between the corruption and foul indices committed by the different teams in the World Cup, it does not necessarily speak of the particular conditions of the players, nor of the specific conditions of the game they had, and that led them to commit certain faults.

For others, even the fact that many players play regularly outside their countries of origin is also an element that should be taken into consideration.

All these are valid thoughts, but even so it is significant that the 16 teams (out of 32) that made the most fouls are countries that have moderate to severe problems of corruption in their interaction with governments. Mere coincidence?

Norbert Elias, the author of Quest for Excitement: Sport and Leisure in the Civilizing Process, explains that individuals reflect in games how actually behave in society. The process of civilization demands the internationalization of rules that constrain violence and misconduct. To cheat the referee, foul an opponent or goal in an irregular manner are in essence a lack of integrity. And probably those attitudes reflect education.

My conclusions: the challenge presented by David Bligh does not allow final conclusions because we still do not have enough sensitive data to correlate specific behaviors with macro social conditions. Measurement efforts are still a proxy of reality and therefore have their limits to describe a reality that will always be much more complex than our measurement tools.

Nonetheless, the basic intuition is still relevant: the societies that have more problems of corruption in their government institutions apparently have more problems for their citizens to internalize one of the most important values ​​for the civilization process: the respect of rule of law. And this can be reflected even in such trivial situations as a football championship.


Arturo del Castillo, pictured above, is an Associate Managing Director of Kroll, a division of Duff & Phelps where he directs the practice of Financial Investigation in Mexico. He has led fraud and corruption investigations for local and transnational companies, banks and financial institutions, oil and gas companies, manufacturing, entertainment, healthcare institutions and educational organizations. He has also led compliance reviews and designed anti-corruption programs for financial institutions. He can be reached here.

Share this post


1 Comment

  1. If instead of focusing on absolute number of fouls, you used a the difference in perception between whatever two countries were on the pitch as an expected differential in fouls … wouldn’t that give you a greater data set to work from? Sort of like the Pythagorean expectation of wins stat for American football (and, I don’t know, maybe that’s a stat kept in other sports as well)?

Comments are closed for this article!