Biased reporting on corruption

Content image - Phnom Penh Post
Speakers talked at the press conference on launching the new annual Corruption Perception Index (CPI) last week. Heng Chivoan

Dear Editor,

Reporting on Transparency International’s (TI) new annual Corruption Perception Index (CPI) has, unfortunately, paid little attention to the details of the index and larger questions about its validity.

TI’s website does not yet carry many details about how the 2018 CPI was compiled, but the method was presumably not greatly different from 2017.

For that year, TI says that the CPI was “based on 13 sources that collect the assessment of experts and business executives on some specific corrupt behaviour in the public sector (i.e. bribery, diversion of public funds, use of public office for private gain, nepotism in the civil service and state capture)”.

At least four aspects of this deserve attention:

1. As “Perception” indicates, the index is based on the opinions of those TI chooses to consult, not on objective criteria. Yes, it would be more difficult to question farmers in Kampot or coal miners in West Virginia, but are their opinions any less relevant, and might they not be different from those of the people actually asked?

2. It is not clear how TI or its sources determine who is an “expert” on corruption. Presumably, it is not, for TI, corrupt officials or their corrupters, although who would know more about the subject?

And business executives are hardly a reliable source of cross-cultural information; an executive in one country may regard as “normal business” the same behaviour that a similar executive in another country considers to be corruption.

Furthermore, are business executives whose companies bribe regulators likely to give an honest opinion on the level of corruption around them?

3. The CPI relates only to corruption “in the public sector”. In many countries, public sector corruption is minor compared to private sector corruption, some of which was exposed after the 2008 global financial crisis and which includes things like stock market manipulation, falsification of financial documents, conspiracy to control interest rates, bribery of financial news media, manipulation or falsification of political parties’ internal procedures and much else.

All of the “corrupt behaviours” occur also in the private sector, if we substitute “private” for “public”, “corporate bureaucracy” for “civil service” and “corporate” for “state”.

It would also be helpful to know where TI draws the line between the public and private sectors. If a company or its owners give a million dollars to a private individual expecting something in return, and this individual is later elected to public office, is that “private” or “public” sector corruption?

4. The specific behaviour of “state capture” is also interesting. According to Wikipedia, “The classical definition of state capture refers to the way formal procedures (such as laws and social norms) and government bureaucracy are manipulated by private individuals and firms so as to influence state policies and laws in their favour. State capture seeks to influence the formation of laws to protect and promote influential private interests.”

The problem here is that this evades the question of when a state was captured. After the passage of time, “state capture” begins to look like “normality”. A state that was captured a century or more ago appears – at least to people like TI – as no longer captive and not corrupt.

But does anyone doubt that laws and bureaucracy in countries like the US, Britain, Australia etc. serve to “promote influential private interests” of a fairly well-defined layer? Conversely, if a government and its laws are new they are more likely to look like “state capture”.

Five years ago, responding to TI’s 2013 CPI, Dr Jason Hickel of the London School of Economics pointed out that TI’s corruption map showed North America, Western Europe and Australia and New Zealand as “clean”, while the rest of the world is “corrupt”. Nothing has changed in that regard.