• What's New

What's New

This paper highlights the importance of collecting information on corruption and delves into the critical aspects of cross-country composite corruption indices. The author states that the impact of most anti-corruption efforts is uncertain because it is often debated on the basis of weak information. This is complicated by the fact, the paper adds, that people involved in corruption seldom speak out about the vice. The author states that incentives to be corrupt are not necessarily removed by more stringent regulation.

The paper summarizes the limitations of the most famous index, Transparency International’s Corruption Perceptions Index (CPI), which is similar to the World Bank’s approach. This index measures what people think about corruption as opposed to the true facts about the actual levels of corruption. It is composed from a number of different sources that provide relevant ranking of countries. It points to the following aspects of composite indicators that it finds problematic:

• It is not known if the CPI refers to legal or illegal activities
• Lack of consensus on an operational definition of corruption make it hard to understand the criteria behind the ranking
• The ratio between the different scores has no significance and is not constant
• Individual perceptions of hidden activities can be systemically biased and not reliable
• The weaknesses are not understood by the public
• The value for statistical studies is uncertain
• The value for the global south where corruption is endemic is not certain.

The paper states that the problem of corruption is widely recognized as an obstacle to the efficiency of development aid. It presents two proposals for the reform of composite indicators as follows:

• There should be a reduction in the precision of the CPI, and the inclusion of information about countries’ cross-border achievements. The paper states that this is a possible and realistic suggestion
  
• New underlying surveys would have to be conducted controlling for the difference in the extent of trade looking into large firms’ country of origin. The paper qualifies that this suggestion is inconsistent with estimation problems and is proposed mainly for debate purposes because the methodological challenges of CPI could have been enlarged if such aspects were to be included.

The paper concludes with the final question of whether it is possible to increase the CPI’s value by creating incentives for states to improve their achievements under frameworks like the OECD anti-bribery convention.

• Full text of document
• Publisher details
• Send to a friend