Corruption: measuring the unmeasurable

Abstract

Purpose

The purpose of this paper is to argue that corruption is inherently unmeasurable. Measures of corruption produced by different sources are highly unsatisfactory. Useful measures may be possible for specific purposes in limited contexts. Reasons why highly unsatisfactory measures are in use have to do with politics and power, rather than utility of these measures.

Design/methodology/approach

The paper shows that different dimensions of corruption cannot be reduced to a single number without serious loss of information. For example the number and size of bribes cannot be reduced to a single index. The difference between perceptions of corruption and the reality also makes most popular indices useless.

Findings

The paper shows that commonly used measures of corruption are useless for their ostensible purposes. Evidence is presented that they are calculated and used for political purposes. In particular, they provide convenient excuses for the widespread failure of the structural adjustment programs of IMF and World Bank.

Originality/value

The paper's findings cast doubt on the vast amount of research which takes the corruption measures as objective and factual, and attempts to find causal factors for corruption, as well as effects of corruption on growth and other variables.

Citation

Zaman, A. and Faiz‐Ur‐Rahim (2009), "Corruption: measuring the unmeasurable", Humanomics, Vol. 25 No. 2, pp. 117-126. https://doi.org/10.1108/08288660910964184