The Corruption Perception Index (CPI) was developed by Dr Johann Graf Lambsdorff for Transparency International in 1995 (Lambsdorff, 2005). The CPI emerged during the corruption eruption as a response to the need for a measurement methodology that could evaluate perceptions of corruption more broadly while simultaneously allowing for comparative analyses to be completed across countries and time. This had the benefit of allowing for the identification of successful policy frameworks, which had a positive impact on perceptions of corruption across countries. Today, the Corruption Perception Index continues to be focused on the degree to which corruption is perceived to exist among public officials and politicians.
From its inception, the CPI has held a privileged role within discussions on the comparison of corruption across countries and time – a privilege held in part due to a variety of factors, not least: ‘first-mover advantage’ (Andersson and Heywood, 2009, p. 761), the simplicity of the model (Evaluation Department of the Norwegian Agency for Development Cooperation, 2010, pp. 31), the ability to compare through a ‘league table’ format (Andersson and Heywood, 2009, pp. 754), and indeed, the centrality of Transparency International in discussions surrounding corruption more generally (Andersson and Heywood, 2009, p. 759). However, as significant components within the methodology of the Corruption Perception Index, these advantages have also led to a situation whereby the CPI is challenged through academic enquiry. Its methodology has been challenged due to an overreliance on a small number of sources, the approach of utilising ‘expert opinions’ alone (Evaluation Department of the Norwegian Agency for Development Cooperation, 2010, p. xv), its focus on public sector corruption alone, and more broadly (and indeed related to the first criticism) a focus on the business case against corruption (Andersson and Heywood, 2009, p. 753). Further criticism addresses the Corruption Perception Index’s perceived Western bias and poor conceptual framing more generally – a problem this thesis seeks to overcome through analysing the Corruption Perception Index in the context of robust control variables (Andersson and Heywood, 2009, p. 749).
Although the Corruption Perception Index methodology has undergone several changes since its development, the measurement methodology still faces a litany of challenges primarily due to the conceptual challenge of understanding how to frame corruption, and more specifically, what to include, as outlined within Thompson and Shah (2005, p. 3). One challenge that is likely to remain is the problem of data collection on corruption (Donchev and Ujhelyi, 2014, p. 318). However, following the notable change made to the methodology in 2012, as outlined within Saisana and Saltelli (2012), the Joint Research Committee (Marcos et al., 2017, p. 21) have stated that “the CPI, besides being appealing for reasons of transparency and replicability, is also conceptually and statistically coherent and with a balanced structure (i.e. the CPI is not dominated by any of the individual sources)”. Although alternatives to the CPI measurement have been proposed and outlined by several scholars (Kaufmann et al., 1999; Donchev and Ujhelyi, 2014; Hart, 2019), the CPI maintains its validity in the context of this research. This is partially due to it being a composite index, but mainly because the impact of policy diffusion, as epitomised through early ratification of international treaties and conventions, is assumed to be captured within the component indices used to calculate the annual Corruption Perception Index score. Furthermore, the CPI serves as an important benchmark metric which is widely used to understand corruption, and for this reason, it maintains validity as a methodological tool when evaluating country performance in relation to addressing corruption (Marcos et al., 2017; Hart, 2019).
However, as stated previously, the purpose of this paper is not to develop on the lines of academic critiques of the methodology but to use the Corruption Perception Index as a key tool in analysing the international nature of corruption. Of particular interest to the core thesis of this paper is how the Corruption Perception Index incorporates changes within the policy framework of countries through early ratification of the UNCAC and its resultant impact on Corruption Perception Index scores. Given the growing internationalisation of corruption, stemming largely from the “corruption eruption” in the late 1980s and early 1990s, it is important to view both the development of the Corruption Perception Index and the drafting and adoption of the UNCAC as part of this changing policy paradigm and to critically analyse the UNCAC through the lens of policy diffusion – focusing specifically on early ratifiers of the convention (O’Sullivan, 1993; Naim, 1995; Shipan and Volden, 2008; Marsh and Sharman, 2009). As such, the CPI largely remains the index of choice when one seeks to measure the prevalence of corruption across countries over time.
Constituent Sources of the Corruption Perception Index.
- African Development Bank Country Policy and Institutional Assessment
- Bertelsmann Stiftung Sustainable Governance Indicators
- Bertelsmann Stiftung Transformation Index
- Economist Intelligence Unit Country Risk Service
- Freedom House Nations in Transit
- Global Insight Business Conditions and Risk Indicators
- IMD World Competitiveness Center World Competitiveness Yearbook Executive Opinion Survey
- Political and Economic Risk Consultancy Asian Intelligence
- The PRS Group International Country Risk Guide
- World Bank Country Policy and Institutional Assessment
- World Economic Forum Executive Opinion Survey
- World Justice Project Rule of Law Index Expert Survey
- Varieties of Democracy (V-Dem)
 A problem which it has gone some way to address through changes made in 2012 (Marcos et al., 2017).
 Who may not even be resident in the countries of which they are giving expert reports – a problem which has been somewhat addressed through the development of the global corruption barometer, which more accurately gauges public opinion.