• Background Information
• Implications of the TI Corruption Perceptions Index
• Methodology of the TI Corruption Perceptions Index
• The CPI in Years to Come

Background Information

Who is responsible for the production of the CPI?

This is Transparency International's product and the organisation takes full responsibility for it. The detailed compilation and development of the CPI is a joint initiative undertaken by TI and Göttingen University, Germany. Launched as an annual index in 1995, the lead operational work is undertaken by Dr. Johann Graf Lambsdorff of the Economics Faculty, Göttingen University. The methodology is reviewed in consultation with an international TI Steering Committee of academics and professionals.

For the purpose of the CPI, how is corruption defined?

Transparency International has tended to focus on corruption in the public sector and define corruption as the abuse of public office for private gain. The CPI makes no effort to reflect private sector fraud. The surveys used in compiling the CPI tend to ask questions in line with the misuse of public power for private benefits, with a focus, for example, on bribing of public officials, taking kickbacks in public procurement or embezzling public funds, etc. Many experts on corruption seek to make further distinctions within this broad definition, as they note both administrative corruption (e.g. illicit payments to tax inspectors) and political corruption (e.g. bribes to politicians and political parties).

The Implications of the TI Corruption Perceptions Index

Governments of some countries have complained about the accuracy of the CPI - do their complaints have merit?

Complaints are understandable. No government wants to see its country depicted as being corrupt. The CPI, as noted in the press release, is a "poll of polls" of perceptions held by diverse individuals. It is an index that seeks to portray perceptions of corruption in countries. Governments that do not like the results should not question the merits of the CPI, rather they should question why their countries are widely seen as being corrupt. Actions they take to address domestic corruption may influence perceptions, which in time may be reflected in the rankings in the annual TI CPI.

How reliable is the CPI as a measure of corruption?

If the perceptions consistently reproduce similar assessments it is a good indicator that they are not just referring to cultural or geographical biases but to a real-world phenomenon. The high correlation between the sources we use has actually increased in the past 3 years. The findings suggest that the CPI is a helpful contribution to the understanding of real levels of corruption. The geographical spread of the contributors also contradicts the notion that the CPI is based on the biased perception of Western business people. The following table reports the correlation coefficients of the sources.

What do the scores mean?

As indicated, the CPI does not measure corruption, it measures perceptions of corruption. And, as also noted, the CPI does not cover all of the countries in the world. It would, therefore, be inaccurate for a headline writer to look at the country that is ranked 85th and declare it to be the most corrupt country in the world. It would be equally wrong to look at the country in first place and call it the least corrupt country in the world. And, even though Denmark scores 10, TI is not willing to say corruption never surfaces in this country--just, that most people tend to believe the country is almost totally free of corruption.

Let's be clear--with a score of 10, is TI saying Denmark is free of corruption?

No. Even though Denmark scores a perfect 10.0 this does not mean that there haven't been any incidents of corruption recorded in the country in the past year. The fact that Denmark obtains such an excellent score must be seen as a good indicator of its overall standing. The fact that this is not an absolutely unanimous view is reflected in the fact that the table shows that there is a standard deviation of 0.7 in the responses relative to top ranking Denmark.

How are the exporting countries contributing to corruption?

The ranking only scores the observed behaviour of public officials and politicians, that is, the receiving end of a corrupt transaction. However, exporting industries also contribute to the development of corruption abroad. This is not taken into account in this ranking. There are also a number of industrialised countries which may be successful at keeping their house in order but whose companies are highly engaged in bribing foreign officials. Therefore, our index is not a fair assessment of the responsibility associated with corruption in international trade.

The Methodology of the TI Corruption Perceptions Index

What kinds of issues are raised in the survey questions used in the polls that form the CPI base?

Each of the sources assesses the "extent" of corruption among public officials and politicians in the countries in question but few of them explicitly distinguish between political and administrative corruption. One source that does, e.g. Gallup International, has a correlation of 0.8 between the assessment of politicians and public officials. This underscores the close relationship between these two main forms of corruption. It can therefore be argued that the CPI seeks to capture the perception of all forms of corruption affecting the public sphere. To be sure, more detailed information can be obtained by performing in-depth surveys in individual countries.

How does the CPI measure the degree of corruption?

The CPI is an assessment of the corruption level in 85 countries as perceived by business people, risk analysts, investigative journalists and the general public. The CPI does not seek to reflect perceptions of private sector fraud in countries. The scores used range from 10 (country perceived as virtually corruption-free), down to close to 0 (country perceived as almost totally corrupt).

Why is the "poll of polls" approach used?

Unbiased, hard data on the actual levels of corruption in multiple countries is not available. International surveys are therefore the most credible means available for compiling a ranking. It is the judgement of the authors of the CPI that, in an area as complex and controversial as corruption, no single source, or polling method exists as a perfect sampling frame, with large enough country coverage, and a fully convincing methodology to produce comparative assessments. This is why the CPI has adopted the approach of a composite index. It is a "poll of polls" that consists of credible surveys using different sampling frames and varying methodologies and is the most statistically robust means of measuring perceptions of corruption.

Which survey sources have been used in the 1998 CPI?

The sources used in 1998 are:

• Economist Intelligence Unit (Country Risk Service and Country Forecasts),
• Gallup International (50th Anniversary Survey),
• Institute for Management Development (World Competitiveness Yearbook),
• Political & Economic Risk Consultancy (Asian Intelligence Issue),
• Political Risk Services (International Country Risk Guide),
• World Development Report (private sector survey by the World Bank), and
• World Economic Forum & Harvard Institute for International Development (Global Competitiveness Survey)

Are old surveys used in the CPI?

This year and henceforth the CPI will be based on all available and credible surveys from the past 3 years. If a survey is not replicated and is more than 3 years old it will no longer be included in the index.

What criteria are used to determine which surveys are used?

A number of additional sources were considered for inclusion in the 1998 CPI, but they had to be rejected. In some cases the data was poorly documented, while in others there was just insufficient data available to permit a judgement on their reliability. The CPI authors are endeavouring to reduce reliance on one-off surveys, both because they tend to be slightly less reliable and because exclusion of these surveys after three years can lead to misleading changes in scores.

Why are sources from the past 3 years used as opposed to just using sources from the last year?

The 1998 CPI combines assessments from the past three years to reduce abrupt variations in scoring that might result from high-level political scandals. These scandals might even be the result of a concerted anti-corruption effort. Such events tend to affect public perceptions, but they are a poor reflection of actual levels of corruption.

In recent years the CPI has covered about 50 countries, but this year the total is 85 - how has it been possible to attain such an increase?

The first point is that, even with 85 countries, the CPI is still capturing less than one- half of the total number of sovereign countries in the world. To be sure, all major trading countries are ranked in the CPI and a very substantial number of countries outside of the OECD, which obtain significant foreign capital inflows, are also listed. However, TI sees it as a priority to strive to accurately include as many countries as possible. The increase this year arose as several CPI sources increased their country coverage and we were able to include the Economic Intelligence Unit's data for the first time. In addition, the Steering Committee agreed that we could reduce the minimum number of sources needed per country assessment from 4 (in 1996 and 1997) to 3. This change had a statistically insignificant impact on the scores. To the extent that the CPI still does not include many countries, it is because an insufficient number of surveys had been conducted.

How reliable is the ranking as a measure of a country's corruption?

In terms of perceptions of corruption the CPI is a solid measurement tool. Key to understanding the results is recognition of the Standard Deviation column. This indicates differences in the values of the sources: the greater the standard deviation, the greater the differences of perceptions of a country among the sources.

Why does this year's CPI only note one decimal place, rather than two in previous years?

We believe that by using two decimal points in a country score, such as 5.67, the CPI provides a perception of a greater level of accuracy than is realistic.

Why do some countries share the same rank?

If the CPI scores used more decimal places then differentiation could be possible. As it was decided to go to just one decimal place, numerous countries now share the same score. These countries are listed in alphabetical order with the same score and ranking on the table.

Can data from one year be compared with that from a previous year?

This is somewhat problematic. The CPI incorporates as many reliable and up-to-date sources as possible. One of the drawbacks to this approach is that year- to-year comparisons of a country's score do not only result from a changing perception of a country's performance but also from a changing sample and methodology. Some sources are not updated and must be dropped as a result, while new, reliable sources are added. With differing respondents and slightly differing methodologies a change in a country's score cannot be attributed solely to actual changes in a country's performance. Comparisons with the views collected in previous years can therefore be misleading. In order to reduce the number of misleading interpretations of the CPI scores, the official CPI table will not include the scores from the previous year. In practice, the sources continue to show a high degree of correlation. So, the impact of differing samples and methodologies on the outcome appears to be rather small.

The CPI in the years to come

Will more countries be included next year?

The goal is to increase the number, but the key is the availability of good data. In the past we have had the problem that some countries dropped out of the CPI because we could not obtain the minimum number of sources required for a valid country assessment. From 1999 we will include drop-out countries, marking them as, e.g. "not assessed since 1998." After three years these countries would be omitted. By this approach we are assured that the 1999 CPI will have at least as many countries as the new 1998 CPI.

What are your future plans for the CPI?

Transparency International recognises that the CPI is widely used as an important tool in evolving policies to curb corruption. Accordingly, the Board of TI has made it an organisational priority to make every effort to support the development of the CPI, to learn from leading experts around the globe in survey work, and to complement the CPI with increased efforts at encouraging national surveys involving TI's national chapters. In addition, TI continues to look at possible approaches for developing some form of index that might be able to provide more information on the origins of bribes and the suppliers of corrupt payments.