ADVOCATES FOR INTERNATIONAL DEVELOPMENT
AT A GLANCE GUIDE TO
Prepared by lawyers from
TABLE OF CONTENTS
1. TRANSPARENCY INTERNATIONAL
• Why is Transparency International concerned with corruptions?
2. TRANSPARENCY INTERNATIONAL CORRUPTION PERCEPTIONS INDEX
• Purpose and aims
• History of the Corruption Perceptions Index
• Perceptions not empirical data
• Conditions for inclusion of a source
• Year to Year Comparisons
• Survey questions
• Who was sampled?
• Is the CPI a reliable measure of a country’s perceived level of corruption?
3. Interpreting the CPI
• Is the country with the lowest score the world’s most corrupt country?
• Example from TI literature regarding the interpretation of the CPI: What is implied by Somalia’s ranking in
the CPI 2007?
4. Positive and Negative Aspects
• Positive Aspects
5. Other Indices
• Transparency International Bribe Payers Index
• Transparency International Global Corruption Barometer
• Transparency International Global Corruption Report
• Gallup Worldwide Corruption Index
Transparency International (TI)1 is a non‐governmental organisation funded by various governmental agencies, international
foundations and corporations. It was founded in 1993 and has more than 90 chapters around the world. TI claims on its
website that it
raises awareness and diminishes apathy and tolerance of corruption, and devises and implements practical actions
to address it. Its global network of chapters and contacts also use advocacy campaigns to lobby governments to
implement anti‐corruption reforms.2
It claims to be politically non‐partisan and to not undertake investigations of alleged corruption or expose individual cases,
but at times may work in coalition with organisations that do.
Why is Transparency International concerned with corruption?
TI believes that “corruption hurts everyone, and it harms the poor the most.” It cites the examples of "the unsuspecting
sick person who buys useless counterfeit drugs, putting their health in grave danger", "a small shop owner whose weekly
bribe to the local inspector cuts severely into his modest earnings", as well as "the prosperous multinational corporation
that secured a contract by buying an unfair advantage in a competitive market through illegal kickbacks to corrupt
government officials, at the expense of the honest companies who didn’t" and "post‐disaster donations provided by
compassionate people, directly or through their governments, that never reach the victims, callously diverted instead into
the bank accounts of criminals". Many more examples of the effect of corruption may be found on its website, but its
conclusion is "corruption hurts everyone".
Transparency International Corruption Perceptions Index3
Purpose and aims
The Transparency International Corruption Perceptions Index (CPI) is prepared by TI, which commissions Johann Graf
Lambsdorff, a university professor based in Passau, Germany, to supervise the methodology and statistical calculations. The
data used to determine the index are taken from already existing polls to rank countries in terms of the degree to which
corruption is perceived to exist among public officials and politicians. It is a composite index, a poll of polls, drawing on
corruption‐related data from expert and business surveys carried out by a variety of institutions. The CPI reflects the views
of individuals polled from around the world, including those of experts who are living in the countries evaluated. The
sample is discussed more fully below in "Who was sampled?".
The CPI focuses on corruption in the public sector and defines corruption as the abuse of public office for private gain. The
surveys used in the compilation of the index are commissioned to ask questions that relate to the misuse of public power
for private benefit, for example bribery of public officials, kickbacks in public procurement, embezzlement of public funds,
or questions that probe the strength of anti‐corruption policies, thereby encompassing both administrative and political
History of the Corruption Perceptions Index
The CPI is one of the best known of TI’s tools; it was first released in 1995 and TI claims that it has been responsible for
putting it and the issue of corruption on the international policy agenda. The first CPI was published in 1995 and included
41 countries; the CPI published in September 2007 ranks 180 countries including 52 African countries. New countries are
included in the index when there are sufficient data on those countries to include them (that is when there are sufficient
independent polls to enable analysis). The absence of a country relates only to the absence of information and has no
bearing on the perception of corruption in that country.
Further information on TI can be found at www.transparency.org and http://www.icgg.org – International Center for Corruption
Resources (a joint venture between TI and Universität Passau)
Perceptions not empirical data
As alluded to above in "Purpose and Aims", hard empirical data are not used to compile the CPI; rather "the experience and
perceptions of those who are directly confronted with the realities of corruption in a certain country" are drawn upon.
Empirical data are not used to assess the overall levels of corruption in different countries, as data on comparisons of the
amount of bribes or the number of court cases brought in a country may not necessarily reflect actual levels of corruption.
For instance, in the latter case it could be argued that the data merely highlights the quality of the prosecutors and the
courts and/or the media in exposing corruption in such a country. Transparency International claims that international
surveys on perceptions are the most credible means of compiling a ranking of nations in terms of corruption.
The CPI has guidelines which have been set up to govern the decision‐making process regarding the selection of its sources.
This section contains a summary of the methodology of the CPI; for more detail please see “The Methodology of the TI
Corruption Perceptions Index 2007” published by TI (the Methodology).4
The sources must meet the criteria set out below and the final decision of whether a country should be included is met with
the help of the Transparency International Index Steering Committee. A full list of survey sources, details on questions
asked, and number of respondents for the CPI 2007 are available in the Methodology.
Conditions for inclusion of a source
The CPI 2007 draws on 14 different polls and surveys from 12 independent institutions. Where the sources are surveys, the
data for the previous two years is included in the CPI (if available) to create a smoothing effect. For a country to be included
in the CPI the following conditions must be satisfied with respect to the data:
• it must be well documented and sufficient to permit a judgment on its reliability;
• there must be at least three surveys available from the approved sources for that country;
• the source must provide a ranking of nations. This condition is not met if a source has varying methodologies applying
to the countries it surveys, as this would result in an inability to compare country with country; and
• the source must measure the overall extent of corruption. Therefore its scope must not be mixed with other issues
such as political instability or nationalism, or if changes are measured instead of the extent of corruption.
Year to Year Comparisons
The index provides a snapshot of the views of business people and country analysts with less of a focus on trends. Where
trends in the index can be identified and these trends are not due to technical factors it is possible to compare the CPI 2006
and CPI 2007 results. Changes which are greater than 0.3 are unlikely to have arisen from technical factors but from actual
changes in perceptions.
Points to note when comparing previous years' results to the CPI 2007:
• comparisons to the results from previous years should be based on a country's score, not its rank, as new entries in the
index may change the ranking of countries;
• scores may change due to a changed perception of corruption in a country, but might also change due to a change in
the sample and methodology, for example several extra sources were included in the CPI 2007; and
• the Global Corruption Report 20065 looks at the trends which are present between 1995 and 2006.
Below is an incomplete list highlighting a few examples of questions which have been posed by the sources, from which the
CPI 2007 was compiled.6
www.transparency.org/policy_research/surveys_indices/cpi/2007/methodology or www.ICGG.org.
This can be found at: http://www.transparency.org/content/download/19093/263155
These are not complete and further examples and more detail may be found in the Methodology.
• The Economist Intelligence Unit asks a panel of experts to assess the incidence of corruption and defines corruption as
the misuse of public office for personal (or party political) financial gain. Integers between 0 (denoting a “very low”
incidence of corruption) and 4 (denoting a “very high” incidence) are used.
• Freedom House Nations in Transit asks its panel of experts to assess, inter alia, the implementation of anticorruption
initiatives; the government’s freedom from excessive bureaucratic regulations and other controls that increase
opportunities for corruption; public perceptions of corruption; laws on financial disclosure and conflict of interest.
• Political and Economic Risk Consultancy, Hong Kong asks expatriate business people to rate on a scale of 0 to 10 how
bad they considered the problem of corruption to be in the country in which they are working as well as in their home
• World Economic Forum (WEF) asks: “In your industry, how commonly would you estimate that firms make
undocumented extra payments or bribes connected with [different industries]”. The respondents rate on a scale of 1
(common) to 7 (never occur) from which a simple average is taken.
• Global Insight provides an assessment of the likelihood of encountering corrupt officials. Corruption can range from
petty bureaucratic corruption (such as the paying of bribes to low‐level officials) right through to grand political
corruption (such as the paying of large kickbacks in return for the awarding of contracts). Scores take the following
values: 1; 1.5; 2; 2.5; 3; 3.5; 4; 4.5; 5. Where 1 means corruption is virtually unknown and 5 means corruption is
pervasive and reaches the highest levels of government.
None of the data in the CPI 2007 emphasise one form of corruption at the expense of other forms. All the sources aim at
the same broad phenomenon. In addition, the sources do not distinguish between administrative and political corruption.
Who was sampled?
The samples used by the sources differ hugely, but they can be broken down into two groups, and TI notes that the data
from these two groups actually do correlate well:
• Non‐residents (expats) – an established network of local correspondents guide the qualitative assessments by
coordination and discussion with staff members at their headquarters. The advantage of this group is that they do not
have a "home country bias" and therefore do not rate the country on local standards which differ from country to
• Resident nationals and ex‐pats from multinational firms – there may be some "home country bias" but the advantage
in these samples is that they are not susceptible to introducing an undue dominance of "western business people's"
viewpoint (where a westerner may lack understanding of local culture).
Please see the Methodology for details of how the sources are standardised and the index is complied.
Is the CPI a reliable measure of a country's perceived level of corruption?
TI states that the CPI has been tested and used widely by both scholars and analysts. However it should be noted here that
the reliability of the CPI differs across countries. Countries with a high number of sources and small differences in the
evaluations provided by the sources (indicated by a narrow confidence range) convey greater reliability in terms of their
score and ranking; the converse is also the case.
Interpreting the CPI
The CPI should be interpreted as a ranking of countries with scores ranging from 0 (highly corrupt) to 10 (highly clean).
While ranking countries enables TI to build an index, a country’s score is a much more important indication of the perceived
level of corruption in a country. A country's rank can change simply because new countries enter the index or others drop
Is the country with the lowest score the world's most corrupt country?
The score a country receives in the CPI does not reflect how corrupt a country is, but how corrupt a country is perceived to
be. As all the countries in the world are not included in the index it is incorrect even to say that the country ranked at the
bottom of the list is the country where corruption is perceived to be the greatest. Obviously, the CPI provides no
information about countries which are not included.
Example from TI literature regarding the interpretation of the CPI: What is implied by
Somalia’s ranking in the CPI 2007?
"Corruption in Somalia has been perceived to be the highest in the CPI 2007. This does not, however, indicate that Somalia
is the ‘most corrupt country’ or that Somalians are the ‘most corrupt people’. While corruption is one of the most
formidable challenges to good governance, development and poverty reduction in Somalia, the vast majority of the people
are only victims of corruption. Corruption by a limited number of powerful individuals, and failure of leaders and
institutions to control or prevent corruption, does not imply that a country or its people are most corrupt."
Positive and Negative Aspects
Below the positive and negative aspects of the CPI and its methodology are discussed. The sources have been found
through internet and journal searches and this section does not purport to be exhaustive.
1) Use of perception as opposed to fact
The CPI has been criticised for representing the perception of a selected few experts and of ignoring the perception of the
wider population. In particular, some Bangladeshi economists have questioned the methodology of the CPI pointing out
that although Bangladesh was placed as the seventh most corrupt country in the CPI 2007, 90 per cent of the country had
no scope to be involved in corruption.7 Another economist Hossein Zillur Rahman, the executive chairman of the Power and
Participation Research Centre, said the report was a "mere perception" and the method followed in preparing the index
could not measure institutional corruption. "It is more important to look into the causes of corruption, to look at whole
perspective," he said.8
As mentioned above, it is apparent that reliable empirical data are not as easy to collect as may at first be imagined. It
seems that TI does, however, recognise this and have noted that this is the rationale behind their other indices, such as the
Bribe Payers Index and the Global Corruption Barometer. These are discussed below under the section entitled "Other
It has been argued in some quarters that the results of a previous year's index could impact the perceptions of the
respondents in surveys in the current year. This hypothesis was tested by TI in 2006 using a survey question posed to
business leaders around the world. Based on more than 9000 responses, TI concluded that knowledge of the CPI does not
induce business experts to ‘go with the herd’. TI concluded that knowledge of the CPI may motivate respondents to
determine their own views. This is, for them, a strong indication that there is no circularity in the present approach.9
It has also been argued that the CPI measures reputation, and reputations take a long time to change.10
4) The CPI judges poor nations when rich nations are the ones who fund the corruption
TI recognises that this is the case and have produced other indices to attempt to address this issue. Please see
"Transparency International Bribe Payers Index" below.
Abul Barakat, economics professor of Dhaka University, http://www.hindu.com/thehindu/holnus/003200709270921.htm,
Further information on the questions asked by TI in order to establish that there was a low risk of circularity can be found at:
5) Government aid may be based upon the ranking in the CPI
There is some concern that some governments have sought to use corruption scores to determine which countries receive
aid, and which do not. TI states in its Frequently Asked Questions that it does not encourage the CPI to be used in this way.
Countries that are perceived as very corrupt can not be written off – it is particularly they who need help to emerge from
the corruption‐poverty spiral.11
1) Although it takes time for reputation and perception of corruption to change in relation to a country, over time trends
in the CPI can highlight positive improvements in some countries.
2) TI does not carry out any investigations into corruption itself, but corruption investigations are and have been launched
on the information provided by TI.
3) It has been noted that the TI indices, including the CPI, are useful tools for Anti‐money Laundering Officers in companies
and organisations, who can use the report as a guide when judging issues in relation to anti‐money laundering. There
are also examples of states using the results of the indices to inform themselves in reports.
TI has commissioned further indices to complement the information provided by the CPI. The Transparency International
Bribe Payers Index (BPI) assesses the supply side of corruption and ranks corruption by source country and industry sector.
The Transparency International Global Corruption Barometer (GCB) is a public opinion survey that assesses the general
public’s perception and experience of corruption in more than 60 countries around the world.
Transparency International Bribe Payers Index12
The BPI is a ranking of leading exporting countries and their propensity to bribe abroad, providing an indication of the
‘supply side’ of corruption. TI indicates in its CPI 2007 Press Pack13 that
corruption by high‐level public officials in poor countries has an international dimension that implicates the CPI’s
top scorers. Bribe money often stems from multinationals based in the world’s richest countries. It can no longer be
acceptable for these companies to regard bribery in export markets as a legitimate business strategy.
The BPI thus attempts to show the opposite side of the coin to that of the CPI, exploring where the money for bribes and
corruption perceived by the CPI are funded. The BPI ranks countries as opposed to companies, as the
supply side of corruption in international business transactions implies a shared responsibility between companies
operating abroad and their home governments. This survey indicates the success and failure of governments to
control corruption abroad by companies headquartered in their national borders. It also indicates the success and
failure of companies to ensure their employees comply with the highest standards of business practice. This is the
particular aspect the BPI is examining.14
As TI states in its introduction to the CPI 2007, "The bribe money that buys a champagne lifestyle for corrupt officials in the
poorest nations often originates in multinational companies based in the world’s richest countries – the CPI’s top scorers."
Transparency International Global Corruption Barometer15
TI states that the GCB 2007:
Further information can be found at http://www.transparency.org/policy_research/surveys_indices/cpi/2007/faq#using2
The most recent Bribe Payers' Index (October 2006), please see: http://www.transparency.org/policy_research/surveys_indices/bpi
Found within the press pack at http://www.transparency.org/content/download/23972/358236.
For the Questions and Answers BPI 2006 published by TI in relation to the BPI 2006 please see:
asks people about their opinions regarding which public sectors are the most corrupt, and their opinion on how
future levels of corruption will evolve in the near future, as well as how their government is doing in the fight
against corruption. The GCB also explores people’s experiences with bribery, presenting information on how
frequently citizens are asked to pay bribes when they come in contact with different public service providers.16
It differs from the CPI in that it focuses on the experiences of the general public and the corruption they encounter, as
opposed to the corruption in the public sector, upon which the CPI focuses. Despite these differences, the statistics indicate
that there is a strong correlation between the two indices.
Transparency International Global Corruption Report17
The Transparency International Global Corruption Report (GCR) presents an in depth assessment of the state of corruption
around the world. It brings together contributions from experts and activists to explore corruption issues in a specific
sector, to present a review of corruption trends and issues in a series of country reviews, and to showcase the latest
research findings with regard to corruption and anti‐corruption reform. The most recent GCR was published in May 2007
and focuses on corruption in judicial systems.
Gallup Worldwide Corruption Index18
The Gallup Worldwide Corruption Index (GCI) measures the general public's perception of corruption within their own
country. The two questions asked by Gallup are (a) Is corruption widespread through the government in your country? and
(b) Is corruption widespread within businesses located in your country?
Mobolaji E. Aluko has suggested the comparison might be more interesting if a further index were proposed which
measured what people within a country considered corrupt; a "Citizen Corruption Index".19 He proposes this from an
insight he came across during his research (but referenced only as 'a discussant on a website'):
People in developed countries are more likely to make a bigger deal out of corruption, so if you look at Gallup's
chart, you'd think that Italy is more corrupt than Nigeria, which is patently ridiculous. Italy is outrageously corrupt
for a sophisticated, developed nation (Transparency International rates it a 4.5/10 in transparency), but is nowhere
near the level of Nigeria (2.2).) What is reasonable to conclude is that Italians are more bothered by their level of
corruption than Nigerians are, and perhaps they should be, because their country is far more corrupt than other
countries with similar economic and social development.
The sources for this guide are cited in the footnotes. Where sources have not been cited the information has been obtained
from Transparency International's websites and does not necessarily reflect a non‐biased view.
"Nigeria ‐ 10 Years of Transparency International Monitoring Perception of Nigeria 's Corruption", Mobolaji E. Aluko, PhD, December 9,