U4 Helpdesk Query

U4 helpdesk reply

The query has been sub-divided into three parts. The responses to the first two parts (estimated scale of corruption in Africa and its impact on the poor in the region) are to be found in the respective sections below. As to the third part of the query, related to specifics for African particular countries, the Helpdesk team would be happy to provide you with an answer, but would appreciate it if you could narrow down your query (perhaps focusing on a particular sub-region or listing some particular countries you are interested in), so that a more targeted response can be produced.

Content

• Part 1: Estimated Figures and Scale of Corruption in Africa
• Part 2: Estimates of the cost of corruption to Africa
  

Part 1

Corruption is a secretive transaction by definition and as such is difficult if not impossible to measure in a reliable fashion. What is more, definitions of corruption may vary, as do perceptions of what constitutes corruption in any given context. Several attempts have, however, been made at quantifying this phenomenon to increase awareness of the cost of corruption and of the degree to which it is prevalent in any given context or agency.

The most commonly employed mechanisms for creating quantitative data on corruption are surveys and econometric analysis, or estimates based thereon.

1. Surveys (estimating the scale of corruption in Africa)

Those surveys frequently used in the context of corruption are household surveys and surveys of different parts of the community, such as businessmen and public officials. They usually aim at drawing attention to the existence and scale of corruption in an institution, or indeed a state. Examples of surveys of the cost and extent of corruption in Africa or African states include:

1.1. Regional Surveys

• The World Bank Governance Indicators 1996-2002 consider the perception of corruption under the heading "control of corruption". Of those African countries included in the survey, only Botswana is positioned in the "best quartile", which indicates that a mere 25% of all 190 countries rate the same or better. By contrast, 27 countries score between 25 and 75%, and 23 countries achieve a percentile rank of below 25 (with Congo, Equatorial Guinea and Nigeria scoring below 5), which indicates that 75% of all 190 countries are perceived to be less corrupt.
  
  
• Transparency International's 2003 Corruption Perceptions Index (CPI) is a league table of nations that are ranked according to their perceived level of corruption. As indicated by its title, the CPI measures the perception, not necessarily the reality of corruption in a country. These perceptions reflect the confidence investors and investment rating agencies have in a country.
  The CPI is a composite index, that is, an index made up of a number of independent surveys. For a country to be included in the CPI, a minimum of three surveys must be available.
  [For more information on the CPI, previous indices and a background paper on methodology, please refer to web site.]
  The 2003 index places the majority of African countries in the bottom half of the 133 countries included in the index. Indeed, on a scale ranging from one (high levels of corruption) to ten (low levels of corruption), only Botswana scores more than five points (5.7), followed by Tunisia (4.9) and Namibia (4.7). Bottom of the list come Kenya (1.9), Angola and Cameroon (1.8), and Nigeria (1.4).
  
  
• The Southern African Democracy Barometer finds that "perceptions of official corruption are fairly extensive across the region [Southern Africa - Botswana, Zimbabwe, Zambia, Malawi, Lesotho, Namibia]" and concludes that even though these perceptions may not be accurate, they nevertheless constitute an important political challenge. The results of its survey show that over half of all Zimbabweans considered "all/almost all or most" officials to be corrupt (with an average score of 3.2 on a scale from 1 to 4, where 4 reflects the belief that all officials are corrupt), while almost half of the Zambians questioned ranked the corruption of their officials at 2.9 out of 4 points. Even Namibian citizens, whose view of the level of corruptness of their officials was the most positive of all countries queried, scored an average 2.2 on a scale of 4.
  
  
• Transparency International's Global Corruption Barometer 2002, a survey of attitudes, expectations and priorities on corruption in over 40 countries, covered three African countries, namely Cameroon, Nigeria and South Africa. Asked how they believed corruption affected their personal and family life, 57.4% of South Africans, 50.6% of Nigerians and 47.5 % of Cameroonians considered it to be very significant. 70.8% of Cameroonians, 68.1% of South Africans and 60.3% of Nigerians perceived corruption to affect the business environment very significantly, and 71.2% of Cameroonians, 65.8% of Nigerians and 65.4% of South Africans considered this to be true also for political life. 39.4% of Cameroonians expected corruption to increase a lot over the next three years, as did 36.1% of South Africans and 27.1% of Nigerians. Asked which institution they would choose for eliminating corruption, 31% of Cameroonians opted for the courts, and 32.1% of Nigerians and 23.8% of South Africans for the Police.
  

1.2. National Surveys

• Ghana: The Centre for Democracy and Development's 2000 Governance and Corruption Survey shows that approximately 75% of households see corruption as a serious problem in Ghana, with a majority (66%) paying 10% of their incomes (within the c 1-5 million income bracket) in bribes to public officials. 44% of firms admit to making unofficial payments to public officials, with 27% frequently or always making such payments.
• Kenya: Transparency International-Kenya's Urban Bribery Index of 2001 presents preliminary analysis of a study by TI-Kenya on the magnitude of bribery in Kenya. Based on a survey in which ordinary Kenyans report their daily encounters with corruption - who they bribe, for how much and for what, the study goes beyond perceptions of corruption to provide benchmarks of integrity based on the actual incidence of corruption. It was conducted in March & April 2001 in Nairobi, Mombasa, Kisumu, Eldoret, Nyeri and Machakos and responded to by 1164 individuals.
  According to the survey, 67% of the respondents' interaction with public institutions involved bribes or negative consequences if the bribe payment was declined. It was estimated that "the average urban Kenyan pays 16 bribes to both public and private institutions in a month. Public servants […receive] by far the most bribes, accounting for 99% of the bribery transactions, and 97% of the value". This amounts to a cost of just under Ksh 8,000 a month per respondent (compared to an average income of Ksh. 26,000 per month per respondent).
  
• Madagascar: Transparency International's Chapter in Madagascar conducted a survey in 2001 in order to identify the nature, scope and frequency of corruption on the basis of perceptions and experiences shared by the business community in the provinces of Antananarivo, Tamatave and Toliary. The results showed that among others 75% of businesses "always" experienced corruption in the public works sector and in courts, while 30% of the informal sector claimed to always experience corruption in their interaction with public services. Depending on the public agency and the economic status of the firm, bribe payments worth up to 100 million Fmg were made by the formal sector, and 5 million Fmg by the informal sector. Customs and traffic police were perceived to be very corrupt by over 90% of formal businesses, and 93% of informal sector respondents perceived the traffic police to be very corrupt as well, followed by those civil servants in charge of registrations and licenses (86%).
  
• Morocco: Transparency Maroc's Enquête Nationale d'Integrité, a national integrity survey conducted among private sector representatives in 2001, shows that in the area of procurement, 22% of respondents believe corruption to be an integral part of the process, while 37% think it is often required. Those indicating a knowledge of corruption in procurement claim that 33% of the corrupt payments made to obtain public contracts are in excess of 10% of the contract's value. The traffic police was perceived to be corrupt by 62% of the respondents, followed by customs officials at 44%, while 42% perceive corruption in politics as occurring always, and 38% as often. However, the survey also showed that firms were less likely to actually bribe (or to admit to bribery) than these figures suggest.
  
• South Africa: The Country Corruption Assessment Report 2003 intends to give a comprehensive picture of the nature and extent of corruption in South Africa. To this end, three surveys on perceptions and experiences of corruption were conducted among households, businesses and the public services and its clients. The results showed 80% of South Africans perceiving corruption to be prevalent, with 41% considering it one of the most important problems to be addressed. 62% of respondents from the private sector perceived corruption to be a serious problem, and public service clients believed between15-30% of public officials to be corrupt. Some public service managers held the view that up to 75% of their own staff were corrupt. These negative views of the scale of corruption were however not matched by the (substantially lower) amount of South Africans actually experiencing corruption.
  
• Tanzania: CIETinternational's Service Delivery Survey of 1996 points to a general perception of corruption in the four services - the police, the judiciary, revenue and lands services - included in the survey. The results from the household survey indicated that 60% of respondents thought there was "very much" corruption in their respective district, with 39% reporting corruption in revenue services, 35% in the police, 32% in the judiciary and 25% in land services. A questionnaire distributed among service workers showed that 57% of respondents thought there was "very much" corruption in public services in Tanzania.
• Uganda: Uganda Enterprise Survey 1999 was conducted by the World Bank and the Ugandan Private Sector Foundation among firms to collect data on constraints facing private enterprises in Uganda. Of the 176 firms answering the questions related to corruption, 81% reported to have been involved in bribe-paying. They reported spending an average of 7.9% of their total costs in corrupt payments, compared to 6.3% for fuel and 6.8% for interest. Of the 167 firms supplying data on both bribe payments and taxes, 70% reported higher bribe payments than corporate income taxes. Those firms not paying bribes operated in sectors with little or no contact with the public sector.
  

Even though methodology and types of respondents vary, these surveys clearly indicate that corruption is perceived to be widespread, both at the national and at the regional level.

[top]

Part 2

2. Estimates of the cost of corruption to Africa

Estimates of the cost of corruption in monetary terms - or percentages thereof - are useful to illustrate the seriousness of the case, but should be handled with care, since corruption - as mentioned before - cannot be measured with certainty. Furthermore, the cost of corruption comprises not just of the sums of money lost but also of retarded development and increased inequalities - which are far less easy to quantify.

National Economy: A report by the African Union, presented before a meeting in Addis Ababa in September 2002, estimates that corruption costs African economies in excess of 148bn dollars a year. This figure, which includes both direct and indirect costs of corruption, i.e. resources diverted by corrupt acts and resources withheld or deterred due to the existence of corruption, is thought to represent 25% of Africa's GDP and to increase the cost of goods by as much as 20%. [BBC News, 18/09/02; The Economist, 19/09/02]

Aid effectiveness: In 1996, it was estimated that up to 30bn dollars in aid for Africa ended up in foreign bank accounts, an amount equal to twice the annual GDP of Ghana, Kenya and Uganda [from Michelle Celarier, 1996: The Cost of Corruption. In: Euromoney, September 1996: 49.]
See also this article on the Asian Development Bank web site.

Revenue collection: Research findings by the African Development Bank indicate that corruption leads to a loss of approximately 50% of tax revenue, which in some instances is a greater amount than a country's total foreign debt. [Gabriel Negatu, African Development Bank (by email)]

Household expenditure: The African Development Bank estimates that lower income households spend an average 2-3% of their income on bribes, while rich households spend an average of 0.9% of their income.

These figures and results illustrate that corruption in Africa is perceived to be both widespread and costly, by diverting assets away from their intended use. While this result does probably not come as a surprise, it may be worth pointing out that although some African countries are perceived to be extremely corrupt, others, such as Botswana, Tunisia, South Africa, Mauritius and Namibia do not fare badly in international comparisons.