Measuring corruption (Extent and Impact)

Corruption has become not only an economic and social problem, but a political and religious problem (Acha, 1993). This makes corruption a function of countries’ policy and institutional environment (Dike, 2003). Studies in corruption have started to draw identifiable indicators that are measurable (Gambetta, 2000; Berg, 2001). There is a need to measure corruption in countries based on the socio-economic and political factors. Also, measurement helps to reveal the extent of corruption so that it could be checked for the improvement. It is hoped that this research will address the measurement of the extent of corruption in education and health, of some income households in Apata, Ibadan.

Daniel Kaufman, the World Bank Institute’s global governance director, in Odious Debts (CIOB, 29 September 2004, p.1) argues that:

“Until very recently, it was virtually impossible to venture an estimate of the extent of annual corrupt transactions. Only few years ago, corruption was regarded as impossible to measure. But thanks to the explosion in measurement approaches and actual data in this field, it is now possible to estimate rough orders of magnitude.”

For instance, Rijikeghem and Weder (1997) investigated the extent to which the level of public sector salaries is linked to the amount of corruption in 28 developing countries. Similarly, the Economic Intelligence Unit (former Business International Corporation [BI]) conducted surveys of levels of corruption in different countries. They rated these countries based on data collected from a network of correspondents and analysts around the world. Its first publication covers a period 1981–83. Berg (2001) and Jain (2001) point out that Mauro (1995) may have been the first to use BI data for empirical analysis because he found a stronger impact of corruption perceptions on investment than growth. This may be because perceived corruption level is a poor indicator of true corruption.

Aina (1982, p.6) based his methodology on judgments derived from study, observations, laws and regulations, newspapers and experience:

“Hence there may be some persons who from their own experiences, would disagree with them. I would venture to add, nevertheless, that a great many people who have had working experience or intensive exposure to Nigerian bureaucracy would support my observations.”

The measurement of the extent of corruption until now is extensively theoretical and full of anecdotes, because according to Miller et al. (2001, p. 26):

“The trouble with over reliance on the anecdotes and perceptions’ approach is not just that it is insufficiently quantitative, and not just that it exaggerates the extent and importance of low-level corruption, but it fails to discriminate sufficiently.”

But Miller et al (2001), argue that official corruption statistics are less reliable because no party to the transaction has much incentive to report it. Since corruption is carried out clandestinely, there is still more information about corruption that is hidden and cannot be known. What the public read about the extent of corruption, hear and experience helps them in drawing individual judgements

However there is still difficulty in knowing the precise extent of corruption (Corey, 2004). The extent of corruption cannot be precisely measured because a large proportion of it is never detected, nor reported (Monteiro, 1966; Caiden et al. 2001; Corey, 2004).

Even if the extent of corruption were measured exactly, there would still be a problem of measuring an act which is clandestine and elusive (Miller, 2001). According to David Silverman (2002), many crimes such as embezzlement are not reported because they are not committed openly but behind closed doors. It is an act that is taken seriously only when the culprit is caught (Agedah, 1993). Hence, everyone is in the dark about the real extent of corruption (Lambsdorff, 1999). It is not only the definition of corruption that is problematic: existing literatures focus on cross-country analysis and data on corruption derived from perception indices. For instance, even bribes paid by various households to access education and health care and other social services can never be correctly measured because so far the extent of corruption has been measured so far from mainly the payer’s perspective. The researcher in the survey in Apata, Ibadan, Nigeria, was able to conduct a survey at a household level for the first time, to assess the extent of corruption on the livelihood of the income groups in education and health care services.

Therefore studying corruption has some analytical problems. Johnston and Wood (1985) identified viable collection of data as a problem, the loaded colourful rhetoric that accompanies it, the emotive nature and the fact that often there is no immediate victim. Similarly, there is conflicting empirical evidence about how the extent of corruption is measured. Caiden et al. (2001) argue that corruption cannot be measured because of its often indeterminate and conspiratorial nature. Berg (2001) and Jain (2001) posit that empirical study on corruption requires a measure for corruption as well as measures for a larger number of social, economic and political variables.

The lack of research on the links between corruption and the cost of misdirected public policies, and lack of a precise management tool, constitute a problem on how the extent of corruption is measured (Jain, 2001, p. 74). Similarly, another problem of measuring corruption according to Morse (2004, p. 133) is that:

“Those benefiting from corruption are unlikely to say so and even more unlikely to say how much they received. Those on the the ‘giving end (ie the payers) are less reticent to talk about the extent of corruption.”

In spite of the initial difficulties encountered by researchers in trying to measure the extent of corruption, there has not been a breakthrough in this regard (Gambetta, 2000; Berg, 2001; Jain, 2001). Gupta et al. (2000) were able to measure the impact of corruption on the provision of health care and education services in countries where private markets for social services were limited.

For example, see Table 4. Transparency International uses a corruption perception index [CPI] to gauge levels of corruption in countries (Lambsdorff, 1999). The objective of the CPI is to provide data on the perception of corruption in countries by applying surveys of businesspeople, risk analysts and the general public on the degree of corruption among public officials and politicians (Eigen, 2001; 2002; 2003). Every country is given a CPI score between 10 (highly clean) and 0 (highly corrupt). TI used 15 sources for data collection and collationin order to authenticate its reliability (Transparency International, 2002). The Corruption Perception Index 2001 reveals that corruption is prevalent in those countries which score less than 2, for example Bangladesh, Nigeria, Uganda, Indonesia, Kenya, Cameroon, Ukraine and Russia. According to the Transparency International Corruption Perception Index 2002, Angola, Madagascar, Paraguay were added to countries with a score of less than 2 points. Argentina, where it is perceived that corruption has greatly increased, joins Panama, Honduras, Guatemala, Nicaragua, Venezuela, Bolivia, Ecuador and Haiti with a score of 3 or less.

Source: Transparency International, Corruption Perception Index [CPI], 2001; 2002; 2003; 2004, and 2005

Note: Haiti scored 1.5 and ranked 145 in 2004 to become the most corrupt country in the world, thus displacing Bangladesh which equally scored 1.5. Chad ranked 158 and scored 1.7 in 2005 to become the most corrupt country in the world, thus displacing Haiti and Bangladesh with 1.8 and 1.7 respectively.

The methodology used by TI to generate data used for the CPI to a large extent does not reflect thorough sampling, because the information collected was based on the perception of a few elites who did not necessarily reflect the true corruption experience. For example, the number of surveys used for countries differ, also the extent and impact of corruption varies from one country to the other. The variation in using a uniform methodolody to survey the extent and impact of corruption still poses a problem. Whatever information collected by TI may be biased.

Bribe Payers Surveys are comprehensive research conducted on bribe- paying in international trade. They examine the factors influencing corruption, unfair business practices, private and industrial sectors. The Bribe Payers Index [BPI] addresses the tendency of companies from top industrialized countries to bribe in the emerging global market (Eigen, 2002). The first ever detailed BPI was released by Transparency International in May 2002. The BPI was based on surveys conducted in 15 emerging markets by Gallup International Association on behalf of TI. The BPI was carried out in the following countries: Philippines, South Korea, Indonesia, Russia, Colombia, Nigeria, Poland, Morocco, Hungary, Argentina, Mexico, Poland, South Africa and Thailand (Lambsdorff, 2002). These countries as described by TI (2002) are among the largest emerging markets involved in trade and investment with multinational firms. From Table 5, it can be deduced that there are very high levels of bribery in developed and developing countries.

Source: Transparency International, Bribe Payers Index and Bribery in Business Sector 2004
Note: 10 .0 is perceived to be a perfect score (for example, least corrupt). It reduces as a country is perceived to have the propensity to pay bribe. This reveals heavy bribe paying by domestic firms in developing economies. TI (2002) agrees that there is a global problem of corporate bribe paying.

Also, TI claims that this study consists of reliable and credible sources that apply different methodologies in order to understand and reveal how corruption is perceived from one country to another. According to TI (2001; 2002; 2003; 2004 and 2005) for example, it used many sources and independent institutions for the surveys covering all the countries highlighted in the Corruption Perception Index (CPI). The data are based on people’s perception. Since these data were specific on interpretation of perceptions and the understanding about corruption, actual levels of corruption cannot be determined precisely. Perception may act as a guide to understand the extent (TI, 2002).

TI has used the Corruption Perception Index (CPI) over the years to inculcate a consciousness of greater accountability and transparency in many countries. That notwithstanding, according to Osuji (2004), Fredrick Galtung observed some flaws such as the CPI which focuses on receivers of bribes; questionable respondents’ base; biased sampling framework; irregular coverage of countries and disregard for conscious reform efforts of governments. How genuine is the conscious reform effort? For example, the efforts made by the Nigerian government to combat corruption through anticorruption agencies seems genuine to outsiders; but to insiders, that is Nigerians, it is not a conscious reform effort, not that the legislators and anticorruption agencies are incapacitated to function effectively, many of them are alleged to be corrupt.

Kaufman (2004, p. 1) futher argues that:

“There is a margin of error in all estimates and they should be regarded as preliminary orders of magnitude. For example, the $1,000 billion estimate of annual worldwide surveys does not include the extent of embezzlement and tainted procurement in the public sector.

For example, Berg (2001) points out that most of the current studies on corruption have concentrated more on one or more perception-based measures of corruption, which are on surveys or polls asking respondents how they perceive the level of corruption in a country. He further points out that the surveys are typically valid and truth worthy, but may not be accurate and are often imprecise. Similarly, Berg (2001), identifies problems associated with the accuracy of perception-based indices as: the perception of general corruption in a country lag behind the actual level, resulting in low sensibility to policy changes and problem that perception indices may be endogenous, for instance media coverage of published corruption ranking may influence people’s perception of corruption in the country.

Treisman (2000, p. 438) supports that:

“Its many likely determinants interrelate in complicated ways. Some can change quickly and may be caused by corruption as well as the reverse. As with other types of criminal activities, it is hard to observe directly, and so researchers must rely on surveys of corruption’s victims, the accuracy of which is often difficult to assess.”

The problem of measuring corruption is linked to the fact that the methods used by various researchers are not explicit and more so majority of them use indicators that measure the perception of corruption rather than the true level of corruption. Svensson (2005, p. 21) argues that:

“Measuring corruption across countries is a difficult task, both due to the secretive nature of corruption and the variety of forms it takes.”



Impact

The measurement of impact of corruption according to Miller (2001) is as difficult as measuring its extent. Since the mid 1970s, corruption has been the focus of increasing legions of researchers in various fields cutting across all disciplines such as Social Anthropology, Geography, Economics, Political Science, Law, History and Accountancy (Corey, 2004). In trying to understand corruption better researchers have involved themselves with cross-country analysis on professional inquiry of the impact of corruption in many countries (Gambetta, 2000; Lambsdorff, 1999).

According to Berg (2001) the methodological differences between these disciplines have markedly made it problematic for researchers to benefit from or coordinate research from other disciplines. Van Roy Edward (1971, p.10) points out that:

“The heart of the matter thus remains unclear: Corruption appears to maintain systematic stability and yet also reflects change; it seems to be both functional and dysfunctional, equilibrating and dis-equilibrating, a permanent fixture of an ongoing arrangement and transient symptom of changing times.”

Tanzi (1994), Ades and DiTella (1997), Mauro (1997), Gray and Kaufmann (1998), Klitgaard (1998) and UNDP (1997, p. 36) have shed light on the problems encountered by researchers in trying to obtain valid data on the impact of corruption with methodologies which are not transparent, because they are based on the opinions of people knowledgeable about the countries concerned. People’s perceptions are commonly used (Miller et al., 2001). For example, the World Value Service surveyed 1000 randomly selected people in each of the 40 countries in the 1980s based on their perceptions, and again in the 1990s, by asking them:

“Would you say that most people can be trusted or that you can’t be too careful in dealing with people?”

Leite and Weidemann (1999) carried out a survey in 72 countries in order to measure the first variable of a country’s exports of fuels and minerals as a share of GNP. Leite and Weidemann (1999) found significant increase in the variables, which increases the level of corruption in a cross section of the countries.

Mauro (1995) analysed the effects of corruption by focusing on the relationship between investment and corruption. Mauro (1995) used Business International [BI] Index of corruption on GDP for the period between 1960 and 1985 for a cross section of 67 countries. His findings reveal that corruption lowers investment. Countries with high levels of corruption have difficulties attracting investors (Cote-Freeman, 1999). For instance, the U.S. Ambassador cautioned that corruption should be tackled in the Philippines or it would lose investors (The Strait Times, 18 July 2002).

Miller et al. (2001) used a combination of methods such as in-depth interviews, large-scale structured-interview surveys and focused-group discussions, and these were used on over three hundred participants spread widely across the regions of Ukraine, Bulgaria, Slovakia and Czech Republic. The focus group discussions were free-ranging, involving typically 6 and 12 participants and lasting between one and two hours and was loosely guided by a prepared schedule of topics. The in-depth interview was carried out in private instead of the usual round table discussion (Miller et al., 2001). The Miller et al (2001) sample covered a wide range of the public, such as ethnic minorities and members of the titular nationality living within the ethnic minority area; and nation-wide quota sample of street-level officials in health, education, social services, the police and legal services. The number of focus-group discussions commissioned was 26 with 187 participants and 136 semi-structured in-depth interviews in December 1996. In 1997 –98, 4778 fully structured interviews with representative samples of the public with another 1272 interviews were added, using a fully structured questionnaire. Miller et al (2001) interviewed about 1,307 junior officials with many of the same questions of “perception, gossip and actual experience” used in 1998 in interviewing the public.


The research reported in this thesis is the first to measure the extent and impact of corruption truly from the perspective of the real payers at household level in education and health care, not only in Ibadan, but Nigeria in general. In the light of these methodological arguments, there is a burning need for more studies that measure extent and impact.


Conclusion

There is commonality in all the definitions of corruption because of its condemnation. The definitions of corruption in this study centres on the notion of ethical functioning of public concern. Good organizational conduct is the sine qua non to every undertaking. The implementation of rules and regulation constitutes a problem in tackling corruption. This has often hinged on the complexity of rules and regulations. For instance, McMullan (1961) and the New Vision (31 March 2003) posit that multiplication of laws could multiply the possibilities of corruption. Similarly, Skeel and Stuntz (2002) argue that those who break laws escape from punishment because they invest in creative ways to circumvent the law.

It could be seen from what is happening daily that corruption is very widespread and has numerous impacts, not only on socio-economic but political aspects of people’s lives. It is a fact that there is diversity of opinion on what corruption is, but everyone condemns it. Although some argue that corruption is beneficial, the negative aspects far outweigh the positive aspects because of its direct consequences. Furthermore, the measurement of impact and extent of corruption is highly problematic. There is a need for more work to be done in this area, especially at the household level in developing countries.