Saturday, October 30, 2004

Bangladesh: TI Corruption Index - A Failure in Journalism


The approach adopted by TI has an inherently dangerous consequence for nations found at the bottom of this index. It may incite some unscrupulous businessmen to resort to further corrupting tactics in these states and on the other side prevent real FDI entering the country. This result, naturally affects all the citizens of the country, as poverty and corruption may become further ingrained if Bangladesh is not provided with the much needed foreign investment. In nearly every respect, Bangladesh faired better than its counterparts in South Asia. If we use TI’s own assumptions (in regard to the GCB results that deal with a very similar range of issues as the SAS) there should be some correlation between the SAS and CPI results - but there is none. What should Bangladeshis make of this discrepancy and why have none of our journalists pointed this out yet? They appear good at defending their rights but useless in doing their jobs.

A FAILURE IN JOURNALISM (THE TI CORRUPTION INDEX)
M.B.I.Munshi


1. The TI corruption perception index (CPI) has in the past, only taken perspectives and opinions of businessmen and not ordinary citizens or residents. In this year’s report the perceptions of residents have also been taken but it is not clear who these residents are e.g. resident businessmen, ordinary citizens, journalists, professionals, politicians, bureaucrats etc. As you have rightly mentioned, how corruption affects different sections of society vary immensely and how this differs from nation to nation is also highly variable. Massive or grand corruption carried out by a few elite families does not affect the vast majority of Bangladeshis who have no experience of it (at least directly) but it may affect big businessmen and foreigners adversely, while small scale corruption affects many people, its significance may be limited and it really depends on who is doing the bribing i.e. the giver or the taker. But as TI points out, “The sources do not distinguish between administrative and political
corruption or between petty and grand corruption.”

2. Instead, they have a Global Corruption Barometer (GCB) that gauges public sentiment about corruption. In the survey the following three questions were asked:

Question 1 looks at how seriously respondents believe corruption affects the different spheres of life, such as their personal and family life, the business environment, political life, and the culture and values of society in their country.

Question 2 investigated respondents’ expectations as to how the level of corruption will change over the next three years.

Finally, question 3 asks respondents for their first choice to eliminate corruption from an institution such as courts, political parties, police, the private sector etc. The results from these surveys conducted by Gallup International are all very interesting, but most important from a Bangladeshi perspective, was that a significant relationship seemed to exist between the TI Global Corruption Barometer and the CPI. In other words, in the case of countries with a low CPI score, that is countries perceived to be high in corruption, the respondents within these countries felt that corruption had a significant effect on the different spheres in their lives. This would have been a useful device to see whether there was any correlation between Bangladesh’s’ standing in the CPI and the GCB but unfortunately Bangladesh was not included in GCB survey of 3 July 2003 but both Pakistan and India were.

3. Instead, a special South Asia Survey (SAS) was conducted to investigate the perceptions of the most corrupt public institutions in Bangladesh, India, Pakistan, Nepal and Sri Lanka. It is interesting that a special note of appreciation was extended to Transparency International – Bangladesh for having coordinated the planning and design stages of the study. As with the GCB we can assume that there will be some correlation in results between the SAS and the CPI - so if Bangladesh has a high level of corruption in the public sector as reflected in the SAS then that will be mirrored in the CPI results where Bangladesh was ranked the most corrupt nation in the world in 2004 (that would be now four years in a row). The report that was published on 17 December 2003 entitled, ‘Corruption in South Asia – Insights & Benchmarks from Citizen Feedback Surveys in Five Countries’ came up with the following conclusions and results,

‘In India, Pakistan and Sri Lanka, 100% of respondents that interacted with the police during the past year reported encountering corruption. In Bangladesh, this figure was 84% and in Nepal, 48%. In their experiences with the judiciary, nearly all Indian (100%), Sri Lankan (100%), and Pakistani (96%) households polled reported paying bribes. Judicial corruption was also significant in Bangladesh (75% of users) and Nepal (42 % of users) ... After the police and judiciary, land administration was identified as the next most corrupt sector across the region, according to the experiences of South Asian households ... In Pakistan, 100% of respondents with experience with the land administration authorities reported corruption and in Sri Lanka this figure was 98%. Land administration was somewhat cleaner in Bangladesh (73% of users reported corruption), India (47% of users) and Nepal (17% of users).”

In nearly every respect, Bangladesh faired better than its counterparts in South Asia. If we use TI’s own assumptions (in regard to the GCB results that deal with a very similar range of issues as the SAS) there should be some correlation between the SAS and CPI results - but there is none. What should Bangladeshis make of this discrepancy and why have none of our journalists pointed this out yet? They appear good at defending their rights but useless in doing their jobs.

4. There is one other report that requires comment and raises further apprehensions and disquiet about the credibility of all these surveys - The Transparency International Bribe Payers Index ranks leading exporting countries in terms of the degree to which international companies with their headquarters in those countries are likely to pay bribes to senior public officials in key emerging market economies. In that sense, it measures the supply side of bribery in the countries where the bribes are paid. Countries are ranked on a mean score from the answers given by respondents to the question "in the business sectors with which you are most familiar, please indicate how likely companies from the following countries are to pay or offer bribes to win or retain business in this country?" This was one of my biggest complaints in my original article so I was pleasantly surprised that TI had in fact already considered this proposition a long time ago (2002). However, with the other reports and surveys this
one also suffers from a major impediment,

‘The Transparency International Bribe Payers Index (BPI) 2002, published on 14 May 2002, is based on surveys conducted in 15 emerging market countries by Gallup International Association. The BPI 2002 was conducted in: Argentina, Brazil, Colombia, Hungary, India, Indonesia, Mexico, Morocco, Nigeria, the Philippines, Poland, Russia, South Africa, South Korea and Thailand, which are among the very largest such countries involved in trade and investment with multinational firms. The questions relate to the propensity of companies from 21 leading exporting countries to pay bribes to senior public officials in the surveyed emerging market countries.’

In other words, Bangladesh was excluded from this survey so any benefit that might have been gained or derived in Bangladesh’s favor in relation to the CPI results is out of consideration. The more observant would have also noticed that India has been designated an emerging market country and was included in this survey (this has probably relegated Bangladesh into the category of a country never to be able to achieve emerging market status – with the CPI result this appears a foregone conclusion).

5. It is also wrong to suggest that the media role is negligible in the CPI report as one of the questionnaires submitted to businessmen requires their impressions to the following statement, “Extent of corruption as practised in governments, as perceived by the public and as reported in the media, as well as the implementation of anti-corruption initiatives.” In an objective analysis, this is a fair question but it has enormous drawbacks when considered in the circumstances I have already mentioned in my original article. In another context, relating to the BPI survey, the question was asked, ‘Please describe where your knowledge about this subject (perceptions about multinational firms) comes from?’ 55% of the respondents answered ‘press and media reports.’

There is another drawback admitted by TI, “In addition, in those cases where government and/or others have made substantial efforts to combat corruption, with demonstrable results, and where there is no improvement in a CPI score, there is the possibility that these efforts – however successful – have not been adequately communicated.”

6. There is an inherent bias in the perspectives represented in the CPI in favor of foreign businessmen and western nations. All the institutions where the surveys are carried out and perceptions recorded are based in the West i.e. Columbia University, Economist Intelligence Unit, Freedom House, Information International, International Institute for Management Development, a multilateral development bank, Merchant International Group, Political and Economic Risk Consultancy, Transparency International/Gallup International, World Bank/European Bank for Reconstruction and Development, World Economic Forum and World Markets Research Centre.

The manner in which the questionnaires are submitted to the business community and the responses are highly touch and go and lack certainty and discrepancies can be immense. The TI itself admits this problem, “The reliability differs, however, between countries. Countries with a low number of sources and large differences in the values provided by the sources (indicated by a high Standard Deviation) convey less reliability as to their score and ranking. Some countries may be overly represented as corrupt and others far less so.” If anyone bothered to look at the web site they would be able to understand my discomfort with this index.

The approach adopted by TI has an inherently dangerous consequence for nations found at the bottom of this index. It may incite some unscrupulous businessmen to resort to further corrupting tactics in these states and on the other side prevent real FDI entering the country. This result, naturally affects all the citizens of the country, as poverty and corruption may become further ingrained if Bangladesh is not provided with the much needed foreign investment.

I completely that corruption for ordinary citizens is like extortion. There is no doubt that corruption is widespread and pervasive in Bangladesh - I am not denying any of this. I do believe that corrupt people in our society should be severely punished and I have been saying so for more than 5 years and have written numerous articles on the subject. However, I believe that there is a direct bearing or nexus between how Bangladesh is perceived abroad and how well it can do in the future. If we blithely accept everything said about the country with an uncritical attitude then we are doing a disservice to the nation. Essentially, this is the role of the journalist and the writers but they have appeared to have taken a long leave of absence when Bangladesh’s interests are involved and most need of defence.

Email to Dak Bangla from M.B.I. Munshi