28 Jul 2010

What does the East African Bribery Index say about #Tanzania?

Who do Tanzanians have to bribe before they get a service? Which institutions demand the most bribes? What actions to citizens take if asked for a bribe? And do citizens think corruption is increasing or declining?

Last week the Kenyan chapter of Transparency International published the East African Bribery Index, to provide answers to these questions and promote debate on corruption within the region. They've certainly succeeded in the second part of this, getting a fair bit of coverage in the press and drawing responses from the heads of several of the institutions found to be most corrupt, including the Permanent Secretary at the Ministry of Health and Social Welfare, Blandina Nyoni.

The index is based on a survey of over 10,000 respondents, 3,231 in Tanzania. It therefore looks at corruption from the perspective of citizens: how do they experience corruption, what services have they been unable to access without paying a bribe?

So what did the survey find? Well, you can read the full report here (pdf) but a few points are worth highlighting:
  • The police were found to be Tanzania's most corrupt institution. 77% of people reporting having tried to get a service from the police said they were expected to pay a bribe in order to access that service.
  • The judiciary (68%), the registrar of births and deaths (66%) and the Tanzania Revenue Authority (47%) also score badly.
  • Petty corruption is reportedly very widespread among almost all the services included in the survey. This includes hospitals, where 41% of people accessing hospital services being asked for a bribe, local authorities (41%), Tanesco (34%), micro-finance institutions (31%) and schools (19%).
  • Only 7% of people asked for a bribe reported this to anyone in a position of authority, with a belief that no action would be taken (39%) and fear of intimidation (20%) the main reasons for not reporting.
  • 85% of respondents felt that the country is either "corrupt" or "very corrupt", and 42% felt corruption had increased in the past year, compared to only 14% who felt it had decreased.
These are pretty shocking figures, and if they are reliable, they show petty corruption to be extremely widespread.

The survey does have some technical weaknesses, however, some of which may be significant. On methodology, for example, the report leaves out important details on how various figures are calculated (e.g. the Aggregate Index), and more detail in the tables (actual numbers as well as percentages, for example) would be more transparent. This is important since if only a small number of respondents had encountered a particular institution during the past year, the effective sample sizes would also be small.

And more worryingly, the demographic data presented on the respondents suggest that the sample was perhaps not truly representative of Tanzania. To experienced analysts of Tanzanian survey data, having 49% urban respondents and 40% of respondents from households with over TZS 300,000/- monthly income, raises some alarm bells.

The report also shows a lack of familiarity with Tanzania. It refers to institutions that don't exist in Tanzania, or refers to them by the wrong name. The "Office of the Ombudsman" is the clearest example of this, but there are others as well.

On the results, the highly respected Afrobarometer surveys ask some similar questions on citizens actual experience of corruption, producing some very different findings. For example, 59% of Tanzanian respondents to the 2008 Afrobarometer survey said they had not (in the past year) paid a bribe to avoid a problem with the police. There isn't necessarily a contradiction here, but it is not clear why there is such a difference. Without going back to the questionnaire or the raw data, it's impossible to know.

But this is nit-picking. The index is a very effective way of stimulating debate and putting a little pressure on the worst performing institutions to improve their act. Nothing holds someone to account like the sense that they're being watched. This index does exactly that, as the response it has provoked demonstrates.

CORRECTION: This post has been updated to correct the a mistake in title accorded to Mrs Blandina Nyoni. Previously the post stated that she was Minister for Health and Social Welfare, which has now been corrected to say that she is the Permanent Secretary.