Feats of the imagination and census fiction: The Mandera mystery figures

Thursday February 14 2013


While releasing the results of the 2009 census, the minister for Planning indicated that there were several counties for which the figures appeared inconsistent. He singled out North Eastern Province where it was considered imperative to undertake a recount to correct the glaring anomalies.

However, this raised a hue and a cry from the communities concerned, incited by MPs from the area and culminating in a court case against the government.

Arguing that the measure was discriminatory, the complainants, were granted an injunction halting the repeat count. Strangely, in spite of having enough grounds to lodge an appeal, the minister decided not to go ahead.

As a result, it is those inconsistent population figures that were afterwards used to decide on the distribution of the 80 additional constituencies. And it is the same Kenya National Bureau of Statistics figures that are now being used by the Commission on Revenue Allocation.

Thus, for instance, with supposedly the same population, Mandera (1,025,756) ends up with about the same level of entitlement to devolved funds as Machakos (1,098,584).

But it is when it came to voter registration that the disparities became most blatant. Taking the same two counties, with a theoretically comparable voting age population, Mandera recorded 121,005 while Machakos is at nearly four times, with 445,819.

One begins to appreciate why the opposition to the biometric voter registration (BVR) system was loudest from the same counties whose inflated census figures the minister was concerned about.

Conversely, former Mathira MP Ephraim Maina feels his county, Nyeri, was shortchanged in the sharing out of the 80 new parliamentary seats as a result of the census figures being well below what he thinks they should be.

The BVR figures seem to bear him out. If we apply the 46.6 per cent co-efficient used by the IEBC to work out the approximate voting age population on the actual number of registered voters in Nyeri County (357,105), we arrive at a total population of at least 766,320, well above the 693,558 census figure.

The same exercise for Mandera County (ie, taking the BVR figure of 121,005 as 46.6 per cent of the total population) gives a population of 259,667 instead of the 1,025,756 from the census. These statistics point to the likelihood of the KNBS figure being off the mark by 50 per cent.

The story is similar at the provincial level. If you consider the rate of population growth for Central and North Eastern provinces between 1999 and 2009, the 18 per cent for Central Province (from 3.7m to 4.4m) does appear closer to reality than the purported 140 per cent for North Eastern Province (0.96m to 2.3m).

Even more startling are the figures for Nairobi, and to some extent Nakuru. For Nairobi, out of a figure of 3,138,369 from the 2009 census, the IEBC was expecting a total of 1,463,105 to register. However, in the end, the number climbed to an astonishing 1,778,903.

Some people felt that the population of the city was higher than the figure given by the census. Here is proof that they were right.

Applying the same formula of 46.6 per cent to the actual (incontestable BVR) number of registered voters, clearly the total population of Nairobi is at least 3,867,180.

Why at least? Simply because it is rare in practice to register any more than about 90 per cent of the voting age population. We are dealing with two crucial issues, one of immediate consequence, and the other of medium- to long-term significance.

The latter relates to the matter of the lopsided revenue sharing which is not our main concern now. The immediate consequence, however, is that of the coming General Election. With opinion polls being bandied around which clearly do not take full account of contrived demographic imbalances across counties, people are likely to be shocked at outcomes that can only be the result of ballot stuffing.

This is more so because strongly TNA-leaning counties such as Kiambu, Murang’a, Nyeri, Kirinyaga, Nakuru and to some extent Nairobi, all showing figures close to, or in excess of, 100 per cent, were very likely undercounted.

Mr Keguro is the chief executive officer, Biz Ideas Management Consultants (www.biz-ideas.biz)