Ideas & Debate

Tap census data for democracy and growth


Just when we thought we might have a quiet week, a senior politician reminded us of our fractious 2017 presidential election by demanding that Kenya National Bureau of Statistics (KNBS) “opens its servers”. The context was this week’s publication of the first volume of the 2019 Kenya Population and Housing Census report, which had clearly caused him discomfort over the numbers contained therein.

The statement was neither necessary nor useful. Today’s KNBS represents light years of positive change and welcome modernisation in comparison to its predecessor, the Central Bureau of Statistics.

Further, this week’s was the first of five volumes promised, with subsequent publications expected to capture population data by administrative units to sub-locational level, age and sex by county, political units, socio-economic characteristics and other themes (say, migration, or inter-marriage?)

However, for this sort of national exercise, we now need to be simultaneously publishing national and sub-national reports in a way that gets each county its own specific report in real time. Kenya is surely ready for a properly devolved KNBS that actively resides in County Statistical Units. More of this later.

Let’s keep it positive and visualise four “mindsets” we need to leverage the census for Kenya’s benefit.


The first is the DATA itself. We’re still growing at roughly a million people a year with falling household sizes (and incomes?). What do “correlation is not causation” analysts make of the drop from a “single political party” inter-censal population growth rate of 3.4 per cent, to a “multi-party” rate of 2.9 per cent - now down to 2.2 per cent over the roughly ten years of a new constitution?

With yesterday’s Kibra by-election in mind, I thought I would torture some data further, focusing on Nairobi (which KNBS classifies into 11, and not 17 sub-counties). Of the 4.4 million population, 2.4 million live in its northern to eastern outskirts. 1.5 million live within the Sh50 peak-time bus fare distance from the CBD, and the rest are in Westlands or Lang’ata sub-counties.

Within this Sh50 fare range, Kibra, Makadara and Mathare are the smallest — combined population less than 600,000. Think too about population density. In a nation with total land space that has reduced from 14 acres per person in 1969, to five in 1999 and three today, Nairobi is the most densely populated place, at 25 people per acre.

The range? Four persons per acre in Lang’ata versus 300 in Mathare. Kibra’s in between, at 61, and less than Makadara at 65. Do the data confirm the daily picture?

This is a simple Nairobi example. But what further data insights could we mine from the census?

We could go to our second and third mindsets. Our “forked tongue” politicians are already obsessing over DEMOCRACY and DEVELOPMENT for the people. All selfishly around the supposed forthcoming “critical event path” of BBI-referendum-boundaries review before 2022. Watch them attend Parliament in order to create or protect or otherwise gerrymander constituencies (or counties).

Then watch their “voter importation or transfer” strategies that result in weird electoral data Iike, how do we have 118,000 registered Kibra voters out of a 186,000 population? In one sub-county (constituency) in Nairobi in 2017, 100 per cent of the population was counted as registered voters! Indeed, “aging” Nairobi has more than half of its population as voters in most of its sub-counties!

Instead, how do we make the census data a “baseline” tool for serving the people? How does the “corporate” end of private sector utilise this census data to better understand that Kenya extends beyond Nairobi, and then do a “Bata” for the Kenyans without shoes? How do we build the “real” end of private sector by better understanding Kenya’s people and our socio-economics? Food for thought.

Which brings me to DIGITAL, the final “D” in the diamond. Remember “Huduma Namba” as a single source of truth? We’ve heard little, and, yes, we had concerns about data privacy and protection.

But, “out of the box”, wouldn’t the utility of this KNBS output be enhanced as a pilot for the development and rollout of digitally devolved/decentralised, yet secure “data and information portals” that support what I’m calling the “Four Ds” - data and digital for democracy and development? Recall the census theme - “Counting our people for sustainable development and devolution of services”.