The Electoral Commission (IEC) held two registration weekends (in March and April). Earlier this month it released a report on the numbers of people who registered to vote. This post will look at that report and tease out a couple of themes from it. All the pictures in this post are screengrabs from the report, which you can download here.
Firstly, the bottom-line numbers seem to be quite good: just under 1.4 million new registrations were recorded over the two registration weekends, beating the IEC’s target of one million new registrations. This brings the total number of registered voters up from around 25 million to over 26.3 million.
However these new registrations only represent a fraction of voter registration activity. Take a look at the table and graph below, which compare this year’s registration activity to the registration weekends before the 2011 municipal elections:
Where the streets have no name
The newly registered voters only made up about a fifth of all registration activity. In contrast, almost half of all the people who went to voting stations this year did so to re-register in the very VDs in which they’d previously been registered.
This suggests that these re-registrations have been driven by the IEC’s need to obtain better address data of registered voters following the Constitutional Court’s ruling that the IEC ‘is obliged to obtain sufficient particularity of the voter’s address‘. If you compare the registration figures from 2011 with those of 2016 you can see that in 2011 there were (relatively) few re-registrations of voters within the very same VDs in which they were previously registered.
A breakdown of the 2016 registrations by province supports this theory: the highest numbers of re-registrations were in the provinces with higher percentages of rural voters:
Most of the re-registrations within the same VDs occurred in KwaZulu-Natal, Eastern Cape and Limpopo. In contrast, Gauteng and the Western Cape had few re-registrations relative to total registrations.
Men and other minority groups
Women are in the majority in South Africa; about 51 per cent of the population is female according to the 2015 Mid-Year Population Estimate. You’d also expect the voting-age population (people 18 and older) to be majority women, based on first-year demography studies: there are more women than men in almost every age cohort and the proportion of women increases as you move from younger to older cohorts.
Still, it was a bit of a shock to find that the ratio of women to men in terms of new voter registrations was as high as 53 per cent to 47 per cent:
New registrations of women outnumbered those of men overall, and the gap was most pronounced in the 20-to-29 age cohort. It’s true that new registrations of men outnumbered women for all age cohorts between 30 years and 60 years, but it turns out that this doesn’t mean much in the context of overall registration numbers.
The gender gap in registered voters is even wider when you consider the total number of registered voters – women make up almost 55 per cent of total registered voters, as the data from the IEC website shows:
Women outnumber men in every age cohort of registered voters. And, with a ten-percentage-point spread, you wonder why politicians and parties don’t do more to woo the female vote.