Tshwane is very likely headed for a coalition government scenario after August 3. An assessment of the latest registration figures and voting patterns from both 2011 and 2014 indicates it is unlikely any party will reach 50% in the metro.
Ipsos’s running tracking poll currently shows all parties far under 50%. The DA leads the poll by a very significant margin (42% for the DA to 23% for the ANC), although the Ipsos poll very likely understates ANC support in Tshwane at this stage (I’ll dive into the details of that in another article).
Previous voting patterns and the latest registration figures also provide pretty good evidence that Tshwane is likely to produce a coalition outcome.
The Tshwane Electorate
Tshwane has roughly 700 voting districts (VDs) and 1.5 million voters. These VDs can be divided into five distinct groups:
1. Solid ANC VDs (where the ANC receives more than 70% of the vote)
2. VDs that lean towards the ANC (where the ANC receives 50-70% of the vote)
3. Marginal VDs (where a party wins with less than 50% of the vote)
4. VDs that lean towards the DA (where the DA receives 50-70% of the vote)
5. Solid DA VDs (where the DA receives more than 70% of the vote)
The 5 groups of voting districts have the following characteristics:
Based on these numbers, the ANC should be able to win easily. There are many more traditional ANC voters in Tshwane than traditional DA voters. The fact that Tshwane is even in play, suggests that all is not well for the ANC. So why is the ANC struggling in Tshwane?
Data from previous elections reveal three damaging trends for the ANC in the metro.
The ANC’s 3 big problems in Tshwane
1. Voter registration
The DA won a significant victory during the voter registration weekends earlier this year. Registration of new voters was much higher in strong DA areas than in strong ANC areas. The data is summarised below:
The DA was able to add over 28 000 new voters to the rolls in its strong areas, whilst the ANC was only able to add 12 000 voters to the rolls in its strongest areas. The DA base will therefore be larger relative to the ANC base in the upcoming election, creating a more favourable electorate for the DA.
2. The ANC is losing votes in its strongest areas
Since 2011, the ANC has steadily lost support in its strongest areas. Consider the table below:
Between 2011 and 2014, a significant number of voters moved away from the ANC in the ‘solid ANC’ areas of the city. This includes traditional ANC strongholds likeSoshanguve, Hammanskraal, Atteridgeville and Mamelodi.
The table shows that the ANC lost more than 10% of the vote in its solid areas, and a full 20% of the vote in the areas that lean towards the ANC.
These shifts were primarily caused by the EFF, with large numbers of voters moving away from the ANC and to the EFF in these areas. The DA also received a portion of the voters moving away from the ANC in these areas, but many fewer than the EFF.
If this trend continues and intensifies the ANC could be down to the low 70s in its strongest areas and into the 50s in the ‘lean ANC’ areas. This would have a massive impact on the final ANC vote tally.
And the preliminary evidence suggests that the trend will continue. The IPSOS poll shows the DA growing very strongly in black communities and the EFF holding onto its 2014 levels of support. This would certainly drive down the ANC majorities in these VDs.
Turnout is a massive headache for the ANC in Tshwane. The problem has two dimensions. Firstly, the DA is very, very good at motivating and turning out its supporters in Tshwane. And secondly, the ANC struggles to turn out its voters in Tshwane. Consider the data below:
In both 2011 and 2014, voters in strong DA areas turned out in disproportionately high numbers, whilst large numbers of voters in the strong ANC areas didn’t vote. This inflates the size of the DA base relative to the ANC base on Election Day and significantly decreases the ANC’s overall percentage of the vote.
You would also notice that the ‘turnout gap’ between ANC and DA areas is especially large in local government elections, and somewhat smaller in general elections. If the turnout gap is as big as it was in 2011 (+14% in the DA’s favour), then the ANC will have a real problem on its hands.
Again, the preliminary evidence suggests that the pattern could repeat itself and even intensify, given the unhappiness over the nomination of Thoko Didiza as ANC mayoral candidate in Tshwane. If even a small portion of ANC voters who turned out in 2011 stay away because of their frustrations with Didiza, it could sway the election.
Expectations for August
If we simulate an election using the latest registration figures, the 2011 turnout pattern and the 2014 voting pattern, the result is as follows:
But that assumes that everything will stay the same. It’s likely that the voting pattern will be less favourable to the ANC given what we know about the IPSOS polling to date and the unrest in Tshwane.
So I simulated the election using the following (conservative) assumptions:
1. DA supporters turn out in slightly higher numbers than 2011, because they think they can win.
2. ANC supporters turn out at the same level they did in 2011
3. The EFF and DA grows marginally in the townships
4. Opposition voters consolidate around the DA, because they think the DA can win.
The result of this simulation was as follows:
As you can see, in this scenario the election becomes quite close and nobody is within reach of 50%.
It is very important to note that this simulated outcome is based on conservative assumptions. More dramatic shifts could take place on Election Day. The DA base could go berserk and turn out at 75%+ levels, or there could be a huge stay away of former ANC voters. Or a number of other dramatic, unexpected shifts could take place.
What the model does show however is that very dramatic developments need to occur for any party to reach 50% on its own. Something big will have to happen.
The ANC is most likely to be able to reach 50%, but it will have to significantly outperform its turnout and margins over previous elections. At this stage, and given the current political climate, that seems like a tall order.
We could therefore be on course for a historic coalition scenario in the nation’s Capital.