South Africa

South Africa

By-Election: Metsimaholo – with no outright winner and entry of SACP, prepare for strange bedfellows

There has been much focus on Metsimaholo and the town of Sasolburg where a by-election took place last Wednesday, though results were only finalised late on Friday afternoon. While it was clear that the ANC had lost ground and were now weaker in Metsimaholo than they are in Tshwane and Nelson Mandela Bay, it took a long time to see how the seats would be allocated. Here is a breakdown of how the different parties fared and what the next governing coalition in Metsimaholo could look like. By WAYNE SUSSMAN.

ANC 16 (19)

The ANC had been hurt by its alliance partner, the South African Communist Party (SACP) and the Economic Freedom Fighters (EFF). Some of the results were devastating for the ANC. The ANC percentage vote share fell in 20/21 wards. In Ward 4 (Refengkgotso Phomolong), the ANC% share of the vote fell from 70% to 44%, in Ward 1 (Zamdela Amelia) they went from 66% to 40%, and in Ward 3 (Refengkgotso Deneysville) they went from 55% to 35%, scraping past the SACP by a few votes.

While being hurt, there was some good news for the biggest party in Metsimaholo. It did enough in each of the 16 wards it defended to prevail and ended up getting 16 of the 42 seats, one more seat than the 35% of the total vote would have given them in the final seat calculation for Metsimaholo. So while the ANC in the Free State would have been upset by the returns on Thursday, their ability to narrowly defend a range of wards from the EFF, SACP and the Democratic Alliance (DA) helped them gain an additional seat.

Earlier in 2017, the ANC defended a seat in Ward 12 (Boiketlong) in Metsimaholo. They scraped home with 31% of the vote and saw off a three-way challenge from the EFF, and the North West-centric Forum 4 Service Delivery (F4SD) and local party Metsimaholo Community Association (MCA). Ward 12 would be the only area where the ANC’s share of the vote would increase from the last election in the ward to now, with their share of the vote going up from 31% to 50%. Their share of the vote would also prove to be sticky in the small village of Oranjeville and the adjacent Metsimaholo township. Here their share of the vote would only go down by 2% from 60% to 58%. This seems to support the trend that the ANC’s support base is more robust in rural areas than suburban or urban areas.

DA 11 (12)

Wednesday was not just a rebuke for the ANC, but also for the second largest party in the municipality, the DA. The DA defended five wards and their share of the vote fell in all but one ward, Ward 18 (Vaal Park Barrage). The Freedom Front Plus (VF+)’s share of the vote went up in all DA-held wards. They also hurt the DA in the marginal ANC-held Ward 20 (Refengkgotso Kragbron). What was worse for the DA was that while the party was propelled by high turnout in the suburbs in 2016, turnout this time round was sharply down in vote-rich areas for the DA, averaging at about 25% less than 2016. There was lower turnout in the ANC-held wards as well, but the lower turnout was not as pronounced. It seems as if the voters in the suburbs did not only have anger towards the ANC, but the biggest party in the governing coalition in Metsimaholo as well. The DA will have to be a part of any non-ANC coalition which governs in Metsimaholo, but they will know that they can no longer just expect their base to show up on election day and/or stay loyal to them. The party will need to ensure that when the new government for Metsimaholo gets elected in council, it shows real delivery to the residents of Metsimaholo.

Of the 14 wards in the townships, the DA was able to increase its percentage share of the vote in 10 of them, but was only able to get more than 10% of the vote in three of them. Their percentage share of the vote remained stagnant in two of the wards and fell in the remaining two. The DA did not have enough growth in the townships to mask their decreasing returns in the suburbs, and ultimately increase their representation on the council.

EFF 8 (8)

The EFF made great strides in some township wards but their growth was not uniform, and in some ANC-held wards their share of the vote went down. They ended up where they began with eight seats. They came very close to winning Ward 7 (Zamdela Transnet) off the ANC. The ANC beat the EFF by 64 votes, getting 37% of the vote against the EFF’s 33%. In 2016, the ANC won 61% of the vote here. Another ward where the EFF ran the ANC close was Ward 21 (Harry Gwala Coalbrook). The EFF lost here to the ANC by 108 votes, getting 34% of the vote to the ANC’s 43%. In 2016 the ANC got 62% compared to the EFF’s 22%.

The second most likely pickup for the EFF was Ward 3 (Refengkgotso Deneysville). Deneysville is a small town approximately 30km from Sasolburg. The EFF got 377 votes less than ANC in 2016, with the ANC getting 55% of the vote, compared to 37% for the EFF. The EFF lost a lot of ground here last Wednesday, shedding votes to the SACP and only getting 21% of the vote. The most likely pickup for the EFF was Ward 12 (Boiketlong), where the EFF lost narrowly to the ANC in a by-election earlier in the year. The EFF’s share of the vote went down from 24% to 22% on Wednesday. The EFF vote share fell in 6/16 ANC held wards on Wednesday.


The SACP had a solid showing in their first electoral foray. If the ANC had lost a ward, they would have picked up a fourth seat (as they just missed out on an additional PR seat). The SACP had to contest against their own alliance partners, and the other political party which favours the colour red, the EFF. They came very close to also winning a ward – Ward 3 (Refengkgotso Deneysville), where the ANC beat them by 34 votes, getting 35% of the vote compared to their 34%. The SACP was able to get over 10% of the vote in six of the 16 ANC held wards. Their second best showing was 19% in Ward 7 (Zamdela Transnet). The question is whether they will continue contesting elections or put that strategy on hold if their preferred candidate wins the presidential race at the ANC conference in a few weeks’ time.

The Rest

VF+ 1

Despite the VF+’s solid growth on Wednesday, they are still stuck on one seat. They were able to hurt the DA on Wednesday, but were not able to attract enough DA voters to get that additional PR seat. It will be interesting to observe whether they will want to sit in a coalition which features the EFF and/or the SACP.

African Independent Congress (AIC) 1

The Matatiele-based party had an impressive foray into Metsimaholo. They won a PR seat and will be an ally for the ANC if they have any chance of ruling in Metsimaholo. The party has assisted the ANC in coalition formations in Ekurhuleni and Rustenburg and would be counted on to support them here again. They will also be satisfied that they beat both the MCA and the F4SD.

MCA 1 (2)

When the first results came in, it seemed as if the local political startup, Metsimaholo Community Association, would be wiped out. They ended up holding onto one of their two seats, and will struggle to be the factor they were in Metsimaholo after the 2016 election. It is not certain who they will work with, but working with the opposition is a simpler path to government than working with the ANC.

F4SD 1

The North West political startup, Forum 4 Service Delivery, made a real first impression in the area when they got 23% of the vote in a Metsimaholo by-election earlier in 2017. They struggled on Wednesday, and while they were unable to replicate what they did earlier in the year, they did enough to win a seat.

Council Composition

Party Seats (Brackets show previous total)
ANC 16 (19)
DA 11 (12)
EFF 8 (8)
VF+ 1 (1)
MCA 1 (2)
F4SD 1
Total 42

Coalition Talks

Scenario 1: DA (11) + EFF (8) + VF+ (1) + MCA (1) + F4SD (1) 22/42

The most likely scenario would see the DA returning the favour to the EFF. This week the EFF lent their votes to the DA to help defeat motions of no confidence against the respective mayors of Johannesburg and Nelson Mandela Bay. The two parties together would have 19 out of the 22 votes needed to form a coalition which could govern in Metsimaholo. The most palatable party for both to work with would be the F4SD. Both parties were willing to work with the F4SD to rule in Rustenburg in North West. The F4SD was formed by aggrieved ANC members and they are more likely to work the opposition then the ANC. The opposition would need another two votes. The VF+ are not going to work with the ANC. The question is whether the EFF would want them in a coalition, and/or whether their supporters would tolerate them sitting with the EFF (and potentially the SACP). The MCA supported the opposition after the 2016 election. While the party was split, the leader who worked with the ANC is no longer in the party, and it is more plausible to expect them to work with the opposition again. So, the EFF and the DA have a realistic path to the needed 22/42 seats. The elephant in the room is the SACP and their valuable three seats.

Scenario 2: ANC (16)+SACP (3) + AIC (1) + MCA (1)-21/42 Hung Council.

The ANC knows that the first party to join them in a coalition in Metsimaholo would not be their tripartite alliance partner, the SACP, but the AIC. The ANC (16) + AIC (1) would leave the ANC five short of the magical number of 22. Let’s assume that Gwede Mantashe can broker a deal between the ANC and the SACP, and even offer the SACP the mayoral chain – that would still only take the ANC coalition to 20. They would be two short. The MCA had mixed feelings last time, and maybe the ANC could persuade them again. We would now be in hung council territory: 21. Unless the ANC offered the F4SD some rich pickings in Rustenburg, where they could offer them a place in the coalition, it is unlikely for the F4SD to work with the ANC. We know that the DA, the EFF and the VF+ will not work with the ANC. Thus it is more plausible that the ANC will not be governing in Metsimaholo.

Scenario 3: DA (11) + EFF (8) + SACP (3)

Ironically, a scenario which would be most stable numerically but the least stable ideologically would be a coalition with the DA (11) EFF (8) and the SACP (3) which would give them 22 seats. The DA would have to resign itself to either an EFF or an SACP mayor. The coalition could count on support from the F4SD, VF+ and possibly the MCA on policies which would suit the smaller parties, but at the same time they would not be held ransom by the smaller parties. The DA would be reluctant passengers here, but no scenario is ideal for them. Neither scenario 2 nor 3 is ideal for the SACP, and of course, scenarios 1 and 3 are not ideal for the EFF. Strange bedfellows will be holding hands in Metsimaholo.

The Other Contests

The ANC and DA also had a tough day at the office in the other contests on Wednesday, with it being worse for the ANC.

The ANC lost Ward 10 in uMlalazi (Mpungose Eshowe) in KwaZulu-Natal to the Inkatha Freedom Party (IFP). The IFP won by 274 votes, getting 54% of the vote, compared to the ANC’s 43%. In 2016 the ANC got 77% of the vote in the Baptist Church voting district; they only got 50% on Wednesday. The IFP won the voting district of Gawozi Junior School getting 58% of the vote. In 2016, the ANC won 61% of the vote here, their share of the vote was reduced to 40%. The IFP won 4/5 voting districts in the ward, only winning 1/5 in 2016. The EFF (2%) and the DA (1%) made up the numbers in this ward. The ANC now has 29/54 seats in uMlalazi, with the IFP going up to 23 seats. The EFF and the DA have one seat each.

The ANC held Ward 32 (Mgojweni Lujizweni near Libode) in Nyandeni in the Eastern Cape. The ANC’s percentage share of the vote fell to 68%, down from 83%. The UDM went up to 32% from the 9% they got in 2016. Turnout was 39%. Nyandeni is part of the ANC firewall of OR Tambo District. This result would concern them.

The DA held Ward 102 in Cape Town (Brackenfell Kraaifontein). This was one of the safest DA seats in Cape Town. The former councillor defected to CAPE Party. She was well known in the area and had served the ward for 11 years. CAPE got a solid 19% of the vote, winning 731 votes. The DA won 2783 votes and their percentage share of the vote was 71%, down from the 94% they got in 2016. The VF+ came third, with 8% of the vote. Turnout was a very low 22%. The DA had similar challenges here as they in Metsimaholo, with more conservative voters choosing to lend their vote to CAPE or the VF+. The party will be concerned about this.

It was a big week for by-elections. There is a short turnaround till the next round of by-elections which will be on Wednesday. There are three contests with the most high-stakes contest centred on Ward 18 in AbaQulusi (Near Vryheid) in Kwa-Zulu Natal. The IFP are defending a competitive seat against the ANC. An ANC win there will result in a hung council. The municipality is currently governed by the IFP and the DA, with outside support from the EFF. DM