Johannesburg: The ANC Council as an Innocent Victim?

Dr Ivor Sarakinsky, School of Governance. University of the Witwatersrand; Johannesburg | 11 August 2016

The 2016 Municipal Election has delivered a cruel outcome to the incumbent ANC City Council Leadership.

Higher credit ratings, and good performance in city management according to several different international benchmarks, improved management and administrative efficiencies introduced over the previous term and yet, the Mayor and his Committee may well be out of office should a coalition between the ANC and other parties fail to materialise. A cruel fate indeed.

The current state of play in terms of council seats stands as follows: ANC 121, DA 104, EFF 30, IFP 5, AIC 4 and FF Plus, ACDP, UDM, Cope and the PA, 1 each. The magic number for a majority is 136. This means that the EFF will be needed by either party to get a clear majority in the Council. The smaller parties added together, come to 15, just on the mark of the magic number needed by the ANC to retain a (slim) majority. The likelihood, however, of all of them joining a coalition with the ANC, is slim.

So, it is a hung Council and the vicissitudes of horse trading shall take their course. The outcomes only known when the spoils are sliced and diced and a coalition agreement or deal is sealed and signed.

How did Johannesburg end up this way?

There are two primary reasons.

First, the Johannesburg ANC has borne the brunt of voter dissatisfaction with the national ANC leadership. This, even though the Gauteng ANC has explicitly questioned the leadership of President Zuma and his cabinet. Poor economic performance, low jobs numbers, continuing scandals around Nkandla and the attempts at manipulating independent institutions, the imprudent replacement of the Finance Minister in December 2015, State owned enterprises in disarray, and the uncritical defense of the ANC Parliamentary Caucus and Ministers of the ANV leader and the Head of State, coupled with the outstanding charges of corruption, appear to have taken their toll. This precipitated an ANC voter stay-away.

So related to this first reason, is the second. A far more complex and interesting one. In simple terms, the Johannesburg outcome is directly caused by voter turnout. Although it is not accurate to confuse national, with local elections, the 2014 national elections is useful in establishing a base-line for comparison. Breaking down that result to Johannesburg, the result is vastly different to what emerged in 2016. With a 76% turnout, the result for Johannesburg would look as follows: ANC 54%, DA 30% and EFF 10%. With Ward and PR seats, the ANC would have won the Johannesburg City Council comfortably.

In the 2016 Local Government Election however, the turnout was 57%, a differential of 20%. How does this translate at Ward and PR level?

2016 election ward map: DA blue shades, ANC green shades
Map source: News24 2016 "Local election results for City of Johannesburg", (accessed 10 Aug 2016).

The 2016 election Ward map above shows the DA (blue and dark blue) and ANC (green and dark green) respectively, with no other party winning wards in Johannesburg. The darker hue of a colour indicates high turnout with a strong majority for a party. The number of dark blues far outnumbers the dark greens (which), but it is the lighter greens that help understand the dynamics of this election.

While the DA has attracted some black support, it still relies on its traditional white support base. The key issue is that the DA has not broken through in attracting significant black township voters in Johannesburg. Moreover, ANC supporters have withheld their vote while the DA has succeeded in getting its supporters to the polls in high numbers. The DA's growth can only be explained in terms of a spectacularly high turnout of its traditional voters. Drilling down into the ward results, there are large numbers of dark blues marking voter turnouts in a range between 70-80% with DA majorities of 75-85%. This explains the jump in DA council seats once the Proportional Representation allocation is made with the top up from the high margins by which they win in the wards where their traditional voter base votes.

In areas with an almost exclusively black population, for example Soweto in the south west and Alexandra in the north east, as well as informal settlements and new townships in the north, there are low turnouts of between 47-55% (below national average of just below 58%). Even with ANC voters staying away from the poll in a sign of protest, the DA could not win a single ward in these areas. In Soweto and other townships, the DA appears to have only marginally increased its ward support compared to the 2011 municipal election. In the far northern and some eastern wards, the DA wins with much smaller majorities and turnouts and the ANC still gets some support in those areas. These wards are low density as shown in the map below, which means that from these wards they do not carry large numbers of votes into the PR allocation.

The ANC still appears to receive support in Coloured and Indian areas even though the wards might be won by the DA. It is perhaps amongst Coloured voters where the DA has taken support away from the ANC in significant numbers. What the ward map above and the demographic map of Johannesburg below conclusively show, is that the DA has yet to break through and win large numbers of black voter's allegiance in Johannesburg.

Large numbers of black voters stayed away and very large numbers of white voters turned out and voted DA. This skews the turnout rate as well as the PR allocation in favour of the DA. If the two maps (above and below) are superimposed on each other, it is clear that voting patterns in Johannesburg correlate meaningfully with the demographic densities of the city and conform to the residual hangover of apartheid era spatial planning and population concentrations.

1 dot per 100 people. Red: White. Black: Black African. Green: Coloured. Blue: Indian.
Data from 2011 Census. Legend: 1 dot per 100. Red: White. Black: Black African. Green: Coloured. Blue: Indian.
GCIS interactive mapping from the GCRO: (accessed 10 Aug 2016)

Given this evidence, it is possible to argue that white voters' attitudes and political support are now rigidly behind the DA while black voters appear to be demonstrating significant flexibility in the exercise of their vote. They are exercising a range of different options. either: withholding their vote, from all parties, in part ANC voters sending a very strong message to the ANC leadership, voting for the ANC, the EFF, in incremental measure for the DA or a range of other parties. Johannesburg therefore illustrates, that black voting patterns are far more interesting and unpredictable.

With the final tally of seats on the Johannesburg City Council being:
ANC 121, Ward 84 PR 37
DA 104, Ward 51 PR 53.

The difference between the ANC and DA in the how their total is constituted confirms the analysis that the ANC was prejudiced by a low turnout of its supporters, while the latter benefitted dramatically from a turnout in the suburbs.

In both instances, race played a role in the choice of party and the decision to queue to cast votes. All of the other parties in the Council have representation due to the nature of the PR seat allocation meaning that they did not win any wards, but had sufficient overall support in order to garner a few council seats. These are:
EFF 30,
IFP 5,
and the FF+ Plus, Aljama, UDM, ACPD, Cope and PA receiving one seat each.

As the smaller parties are too numerous and diverse to form a coalition with the ANC, in order to give it a slender majority in Council, the horse trading to build a coalition is a three horse race with the ANC, DA and EFF being the major players. Who will get into bed with whom is the remaining issue. It appears that animosity and personality issues will prevent an alliance between the ANC and EFF even though policy wise they might be closer together. There is a strong possibility that the EFF and DA will get together to govern Johannesburg and Tshwane - with a trade-off deal struck between the two parties entailing these two important cities.

This will be a rocky marriage - as municipalities deal with property rights, land use and zoning as their core business. The EFF has made unambiguous policy commitments to returning the land to the people while the DA, at least in Cape Town, is open to private sector property developers accessing public and private land for high end development and upscale developments, malls and unaffordable gentrification of areas. This policy fault line is bound to emerge in the running of Johannesburg - once the nuts and bolts of municipal management and administrative affairs gets under way. It remains to be seen how, if at all, the newly-weds will resolve these fundamental tensions.

If the ANC finds itself in the opposition benches in City Council, then it has the advantage of having a detailed understanding of municipal affairs and will use that knowledge to hold the new alliance to account, through every means possible. The prospect of more effective and efficient local government pitted against procedural obstructionism in this likely to characterize the scenario going forward.

Finally, the IEC has correctly received high praise for the manner in which it has managed the 2016 Local Government Election. However, there is another independent body relevant to this election that has not yet received the credit it deserves. The Municipal Demarcation Board is responsible for setting the boundaries of cities and wards within cities. A close look at the wards in Johannesburg correlating as it does with racialised population densities and demographics, shows that there are a number of wards in the north, north east, east and south east where gerrymandering the boundaries would easily have swung the wards away from the DA to the ANC. The same applies to former Coloured and Indian areas in the south of the city. The fact that this has not happened, shows that the MDB has attempted to demarcate wards based on more objective criteria, while upholding its statutory independence.

Johannesburg shows that we are a long way away from re-jigged local boundaries and manipulated local elections. In the United States, towns and parts of towns will forever be governed by either Republican or Democratic representatives due to the gerrymandering of boundaries. The 2016 Local Government Election shows that two of the key independent bodies responsible for establishing the rules of competitive democratic contests; have discharged their responsibilities creditably.