Pierre de Vos teaches Constitutional law at the University of Cape Town Law Faculty, where he is head of the Department of Public Law. He writes a blog, entitled ‘Constitutionally Speaking’.
As I write this column, the judges of the Constitutional Court are probably working feverishly to finalise their judgment that will determine whether the local government elections proceed on 27 October or will be postponed until early next year. The Court is facing an impossible choice, either to postpone the election despite the fact that the Constitution does not explicitly provide for this, or to decline to do so despite the possibility that the election will not be free and fair.
The IEC told the Constitutional Court last month that it was not to blame for this state of affairs, and that budget constraints and the difficulty of organising a safe, free and fair, election during the Covid-19 pandemic made it impossible to hold the election within the time frame provided for by the Constitution. I disagree. The Constitution imposes a positive obligation on the IEC to organise free and fair elections. It is required to act proactively to do what is necessary to fulfil this constitutional obligation, but it failed to do so.
The IEC failed to conduct in-person voter registration drives between the second and third Covid-19 waves, despite clear indications that a third Covid-19 wave was likely to hit in winter. It failed to ask Parliament to adopt new regulations to manage safety concerns during the election campaign, on voting day, and during the counting of votes. It failed to schedule the election before the five-year term of councils expired — which would have given it a 90-day grace period to conduct the election if Covid-19 made this necessary. It also waited far too long before asking Deputy Chief Justice Moseneke to assess whether the election could proceed in October, only to pass the buck to the Constitutional Court by asking it to make a decision that unelected judges should not be asked to make.
I have no idea whether the Constitutional Court will grant the postponement of the election, and if they do so, whether the reasons they provide will be legally plausible. But I am able to explain why — if the election does proceed in October — the failure of the ANC to register ward councillors in various municipalities and to submit proportional representation lists in some municipal councils is likely to have a significant impact on the number of seats won by the ANC in specific municipal councils.
Ironically, this is partly because the current formula used to determine the number of seats won by each party contesting metropolitan or local municipality elections, benefits the ANC, not least because it is a large and (previously well-resourced) party that is usually able to field ward candidates in many more wards than any other party contesting local government elections.
To understand why this is so, one must first understand that 50% of councillors in metropolitan councils and local councils with ward representation are elected in single-member ward elections, and the other 50% are allocated from the proportional representation lists submitted by political parties. This is done in accordance with a complicated formula to ensure that the number of seats won by a party in a specific metro or local council, is more or less proportional to the percentage of votes received by that party in that council election. Each voter has two votes, and may vote for not more than one ward candidate in the ward where the voter is registered, and for not more than one party who has fielded a proportional representation list in that municipal council.
For ward elections, each party may nominate one candidate per ward. Independent candidates may also stand for election in ward elections. In terms of section 8 of schedule 1 of the Local Government: Municipal Structures Act, the candidate who receives the most votes in a ward is the elected councillor for that ward. (If two or more candidates tie for first place, the result will be determined by lot.)
This means that a candidate could be elected as a ward councillor with far less than 50% of the vote, as long as he or she receives more votes than any other candidate standing for election in that ward. Ward elections tend to favour larger parties whose support is geographically concentrated. Smaller parties, whose support is geographically spread out (the Economic Freedom Fighters (EFF) is a case in point) struggle to win ward elections. A party could win 20% of the total number of votes cast in an election for a municipal council, but end up with no ward seats because its voters are spread more or less evenly across wards.
The 50% of proportional representation council seats are distributed to political parties in accordance with a complicated formula to ensure that the total number of seats held by each political party in a council is more or less proportional to the percentage of votes received by that party in that municipal council election. For example, if the ANC wins 30 of the 50 ward seats in a 100-seat council, but only wins 50% of the total percentage of votes in that council election, it would be allocated an additional 20 proportional representation seats to add to its three ward seats to ensure its total number of seats (50 seats) is proportional to the percentage of votes the party received overall (50%).
This means the number of ward seats a party wins is not always a good indication of its overall electoral performance in that municipal council election. In the last local government election, some ANC leaders in Nelson Mandela Bay who did not understand the system, complained that they were cheated because the party won 36 ward seats compared to the DA’s 23 ward seats, but after the allocation of proportional seats, the DA’s final number of seats was 57 versus the 50 seats held by the ANC. But the reason for this was that the DA had won 46.7% of the overall vote as calculated below, while the ANC only won 40.9% of the votes.
But this is where things get interesting. To determine the actual percentage of votes cast for each party in a municipal election, one must add all the votes cast for that party’s ward candidates (whether the candidate won the ward or not), to the total number of votes cast for that party’s proportional representation list and divide this number by 2. (I am simplifying somewhat to avoid the complicated technicalities.) This formula benefits political parties with ample resources and organisational muscle who are able to field ward candidates in each or most wards in a municipal election, and disadvantages parties who are unable to field candidates in each or most wards.
If we assume that voters who cast their ballots for a political party’s proportional list will mostly also cast their votes for the party’s ward candidate, parties who field ward candidates will have their votes “double counted” for purposes of the final percentage of seats it will be allocated, while parties who do not field a candidate in that ward will only have their votes counted once. Moreover, where a party does not field a ward candidate, its voters may cast their ward vote for the candidate of another party, thus increasing the percentage of votes garnered by the latter party.
This is why in municipal wide elections, political parties with sufficient resources will nominate ward candidates in most or all wards — even those wards where they have no prospect of winning the seat. For example, in the ward in Cape Town where I live, the ANC candidate usually wins between 5% and 8% of the votes in the ward election, but this is not insignificant as these votes are added to its proportional representation votes to determine the overall percentage of seats the ANC will win in the Cape Town city council.
This all means that the failure of the ANC to nominate ward candidates in some wards may have a significant impact on the outcome of the election in the affected municipal councils. Where the ANC failed to nominate ward candidates, it would “lose” the votes its supporters would have cast for the ANC ward candidate in that ward. Assuming these voters would have voted for the party list and the ward candidate, this in effect means the number of votes for the ANC in that ward would be halved for purposes of calculating the final number of seats held by the ANC in the council. While the negative impact on its overall number of seats in a council would be more dramatic where the party failed to nominate candidates in wards it is popular in, it would still impact on its overall performance where it failed to nominate candidates in wards it would normally receive only a modest percentage of the vote.
As I read schedule 1 to the Local Government: Municipal Structures Act, the situation is even more dire where the ANC has failed to submit a proportional representation list for a particular municipal council election, as such a failure would mean it does not qualify to have the party on the proportional representation ballot at all. In effect, this could more or less halve the percentage of seats that the ANC would win in that council. (It must be noted that where the ANC did submit a list, but the list is incomplete, it would not be a problem, because section 17 of schedule 1 does allow a party to supplement its list if it contains fewer candidates than the party is entitled to.)
All this means that a postponement of the election will probably be of immense electoral benefit to the ANC. If this happens, populist political leaders like Helen Zille and Julius Malema are likely to exploit this fact to try and discredit the Constitutional Court judgment or even individual justices of the Court. This means that if the Constitutional Court decides to postpone the election, it is imperative that it provides compelling reasons why it believes it is not possible to conduct a free, fair and safe election on 27 October, and that it does not let the IEC off the hook for its failure to deliver a free and fair election by the deadline set by the Constitution. DM
The sound of Krakatoa exploding travelled around the earth three times.
Daily Maverick © All rights reserved