This is the third instalment of a series on coalitions and 2026 local government elections. Read Part 1 and Part 2*. *

The meaning of the vote in South Africa in 2026 includes the acceptance that the winning party or parties will have to form coalition agreements with other parties to secure a majority in order to govern. Voting for a party also means voting for that partyâs coalition-partner choices.

This is the reality South Africans face in the local government elections of November 2026. A significant increase in the number of coalition governments is the most likely outcome, whichever party gets the largest share of the vote in any particular municipality. Few will gain enough votes to govern on their own.

Deals, compromises and consensus-seeking are the rules of the coalition game. One canât be sure which party will link up with whom, or whether post-election coalition arrangements will align with partiesâ election promises.

Coalition politics in South Africa are immature. There is little culture of policy-based, programmatic cooperation based on systematic negotiation and compromise between parties. Legislation to give order and legal standing to such deals has not come to fruition in time for LGE2026. Parties are more likely to use coalitions to undermine the opposition by enticing smaller parties into coalitions.

A vote in LGE2026 is a long way from a simple decision to vote for the party with which the voter identifies, or the party the voter believes has the best policy proposals or governance programme. And the voterâs individual choices across the two (in metros) or three (local municipalities beyond the metros) election ballots will not necessarily restrain or even guide the chosen party in its coalition formation.

Pre-election, most parties baulk at specifying their preferred or likely coalition partners. The DAâs âMoonshot Pactâ in the 2024 general elections, or the same partyâs 2026 promise to choose the âleast evilâ partner, are exceptions to the rule. (Note that the âMoonshot Pactâ fell apart in the light of sobering election results.)

In practice, political parties take calculated risks to achieve a joint total of â50 plus oneâ council seats. All South Africaâs parties have shown, in the coalitions formed thus far, that they will align with any other party, if need be, and will not shy away from coalition-crossing when they want to.

A notable exception is the ANC, which refused to align with the uMkhonto Wesizwe party (MK party) or with the Economic Freedom Fighters nationally. The ANC said it would not align with parties âfishing from the same pondâ, at least not nationally â in other spheres, especially locally, the ANC and EFF have co-governed in multiple municipalities.

This co-governance often takes the form of âsoftâ coalition agreements, such as de facto âconfidence-and-supplyâ arrangements, rather than formal coalitions.

Power, not policies

A few threads run through the incongruous bundle of forces driving coalitions, offering a small measure of predictability in LGE2026 coalition formation. The numbers are key here: parties will first check results, survey their alignment options in constructing an outright majority and enter the coalition that will provide that majority.

The smaller (and weaker, less demanding) the partners entering coalitions with bigger parties, the more operable the big partiesâ coalition government will be. The lead party will retain power, privileges and patronage to dispense.

Tracking by the multiyear Coalitions Barometer shows that the choice of coalition partner depends on the partiesâ exact vote proportions and the seat deficit of the lead party. Also influential are inter-party and inter-personal dynamics, as well as council histories, scandals and failures.

There is an âany party goesâ attitude to coalition partnerships. Numbers count more than policies. Locally, no independent councillor, community or other microparty that can contribute a vote toward a council majority is a party non grata.

For instance, nationally, the DA is the ANCâs main partner in the 10-member Government of National Unity (GNU), whereas in KwaZulu-Natal it is among several partners in a fragile, GNU-style Government of Provincial Unity (GPU). In Gauteng, the ANC and smaller parties formed a minority coalition government that relied on the EFFâs confidence-and-supply support until April 2026, when the EFF formally joined the Gauteng government by means of executive appointment. In local coalitions, the DA is largely not the ANCâs coalition partner of choice.

The extent of party support and thus a partyâs relative contribution to an outright majority will determine what happens after 4 November. There are now fewer places where one party wins outright power and can guarantee representation to their voters. By-elections have shown that the ANC is increasingly unlikely to gain an outright majority to govern on its own.

The DA will remain a major player and could achieve modest growth, but will only occasionally be a majority party. The Inkatha Freedom Party (IFP) appears destined to remain a big player or medium-sized kingmaker in parts of KwaZulu-Natal, but will win very few majorities. The EFFâs blocs of councillors could make or break alternative coalitions.

The MK party is likely to be a power-broker when itâs not an actual power holder in KZN municipalities. The Patriotic Alliance (PA) looks likely to grow and become a kingmaker; it has played a key role in earlier coalition deals.

Opportunism has elevated small and community parties to power-broker status, which seems counter democratic. But many do not have a long life: many parties that won a seat in LGE2021 have faded.

Whose vote counts?

The Coalitions Bill, which seeks to regulate and stabilise coalition agreements, is making very slow progress through Parliament. The proposal that parties winning less than 1% of the vote get no representation in council is highly unlikely to take effect this year, even if it is retained in the Bill. Hence, microparties will probably keep their kingmaking positions in coalitions and municipal executives.

The latest draft of the Bill has reportedly rejected the proposal that municipalities without an outright majority must have proportional mayoral committees. That would penalise the minor parties on whom the major ones have come to depend in governing by coalition.

In South Africaâs new era of coalitions, voters have learnt that coalition governments may trigger doubts about whose vote counts. They may find it is near impossible to fathom which party and/or councillor is to be held accountable for particular aspects of non-delivery or service failure.

Voters know that even well-intentioned councillors may hold zero sway over recalcitrant municipal managers and associated workforces that go missing in action.

The electorate has been stripped of the belief that votes are the key to favourable policy and governance, that elected representatives will be there to serve, will uphold constitutional guarantees, or even simply drive service delivery.

Voters are now disillusioned. In many cases, their votes have elevated power-hungry, corrupt, dishonest and opportunistic councillors. Candidate councillors that shine on the campaign trail can turn into, at best, non-performing ghosts and, at worst, malicious council criminals.

It is challenging to vote with the awareness that votes are cast into a failing municipal system of government.

Is a vote âwastedâ if the chosen party does not get an outright majority or become part of a coalition government? For many voters, even if their party lost, the meaning of their vote will be that they participated â that they had a say. DM