The first observation is that coalitions at municipal level are here to stay. The electoral system for local government is, as instructed by the Constitution, based on proportionality. It is simply not designed to produce an outright winner. While there often are outright winners in municipal elections, it is a mistake to think that coalition are an anomaly. The quickest we adjust our thinking from “condemning” coalitions to examining how to improve them, the better. Second, coalitions are not bad news per se. For example, scholars like Lijphart conclude, based on global comparative research, that consensus democracies produce more inclusive and compromise-based policies, and, wait for it, more stability, than majoritarian governments do. However, and thirdly, coalitions at municipal level in South Africa are mostly chaotic, with devasting impacts for municipal administration and service delivery. We are doing something wrong.

The analysis of why we are getting things wrong is best left to political scientists. But I am confident in making this observation. The failure of dominant political parties (ANC, DA, EFF and ActionSA), to find each other in pragmatic local coalitions hands enormous and unwarranted power to the smallest parties. The smallest parties have our major political movements by the short and curlies because they outwit the dominant parties in coalition negotiations. As a result, many coalition agreements are gross distortions of the will of the electorate. When a political party with less than one percent of the voters behind it, is handed the mayoral chain or the speaker’s hammer, it has nothing to do with democracy. That is a subversion of the will of the electorate. Furthermore, the coalition government is then doomed to fail as we observe time and again. There is more to be said about coalition politics, but the rest of this contribution focuses on what the law can do.

Before suggesting some ideas for law reform, three remarks can be made. The first is that the solution to the coalitions crisis is not a legal one. It is not possible to legislate political behaviour, or force political parties into stable coalitions. Much of such efforts would be unconstitutional anyhow. On the other hand, when the current laws were drafted in the late 90ties, coalitions were not on the radar. Third, while it is not possible to legislate stable coalitions, it is possible to change those legal aspects that contribute to instability.

In that vein, I make six proposals. They are not worked out in detail, and there may be good arguments against them. However, they are presented to stimulate and raise the quality of the debate around coalitions.

Introduce electoral thresholds
We must introduce electoral thresholds to limit the number of political parties that gain seats in municipal councils. This argument has been made convincingly by Sutcliffe and Bannister in detail that needs no repetition here. The current legal framework is an “Oprah Winfrey” system in which “everybody gets some”. Yes, it results in a multiplicity of voices in the council, but this has mostly produced chaos and stagnation in hung councils. The municipal council is not just a collection of disparate political views. It is also a body with the responsibility to govern and make decisions. The current indecision has nothing to do with inclusivity or progress, because it retards and frustrates democratic decision making. The proposal for thresholds can be implemented relatively easily with small amendments to the electoral laws for local government.

Change to executive committees
A further suggestion is to replace the executive mayor system with the executive committee system. Executive mayors are elected (and removed) by council majorities, and handpick their mayoral committee members. When the executive mayor vacates office, the entire mayoral committee vacates office too. In a hung council, this last feature, and the fact that each seat on the executive is subject to the whims and failures of coalition negotiations, is fertile ground for instability. Executive committees work differently: the political composition of the executive committee is largely fixed by law, and thus not subject to majority rule. Coalition negotiations can then focus on the position of the mayor, the chief whip and committee chairs. Importantly, when the coalition collapses, and the mayor is removed from office, the rest of the executive committee stays on. It thus makes for a more stable governance system for hung councils.

There are various ways in which this can be implemented. The law already provides for executive committees, and approximately 50% of our municipalities have them. All municipalities in KwaZulu-Natal have executive committees because executive mayors are outlawed there. The MEC for local government may change a municipality from an executive mayor system to an executive committee system. However, this should not be done in the middle of a council term, as it throws a municipality into disarray.

Municipalities must limit motions of no-confidence
The motion of no-confidence has become the local politicians’ toy of choice. The most troubled councils experience endless motions of no-confidence, which are paraded whenever the chips fall differently compared to the week before. They are tabled and, if carried, celebrated as a victory, because the enemy mayor, speaker or chief whip is gone, right? However, these motions can throw the municipality into prolonged, leaderless paralysis, as happened in the City of Tshwane in March 2023. This is devastating for the municipal administration, and for residents. The law can limit the use of motions of no-confidence. One suggestion is to introduce the so-called “constructive motion of no-confidence”, which is a rule that only allows a motion of no-confidence if it is accompanied by (the votes for) an alternative. In simple terms, you may only dismiss the mayor or the speaker, if you have the votes to install a new one. This could be introduced by a change to the Municipal Structures Act of 1998. However, municipalities may also take the initiative, and include this rule in their own rules and orders. They have the constitutional authority to govern their own internal arrangements, and need not wait for the national government. Other suggestions are for a municipality to firm up its rules and orders to disallow multiple “similar” motions of no-confidence, or even limit the tabling of such motions to certain window periods.

Remove the secrecy surrounding election of office-bearers
In a hung council, the election of the (executive) mayor, speaker and chief whip is a key expression of a coalition agreement, however fickle that agreement may be. Currently, the law insists that these elections are held by secret ballot. It’s not clear what great democratic principle is served with this secrecy. It is defended, also by the courts, with reference to the secrecy an ordinary voter is entitled to, when voting in a general or by-elections. But this is a false equivalence. There is a difference between an ordinary voter who casts his or her ballot on election day, and a public representative voting for the election of an office-bearer in an open council meeting. The ordinary voter must be free to vote according to his or her conscience, and their vote is nobody’s business but theirs. The councillor, however, has been elected into office to carry out a party-political or independent mandate. The public is entitled to know how he or she carries out that mandate. Furthermore, party discipline, whether we like it or not, is essential for the stability of coalitions. There is nothing undemocratic about parties enforcing discipline within their ranks, provided it is done within the limits of the law. The secret ballot pulls the rug from under a party’s ability to enforce discipline, and undermines the stability of coalitions. Furthermore, it creates a fertile ground for sinister backroom dealings. It is thus suggested that the Municipal Structures Act must be changed to ensure that councillors cast their votes for office-bearers in public. The somewhat elusive democratic benefits of the secret vote do not weigh up against the havoc of perennially unstable coalitions.

Force coalition partners to publish coalition agreements
It has been argued before on these pages by Beukes, that coalition agreements should be made public. There are two main reasons. First, the coalition agreement is the political programme of the incoming municipal government, and the public is entitled to know the plans of its government. Second, having the agreement published makes it more difficult for coalition partners to violate the agreement, because the public can hold them accountable for their behaviour. Currently, coalition agreements mean little to those who signed up to them, and this is partly because the public does not know what they say.

Extend the 14-day period at the beginning of the term
After the general elections, the law insists that the newly elected council elects its main office-bearers within 14 days. In a hung council, this period is woefully short. Negotiating a proper coalition agreement takes longer. In the aftermath of the 2021 municipal elections, the result of this unreasonable deadline was that “the tail was wagging the dog”. Office-bearers were elected, based on the politics of the day, before a proper coalition agreement could be hammered out. Such arrangements stood very little chance to succeed. Indeed, most of them collapsed shortly thereafter. Therefore, it is suggested that the 14-day period must be extended to allow more time for negotiations, hopefully leading to coalition agreements that last. If it is not extended for all municipalities, then it must be done at least for hung councils. No municipality can afford a prolonged power vacuum at the top, but it must be possible to legislate a holding pattern after the general elections, in which either the outgoing municipal executive, or the municipal manager, is empowered to take the decisions necessary to keep the municipality going.

Conclusion
None of the above suggestions will solve the coalitions crisis, and there may be very good arguments against them. But hopefully, they will benefit the debate about coalitions. Because one thing is clear: if the past few months are anything to go by, leaving the crisis exclusively to the political actors in municipalities to solve in a mature fashion, would be to abandon the residents of those municipalities where governance is collapsing.

This article follows a series of articles by Jennica Beukes in this Bulletin on coalitions, and the publication, in partnership with SALGA, of a Framework for Coalitions.

By Jaap de Visser