Coalitions have rapidly changed South Africa’s political landscape. The chaos that they have wrought on many local governments now threatens to spread to the national and provincial spheres after next year’s elections, threatening to do even more harm to citizens and the economy. 

Convened by Deputy President Paul Mashatile, last weekend’s National Dialogue on Coalition Politics was intended to start the process of bringing some order to the instability.  

Having been working on the issue for several years, often closely with political leaders who are wrestling with the challenge of building a new fit-for-purpose political culture, the national dialogue represented a necessary, but very far from sufficient, step in the right direction. 

Early polling suggests that the ANC may well fall below 50% of the national vote in 2024. The possibility is more definitive at provincial level and I am prepared to bet right now that the provinces of Gauteng (where no party received more than 36.1% in 2021) and KwaZulu-Natal will be hung after the 2024 election – meaning that no single party will win a majority.

The Free State, Northern Cape and even the Western Cape could possibly be added to this list.

There is little to suggest that the experience of coalitions at national and provincial levels will be markedly different from that which we have seen in the local sphere. There are admittedly some differences in the “rules of the game” between the different spheres of government, but the general principles of how coalition politics is “done” – as well as the political actors and personalities involved – remain the same, given that the national leadership of parties heavily influence coalition decisions at local level.

Such instability is concerning for South Africa’s prospects of recovery over the next five-year electoral cycle where we may see more than one president hold office, shifting government formations, and periods of political deadlock that push Parliament into a stalemate, potentially paralysing government and the public service.

But this instability, I would argue, is largely the short-term price that South Africans will need to pay as our democracy fundamentally shifts from a dominant party system and grows through its adolescent years into a more mature political system.

Genuine intent to govern collectively

Since 2018, I have conducted research on coalition politics as part of an initiative on coalition building, leading to two study tours abroad for South African political leaders.

This research has revealed several voluntary good practice standards of “doing coalition politics properly”, some of which could be translatable into the South African context.

But before any of this becomes relevant, there is one rather obvious condition for a well-functioning coalition to be established: a group of political parties in a legislature accounting for the majority of seats must come together with a genuine intent to govern collectively, setting some of their ideological positions aside, in order to deliver a stable government that can at least deliver uncontroversial basic services for the duration of the coalition’s term.

In many South African cases, the “coalition” lacks this indispensable founding premise.

South Africa’s three largest parties, the ANC, DA and EFF, together accounted for roughly 89% of the vote in the last national general election in 2019 and roughly 77% in 2021. 

Of these three, the EFF has repeatedly proved itself an unreliable coalition partner which does not genuinely want a stable governing solution.

With a political strategy built on instability and disruption, any coalition involving or supported by the EFF is almost guaranteed to be unstable. 

Both the ANC and DA are now publicly touting the idea of a “grand coalition” – ie a coalition between the two largest parties. But this is still an option of last resort to both parties, who will struggle to sell the idea to their respective bases.

In such a context, where the three largest parties who account for such a large majority of the national vote are unable to be stable and reliable coalition partners to each other, it is almost impossible to forge a lasting majority for any stable coalition to thrive.

The only path to a majority in these cases has typically been for a coalition, “led” by the ANC or DA, to be formed with a range of smaller parties that have representation in the particular municipal council. The result of these arrangements is a high number of political parties in a coalition, making it extremely difficult to manage and with huge potential for instability.

Political parties are inherently self-interested and motivated by the attainment of power which they need in order to translate their manifesto promises into government policy.

A large dose of patience

Political parties are not going to play nicely simply because we plead with them to do so; they will behave the way that benefits them the most. The best way to change the behaviour of political parties is the threat of electoral consequences. 

In the countries that have been our destinations for the study tours, Germany and Denmark, voters are in step with coalition politics. They understand it and expect political parties to behave like adults and ensure effective joint government. Parties that cause instability or break the coalition agreement are punished at the polls. 

Nice idea. Of course, this level of voter engagement with coalitions is the product of the many decades of coalition practice that exists in these countries – Denmark has not had a majority government at national level for more than a century – and a political culture that still seems quite foreign to the idiosyncratic South African case.

So, a large dose of patience will be needed as South Africa’s democracy matures towards a position where voters demand good conduct and political parties are held electorally accountable for actions that cause instability.

Can we afford to wait patiently? Can our political leaders raise their games and expedite the journey to a different political culture? And if not, what is to be done in the meantime? 

Legislating stability

Arguably the second-best way of managing the problem of unruly coalitions, although I would argue of far lesser value, is through legislation which tells political parties and their leaders how they must behave. 

Interestingly, a dominant part of the dialogue last weekend was preoccupied with the question of how to regulate coalitions. I am sceptical of the ability of legislation to plaster over the cracks in South Africa’s political culture, because such legislation has little international precedent, and attempts to legally restrain the natural course of inherently political actions often lead to bad law.

Coalition management should be governed by voluntary conventions developed by political parties over time, using lessons from international best practice. To facilitate this, an independent body may be formed. But it should serve only to encourage best practice and provide a forum for dispute resolution rather than attempting to “govern” public representatives.

However, there are some pieces of supporting law that can and should be amended to enable better coalition practice, but they cannot ensure it. 

Most glaringly, the time frame for the formation of coalition governments in South Africa (currently 14 days) is insufficient to negotiate proper coalition agreements and is far out of keeping with international practice where coalitions can take months, even years, to form.

Read more in Daily Maverick: Stable coalitions depend on politicians and political parties being honest and principled

Legislative measures that draw a clearer line between politics and public service should also be introduced, such as removing the municipal manager or accounting officer from the political crosshairs at local government level.

There may also be a strong case for limiting how often a vote of no confidence can be brought, to prevent the frequent chopping and changing of government that has been seen in some city halls.

The threshold question is the most controversial rule change that has been tabled. South Africa has no electoral threshold compared with an international average of around 3-5%. The result is a proliferation of political parties in legislatures.

In eThekwini, for example, there are 24 political parties in the 222-seat council, of which all but five have two seats or fewer. This leads to large-sized coalitions and creates extreme difficulties for stability.

Against the potential for enhanced stability must be balanced the original rationale of low electoral thresholds in South Africa, its uniquely diverse society and the potential for reconciliation and legitimacy presented by such wide-spanning representation.

Is the legislated extinction of these smaller parties justifiable collateral against the desire to give larger parties an easier time in forming and managing coalitions?

This issue will be hotly debated and, as some suggested at the national dialogue, may be better addressed in the next Parliament, given that it would seem unfair on these smaller parties to shift the goalposts within a year of the next general election in a way that would be so significant to their prospects.

Legislation is only a small piece of this puzzle – shifting political culture is the largest.

South Africans will have to stomach some short-term instability and, if this culture is to evolve, vote for the adults in the room when it comes to election time. DM

Mike Law is senior researcher at the Paternoster Group: African Political Insight and has since 2018 been a researcher and co-ordinator of an initiative on coalition politics convened by Prof Richard Calland, involving comparative study for senior South African political leaders.