The recent local government elections resulted in a number of watershed moments for South African politics, notably the decline of the ANC’s share of the national vote to below 50%.

However, another key trend which is likely to accelerate was the proportion of voters who did not vote for any of the ‘Big Three’ parties, the ANC, DA, or the EFF. In the previous 2016 election just over one in ten voters cast a vote for a party other than one of the Big Three. However, in this election, this number more than doubled, with more than 20% of voters supporting a smaller party.

The parties that benefited from these defections from the big parties ranged from established parties such as the IFP and the FF+, to coloured-interest movements (the PA and Cape Coloured Congress), to local interest parties (including Cederberg Eerste and Lekwa Community Forum).

This has seen an unprecedented number of parties represented on councils across the country. Given that the number of hung councils recorded (nearly 70 of the just over 200 local municipalities) is also a record, small parties will be part of governing coalitions in a number of places across the country.

At the same time, the number of small parties represented in some places can make a person’s head spin. In eThekwini, where the ANC saw its vote share plummet to 42%, there was a plethora of small parties which secured seats on the council with very small vote shares.

Five parties secured two seats on the city council (with vote shares ranging from 0.63% for the Democratic Liberal Congress to the 0.92% for the Active Citizens’ Coalition). However, a whopping 14 parties won one seat each. Al Jama-ah’s one seat was thanks to 0.19% of the vote, while the African Transformation Movement won one seat with 0.62%. A party called the KZN Independence Party (presumably a secessionist outfit, though there does not seem much public information available on the party) also managed a seat on the city council.

Thanks to a bloc

In Nelson Mandela Bay the ANC also had its mayoral candidate elected thanks to a bloc of seven other opposition parties, which supported Eugene Johnson. Five of the parties had one seat each, ranging from the 0.5% that gave the PAC its lone seat, to the 1.1% which was enough for the Abantu Integrity Movement to get a seat on the council.

As South Africa’s politics continues to fracture, expect these kinds of outcomes to eventually play out at national and provincial level, perhaps as early as 2024.

Once the ANC loses its national majority, and majority in the various provinces, expect unwieldy and strange coalitions, as we have already seen play out in various municipalities, with small parties wielding outsize influence.

Consider, for example, that Al Jama-ah, the smallest party in the National Assembly, made it to Parliament with 0.18% of the vote, winning just over 30 000 votes of the 17.4 million cast. In a future election where the ANC loses its majority and has to rely on outside support to remain in power, small parties like Al Jama-ah could play an outsize role in the affairs of the country. 

At the same time, hypothetical coalitions could also rely on a number of small parties to remain big enough to stay in power, with smaller parties making various demands to ensure continued support.

Electoral threshold

Which brings me to my question – should South Africa look to implement some form of electoral threshold which keeps smaller parties out of legislatures, whether these are Parliament, the various provincial legislatures, or the various city councils?

A threshold would mean that a party needs some minimum form of representation to secure seats in a legislature.

Many countries which use proportional representation for elections (like South Africa) have some form of electoral threshold. In Germany a party must win at least five percent of the overall national vote or three directly elected constituency seats to gain representation in the Bundestag. This threshold was first introduced in the 1950s in West Germany, partly to prevent the election of extremist parties as well as to prevent party fragmentation, which had played at least some part in the instability experienced by the Weimar Republic in the 1920s and 1930s.

Germany is far from the only country in the world to have some sort of threshold. New Zealand also has a threshold of five percent, but also allows parties to gain representation in Parliament if they win at least one directly elected seat.

Israel, too, has an electoral threshold. Israel uses a pure proportional representation system (with all MPs being elected from party lists, with no directly elected seats, the same system South Africa uses at provincial and national level). Its threshold has been increased over time, from one percent in the 1980s up to the current 3.25% (equivalent to about four seats in the 120-member Knesset).

These are but three examples of electoral thresholds which are common across many of the world’s democracies.

More stable

If South Africa had some form of explicit threshold there would likely be fewer parties in municipalities around the country, meaning that coalitions that need to be formed to govern would involve fewer parties, and would be (in theory) more stable.

However, there is the question of whether having a threshold is fundamentally undemocratic. Senior colleague and outgoing IRR CEO Frans Cronje dismissed thresholds as illiberal when I raised the question in a meeting this week. And perhaps he has a point.

South Africa has an implicit electoral threshold – but an explicit one may be a leap too far, especially given our country’s history of political exclusion.

In addition, thresholds can be misused and can also lead to large numbers of wasted votes, something which proportional representation is, in theory, designed to prevent.

For example, Turkey has a threshold of 10%, primarily to keep out parties which represent that country’s Kurdish minority.

In an election held in Ukraine in the early part of this century, over 20% of voters were effectively disenfranchised when a large number of parties failed to meet the 3% threshold. Thresholds can also see parties which fail to win an outright majority of votes manage to win an outright majority of seats, again something proportional representation systems are designed to prevent.

The implementation of electoral thresholds is unlikely in the near future in South Africa. However, when unwieldy coalitions are necessary to govern South Africa (a day that is not far off) the outsized influence that small parties are likely to wield over these coalitions may mean that the idea of thresholds could start looking pretty attractive to voters.

[Image: Jonathan Beech,]

If you like what you have just read, support the Daily Friend


Marius Roodt is currently deputy editor of the Daily Friend and also consults on IRR campaigns. This is his second stint at the Institute, having returned after spells working at the Centre for Development and Enterprise and a Johannesburg-based management consultancy. He has also previously worked as a journalist, an analyst for a number of foreign governments, and spent most of 2005 and 2006 driving a scooter around London. Roodt holds an honours degree from the Rand Afrikaans University (now the University of Johannesburg) and an MA in Political Studies from the University of the Witwatersrand.