voters – not just by the majority of voters, or even by a majority of a minority. It does this by assigning seats in Parliament in direct proportion to the number of votes any party gets.
In fact, under South Africa’s proportional representation system, any party that wins as little as 0.25 per cent of the national vote will get at least one seat in Parliament. (In the 2014 election, this meant that a party needed only about 45 000 votes to get a single seat.) This is the lowest threshold possible under any electoral framework.9 The system’s inclusivity contrasts starkly with the exclusionary outcomes in winner-take-all countries such as Britain or the United States, and ensures that no one’s vote is wasted.
In South Africa, if party A wins 40 per cent of the national vote, it gets 160 of the 400 seats in the National Assembly. If party B wins 35 per cent of the nationwide vote, it gets exactly 140 seats, while 25 per cent for party C would guarantee it 100 seats. Unlike winner-take-all, proportional representation turns elections into positive-sum games: one party’s victory does not automatically mean that other parties do not make it into Parliament (or into government). It is easy to see why this system is better suited to plural and diverse societies such as South Africa.
But the real power of the proportional representation model kicks in when no party wins more than 50 per cent of the national vote. Let’s assume that the ANC dips just below 50 per cent in the 2019 national election, while the Democratic Alliance (DA), the Economic Freedom Fighters (EFF), and other opposition parties marginally increasing their shares of the vote. For the first time in post-apartheid South Africa, no single party would be able to form a national government without the support of at least one other party.
This is because the crucial step in forming a government is when the National Assembly elects one of its members as president, who then proceeds to appoint a cabinet. For this to happen, someone needs the support of at least 201 of the 400 MPs. Since 1994, the election of the president has always been a mere formality, as the ANC never had fewer than 249 MPs. But once the party’s share of the popular vote drops below 50 per cent, the ANC will have less than the required 201 seats in the National Assembly. This is the moment when coalitions will move to the epicentre of national politics. When ANC support drops below 50 per cent, any combination of parties will be free to form a government as long as they are able to get 201 MPs to vote for their preferred presidential candidate.
Even if the ANC was still the largest party – even if it had 199 seats, and its closest competitor only had 120 – it would still be out of power if opposition parties managed to form a coalition that added together all of their 201 seats. However, the ANC would remain in power if it managed to convince a party with just two seats to join it in a coalition, and support its presidential candidate.
The rule is simple: whichever party’s candidate gets 201 or more votes becomes president. The president then appoints a deputy president, ministers, and deputy ministers to his or her cabinet, thereby formally assembling a government.
Once coalitions become the norm, seemingly minor political parties will suddenly begin to play tremendously important roles. In the above scenario, the ANC would need to convince a minority party with only two Parliamentary seats (equal to only 0.5 per cent of the vote, or about 90 000 votes) to join it in a coalition. This means that the two parties would need to strike a deal. Based on experiences in other countries governed by coalitions, such a deal usually entails that the junior partner gets a few important portfolios, like foreign affairs or even the deputy presidency. The senior partner will generally also have to agree to some policy concessions to convince the smaller party to join it. This means that parties representing minority interests, such as the African Christian Democratic Party (ACDP) or the Freedom Front Plus (FF+), could soon play an important role in governing South Africa.
But there is another potential outcome. If no group of parties in the National Assembly is able to cobble together a formal alliance that would take them over the 201-seat mark, the country could get a so-called minority government. Similar to what happened in Johannesburg and Tshwane after the 2016 municipal election, a group of parties may agree to support the same candidate for president without constituting a formal coalition.
The proportional representation system therefore precisely reflects the will of the voters, even if their will is that no one party should have enough power to form a government on its own. In a plural society like ours, proportional representation will soon lead to a fractured political landscape in which it will be very difficult for any one party to get consistent majorities.
Defying destiny
But this prediction also raises an obvious question: if South Africa’s electoral framework encourages coalitions, why has the ANC been able to rule on its own since 1994? Indeed, in countries with proportional representation systems, coalition governments are the rule, and one-party majorities the rare exceptions. This makes the ANC’s two decades of domination even more impressive, as its majorities were not artificially inflated by the vagaries of winner-take-all. Instead, the opposite happened – the ANC achieved its majorities in a system explicitly designed to encourage plurality and coalition governance.
The ANC’s dominance since 1994 is a powerful illustration that although rules and structures favour certain outcomes over others (in this case, coalition rather than majoritarian government), they are never enough to guarantee that those outcomes will materialise. Although the rules of our electoral system favour multiparty coalitions over single-party majorities, democracy means that those rules are not absolute. They can be overridden wherever voters strongly prefer a single party.
This is precisely what happened during South Africa’s first two-and-a-half decades of democracy, as the ANC’s liberation narrative proved powerful enough to overcome the logic of the electoral system. The ANC was so overwhelmingly popular that it was able to engineer single-party rule in an environment that actively discouraged it.
But once the power of the ANC’s liberation narrative is sufficiently weakened, and its mighty majorities vanish, South African politics will be transformed. This will be no accident, as the proportional representation of every cultural, ethnic and political group, as well as forced coalitions among them, is precisely what constitutional framers from opposite ends of the political spectrum envisioned when they wrote the rules in the mid-1990s.
The ANC’s ability to overcome the tendency towards coalitions meant that citizens wrongly became used to the idea that South African democracy entails clear and unbending distinctions between one powerful governing party and an amalgam of minor opposition parties nipping at the ANC’s heels. One indication of this perception is audible whenever someone speaks of the ANC as a ‘ruling’ party.
This term has authoritarian overtones; it may be appropriate to describe a dictator as a ‘ruler’, but a democratically elected government does not rule over its citizens. It governs on their behalf. That ‘ruling party’ instead of ‘governing party’ has become the default way to describe the ANC in public discussions says a lot about South Africans’ perception of the party as an all-powerful entity capable of enforcing its will. The widespread use of the authoritarian-sounding ‘ruling party’ is not a trivial detail: it shows how South Africans have come to regard the exceptional situation of single-party dominance as the norm.
This perception of the ANC will soon come crashing down. While the party has thus far used its powerful history to thwart the electoral system’s tendency towards coalitions, its fast declining electoral fortunes means this domination is in its death throes. In its place, coalition governments will rise.
Roadmap to the future
This book’s analysis of our coalition future proceeds from three assumptions. The first is that the trend outlined in Figure 1 is set to continue – in other words, that the ANC’s electoral decline will not be arrested in the short term. This means it will become increasingly difficult for the party to garner majorities in metropolitan municipalities like Tshwane, Johannesburg, Ekurhuleni and Nelson Mandela Bay, as well as in larger rural centres like Rustenburg and Polokwane. The result is that, at the municipal level, the ANC will increasingly resemble a rural-based party, while urban competitors like the DA and EFF will continue to make inroads in the cities.
This bottom-up process