In New Nation Movement v President of the Republic of South Africa and Others, the Constitutional Court ruled that independent candidates have a right to participate in elections at a national and provincial level, but did not venture into which electoral system better affords electorate accountability. It is up to the National Assembly to redesign the electoral system by June 2022 to accommodate the election of independent candidates.
Surveys of public opinion confirm a pervasive level of cynicism and mistrust in politicians, their motives and their general failure to deliver on election promises. Accountability of elected representatives remains the Achilles heel and a major stumbling block towards substantive democracy in governance systems. One-party dominance of the ANC and weak opposition have resulted in voter apathy and even withdrawal among some sections of the electorate, who prefer not to vote at all.
These difficulties emanate partly from the current electoral system and partly from the calibre of officials holding public office. As Professor Susan Booysen observed, political parties are the fly in the accountability ointment.
The pure proportional electoral system with closed lists, which is currently in force, has major disadvantages. First, the direct link between eligible citizens running for office and being elected by voters to represent them is severed. Voters are expected to vote for a particular party and the seats are allocated to the parties in proportion to the votes polled by them. The electoral system is therefore party- and not candidate-orientated. Thus, the focus shifts to representation of political parties (indirect representation) instead of representation of citizens as constituent power of the representative bodies (popular sovereignty) as is required by section 42(3) of the Constitution.
Secondly, a pure proportional system tends to be impersonal since most candidates on party lists would be faceless as far as the average voter is concerned. Such a system can easily be manipulated by party elites to create a system of political patronage that wipes out democracy at base within political parties.
Thirdly, members of Parliament in a pure proportional system cannot be held accountable by the electorate in a personal capacity. Voters do not have the power to vote corrupt politicians or party officials with little popular support out of power. Parliamentarians derive their position not from any necessary connection with the electorate but from their status within the party. The crucial thing is not knocking on doors, fixing the problems of constituencies, or even just representing concerns of people living in any identifiable geographic region. The voting for parties without the option to select specific candidates thus erodes the principle of democratic representation.
Parliament therefore faces several difficulties to achieve better responsiveness to popular will with electoral reforms. The principal challenge for the legislature in redesigning the electoral system will be to give effect to the right of voters to select candidates of their choice to represent them in the national and provincial legislatures while doing justice to proportional representation of political parties.
Parameters and logistics of a new electoral system
The sheer number of candidates that were nominated in the previous election makes it impractical to have an electoral list with hundreds of independent candidates running for office in a single huge constituency alongside those nominated by parties.
In the 2019 elections a record number of 48 parties registered candidates for the national parliamentary election. This is 19 more parties than contested the 2014 national elections. The number of candidates running for office only in the National Assembly amounted to 3,535. This development already strained the manageability of a ballot to the extreme.
Thus, given the parameters and logistics, a major electoral reform is inevitable. The legislature will hardly be able to get around a mixed electoral system, which is constituency-based to provide for direct election of candidates combined with a compensating proportional list system. Two previous studies into electoral reform have already recommended that – the Van Zyl Slabbert Electoral Task Team (2003) and the Independent Panel of Assessment of Parliament (2009).
Proposals for multimember constituencies
The Inclusive Society Institute is promoting a multimember constituency system that converges in some respects with the local government electoral system. The proposals are similar to the Van Zyl Slabbert proposals of 2003, except that 66 instead of 69 constituencies are proposed.
It foresees that 300 of the 400 members of Parliament will be elected to multimember constituencies, with between three and seven MPs for each constituency. The MPs for each constituency will be divided proportionally between the parties or individuals. The remaining 100 MPs will be elected on a proportional list basis, to ensure that the electoral system results, in general, in proportional representation.
These proposals have several disadvantages. As Susan Booysen also pointed out, these proposals have no solution to one of the problems of political parties: it is still the parties that will put forward their lists of candidates and thus there will still be no unmediated link between community and representative/s.
The 100 list candidates to ensure proportional representation would be in compliance with sections 46(1)(d) and 105(1)(d) of the Constitution, but the difficulty is that the idea of multimember constituencies gives proportionality a double whammy at the cost of individual political rights guaranteed by section 19.
Section 19 of the Constitution guarantees the right of every adult citizen to make political choices, to stand for public office and to vote for an elective body. It will therefore be hard to justify why voters may select a candidate of their choice only when such a candidate runs as an independent but not when a candidate elects to run on a party ticket.
A further difficulty is that it is not clear how the multi-member seats in constituencies will be allocated to the competing candidates. Can one party win more than one seat per constituency and which candidates will win if voters cannot select their candidates directly to determine who got the most votes? How will the quotas be determined to ensure that candidates running for office in a constituency with three representatives are not subject to different electoral parameters than those where seven candidates can be elected?
To put it differently, the weighting of votes of registered voters in different constituencies will not be equal. This model would probably not be able to take the hurdles of sections 9 (equal treatment of candidates) and 19(3) (individual rights to make political choices) read with section 3(2)(a) of the Constitution. The latter provision stipulates that all citizens are equally entitled to the rights, privileges and benefits of citizens and the section 19 rights are rights of individual adult citizens, not political parties.
In 2013 the Democratic Alliance submitted an Electoral Reform Bill with similar proposals, except that the number of representatives per constituency would be the same. The bill proposed the establishment of 100 three-member constituencies, each with about the same number of voters. In practical terms, political parties would submit a ranked list of five names to the IEC for inclusion in the constituency contest. These were to be closed lists – not open lists where voters could select the candidates of their choice. The three members who obtain the requisite quota of votes or largest surpluses would then be elected as the members of Parliament for that constituency.
The bill contemplated a further 100 members of the National Assembly being elected from national lists submitted by the various parties. A second ballot paper would contain only the list of parties contesting the election and voters could vote for the party of their choice.
If parties failed to have sufficient concentration in particular geographic areas to ensure equitable representation in the National Assembly, it would be corrected by seat allocation from the party list. Thus, once the constituency representatives have been elected, the chief electoral officer would calculate the number of seats in the National Assembly to which each party is entitled based on the proportion of total votes they obtained in the constituency-based election, and would allocate to parties seats from the lists so that the overall composition of the National Assembly reflects, as closely as practical, the proportion of votes obtained by each party.
The DA’s proposal that candidates should be elected on the basis of a closed list of five candidates for each party where the party determines the order of preference faces two difficulties: first, the right of individual candidates to stand for office might be unconstitutionally ousted; and second, the ballot might become overly convoluted should the trend persist of 48 parties participating, with each of them putting up five candidates for election next to all the potential independent candidates. This would hardly enable clear-cut and viable choices for voters or make it possible for them to hold a representative accountable.
The single-member constituency option
If the ballots should still be manageable, it might therefore be preferable to have 300 constituencies where party candidates and independent candidates contest elections in single-member constituencies with a second ballot paper where voters could vote for a party of their choice for purposes of proportional representation that are then allocated to 100 list-candidates to ensure proportional representation of political parties.
Such a personalised election campaign might be a very effective mechanism to weed out corrupt politicians. If a voter chooses a constituency candidate on strength of character, it is unlikely that the same voter would vote blindly on the proportional representation ballot. Unless the party has an equally credible list of leaders, it is not likely to get the proportional representation ballot.
A threshold to reduce splitter parties
There is another factor that should be taken into account. When there is no threshold for political parties to enter Parliament, proportional representation could give rise to a proliferation of small parties or splinter groups without any real political bargaining power in Parliament.
Although proportional electoral systems usually result in three or more relatively strong political parties, this process is evolving very slowly in South Africa. After 25 years elections still reflect the fault lines of a deeply divided society. The closed-list pure proportional electoral system might even have enhanced this state of affairs.
South Africa faces a real problem with the proliferation of small parties. Currently 323 parties are registered at a national level. In the 2019 election, the ANC got 62.2%, the DA 22.2% and the EFF 6.4% of the votes for the National Assembly. All other parties together only got 5.3% of the votes.
Many tiny parties are really vehicles for one person to get elected. The option to run as independent candidates might therefore have the effect of reducing the number of small parties, but one cannot assume that it will be the case.
If the ballots should still be manageable, the number of parties participating in elections need to be reduced to those who have actual chances to make it into Parliament and have potentially enough bargaining power to form stable coalition governments. It would therefore be sensible to introduce a threshold for political parties to enter Parliament. A threshold of 2% to 3% would probably suffice.
Nomination of candidates
The legislature also needs to conceive of an entrance threshold to avoid a proliferation of independent candidates and/or candidates on a party ticket to be elected directly, otherwise ballots may become impossible to manage.
There are various ways to achieve that: one option is to require that all candidates for direct election must pay an electoral deposit to make sure that they are serious about running for office since the deposit will be lost if the candidate is not elected. In all probability, it will not be reasonable and justifiable in an open and democratic society based on equality and freedom in terms of section 36 of the Constitution to differentiate between candidates standing for office on a party ticket and those running as independent candidates. This implies that a deposit paid by a party would only be in relation to list candidates.
Obviously, such deposits payable by individual candidates will have to be lower compared with those payable by political parties on behalf of their candidates on the proportional representation lists. This is so because a less restrictive means would achieve the same purpose in terms of section 36(1)(e) of the limitation clause. Thus, a deposit per candidate of between R10,000 and R15,000 might still qualify as reasonable.
To prescribe the same electoral participation hurdle to independent candidates and candidates running for direct election on a party ticket will have the additional advantage that parties will have to consider carefully which candidates they put up for election and whether they have a viable chance of being elected because the candidate might risk losing the deposit.
Unfortunately, many politicians no longer enter politics with the intention of public service, but with the intention of getting rich. The result is that political battles are a kind of proxy for deciding not how socioeconomic issues are to be addressed, but which faction running on the slate of a lobby group will gain the ability to insert itself into the circulation of money streams.
Another option, if a candidate has substantial support in a community but is too poor to afford the deposit, is to require that such a candidate must be nominated by a reasonable number of registered voters in that constituency, for example 1,000 voters. It will surely enhance inclusiveness of the poor and destitute and make it possible to give them a voice in Parliament. In the past, party membership has often been faked with false identities. The legislature will therefore have to regulate this carefully to curtail potential fraud on such supporting lists.
Ideally, the two options could be combined as alternatives.
Section 46(1) of the Constitution limits the number of seats in the National Assembly to no fewer than 350 and no more than 400. Section 105(2) of the Constitution also limits the members of provincial legislatures to between 30 and 80 members and is to be determined according to a formula prescribed by national legislation.
The largest remainder method and the Droop quota are currently used to allocate seats at both the provincial and national level, with the national list seats allocated by subtracting seats won at the provincial level from a party’s allocated total seats to give a more proportional result. The calculation of quotas according to this method could be retained, but a constituency-based electoral system would make the provincial party lists for election to the National Assembly redundant.
Since the electoral system must now also make provision for the election of independent candidates, quotas for seat allocation will have to be invoked to first allocate seats to directly elected candidates by dividing the number of votes by 300 (if that should be the number of directly elected members).
Thereafter, the number of votes for independent candidates will have to be deducted from the votes for party candidates to arrive at a quota for compensatory seat allocation to ensure proportional representation of political parties. Having done that, the Electoral Commission will have to determine which parties took the hurdle for entrance into Parliament. The remaining 100 seats would then be allocated to these parties according to their proportional share of the vote.
It would help not to merge elections at a national and provincial level. The issues at stake in these elections differ vastly anyway. The merging of the elections has had the effect in the past that national issues completely subsume the provincial ones in election campaigns, exerting a down-ballot pressure. In fact, this practice might actually be unconstitutional because it distorts elections at a provincial level. DM
This is an extract from a peer-reviewed article that appeared in the South African Law Journal. It can be accessed online free of charge.