The Constitutional Court and the Electoral Act
23 June 2020
HISTORICAL BACKGROUND
The framework for the 1994 election was specified in Schedule 2 of the Constitution of the Republic of South Africa Act, 200 of 1993 (the ‘interim constitution’). This schedule established the closed party list proportional representation system which has persisted until the present.
When the interim constitution was replaced by the Constitution of the Republic of South Africa Act, 108 of 1996, Schedule 6 set out transitional arrangements. Section 6 of the Schedule dealt with elections of the National Assembly and specified that the first election held after the passing of the 1996 Constitution (i.e. the 1999 election) would be held in terms of Schedule 2 of the interim constitution.
The framework for elections in the long term is set out in Section 46 of the 1996 Constitution:
1. The National Assembly consists of no fewer than 350 and no more than 400 women and men elected as members in terms of an electoral system that –
a. is prescribed by national legislation;
b. is based on the national common voters roll;
c. provides for a minimum voting age of 18 years; and
d. results, in general, in proportional representation.
2. An Act of Parliament must provide a formula for determining the number of members of the National Assembly.
Accordingly, the Electoral Act, 73 of 1998 was passed. It is accompanied by the Electoral Commission Act, 51 of 1996 and regulations made under both Acts. These acts and regulations constitute the law on elections[1].
It was constitutionally possible to reconsider the electoral system after the 1999 election. On 20 March 2002, the Cabinet resolved to establish an Electoral Task Team (ETT) to draft new electoral legislation. The ETT was to consult stakeholders, including political parties. It was chaired by Dr van Zyl Slabbert. The ETT submitted its report in January 2003. The Team reported both a majority and a minority view. The majority view was that the country should be divided into multi-member constituencies. The number of representatives to be elected in such a constituency would vary, depending on the number of voters, from three to seven for a national election, and 300 of the 400 members of the National
Assembly would be elected from closed constituency lists in this way. A further 100 representatives would be allocated from closed national lists in order to restore overall proportionality. The minority view was that the system should remain unchanged. The government never responded to the report, implicitly adopting the minority view.
Unease with this outcome has persisted. The government has introduced a system of constituency offices and members of the National Assembly are assigned to them. Members of the public may approach these offices for assistance and to make representations to them. But there are no constituencies to accompany the offices and therefore no accountability of MPs to defined groups of people on whom they depend for continuing support.
THE CONSTITUTIONAL COURT RULING
It is against this background that a judgement handed down by the Constitutional Court on 11 June 2020 has to be understood. The applicants in the case successfully sought to remove the Electoral Act restriction that candidates for election to the National Assembly and provincial legislatures must do so as members of political parties. In its judgment, the Constitutional Court declared that:
The Electoral Act 73 of 1998 is unconstitutional to the extent that it requires that adult citizens may be elected to the National Assembly and Provincial Legislatures only through their membership of political parties[2].
The Electoral Act will have to be amended, and the Constitutional Court has given Parliament 24 months to do so[3].
How could the Act be amended? This brief considers two possibilities. For ease of exposition, only national elections will be considered here, but the argument will apply mutatis mutandis to provincial elections.
THE SIMPLER OPTION
In this case, independent candidates could stand for the National Assembly. For the purpose of allocation of seats, each independent candidate would be treated as if he or she were a separate political party. All the existing rules about the allocation of seats would apply. The possible outcomes for each independent candidate would be the award of zero, one or more than one seat. Zero would mean no election. Otherwise the candidate would be elected. The question arises: what would happen if an independent candidate is awarded more than one seat.
Section 46(1)(d) of the Constitution supplies the principle. The candidate would be awarded more than one vote in the National Assembly. An independent candidate winning 1% of the national vote would be entitled to four votes as a member of the National Assembly. Such a candidate would contract the size of the National Assembly by three[4]. There would then be a distinction between members of the National Assembly, which could be less than 400, and votes in the National Assembly which would remain at 400[5]. Section 47(3)(c) of the Constitution[6] would not apply to independent MPs, since no party had nominated them. In terms of membership of committees, an independent would be treated as if he or she were a party with the same number of votes, implying that an independent with more than one vote would have more committee work on average than an MP who is a member of the party. All these details would have to be taken care of, but for the foreseeable future, it is unlikely that any independent MP would command more than one vote.
THE BETTER OPTION
Since the Electoral Act has to be revised, it would be appropriate for consideration to be given to Recommendation 6.8 in the Report of the High Level Panel on the Assessment of Key Legislation and the Acceleration of Fundamental Change published in November 2017.
The Panel recommends that Parliament should amend the Electoral Act to provide for an electoral system that makes Members of Parliament accountable to defined constituencies on a proportional representation and constituency system for national elections.
Essentially, this advocates the adoption of the ETT majority recommendation. The option presented here is in line with the principles in the majority recommendation, but takes into account developments since the report was written. The outline of the better option is presented first, and its justification follows.
Each voter would be issued with two ballot papers. The first would require a choice between candidates in his or her constituency and the second would require a choice between parties contesting the election. The first vote represents voter preferences about representation by individuals. The second represents voter preferences about representation by parties.
Seats in the National Assembly would be divided into two categories: those filled from constituency elections and those filled from party lists.
The number of the party list seats would be twice the number of constituencies and the number of the constituency seats would be 400 minus the number of party list seats. For example, if there were 55 (multi-member) constituencies, there would be 110 party list seats and 290 constituency seats.
Each constituency would be allocated a number of seats proportional to the number of registered voters in it. Constituency candidates would be ordered by the number of votes received and seats allocated going down the list until all constituency seats are filled.
Constituencies should follow metro and district authority boundaries as far as possible to avoid delimitation disputes. Some sparsely populated districts would need to be combined into single constituencies and the five large metros would need subdivision.
Each party contesting the election would draw up an ordered list of party list candidates, as they do at present, up to a maximum of the number of party list seats.
Parties would be represented in proportion to the votes cast for them on the second ballot paper. The number of seats party representatives would occupy would be 400 minus the number of independents elected on a constituency basis. Thus if four independent candidates were elected and a party received 50% of the second ballot vote, it would be allocated 198 seats in total.
The number of party list seats allocated would be the total allocation of seats minus the number of constituency seats won by members of that party. Thus, if the party in point 7 won 150 constituency seats, it would be entitled to fill a further 48 seats from its party list.
Most members would be elected by defined groups of constituents and would be dependent on continuing constituent support. In addition, two party list MPs could be assigned to each constituency, using an algorithm designed to maximize the probability that voters have at least one of the MPs associated with their constituency belonging to the party they support.
It should be stressed that the essential purpose of the party or second vote is to determine the overall number of seats a party receives in parliament. It does not change the results of election of constituency representatives, but it is able to rectify any imbalance that the constituency votes may have caused to the proportionality in the aggregate number of seats. It therefore plays the role of a balancing mechanism, to ensure proportionality in parliament. For example, if a party receives a significant percentage of votes in the constituency elections, but does not win any seats in those constituencies, the second vote will nevertheless provide it with the number of seats to which it is entitled from its proportion of second votes.
This balancing mechanism means that the precise boundaries of constituencies are not important, as the final composition of parliament will reflect the proportion of total votes each party receives on an aggregate basis. There would be no point to United States-style gerrymandering.
CONCLUSION: WHY IS THE BETTER OPTION BETTER?
It is sometimes thought that there is a simple choice between electoral systems: a single member constituency system, with first past the post winners, and a national proportional representation system without constituencies. The single member constituency system works well enough when there are two parties competing for the political centre, with alternations in power. But it gives seriously distorted results when these conditions are not met, with far from proportional representation in parliament. If these were the only alternatives, proportional representation, as enshrined in Section 46(1)(d) of the Constitution is appropriate to South African circumstances.
But there is a way of having both the advantages of proportional representation and accountability to constituencies. This is what the ETT majority and the better option propose. It essentially asks voters to make two decisions: one about the individuals they would like to represent them and one about the party composition of the National Assembly. As is currently the case in municipal elections, split voting would become possible. Judgments about individuals are not necessarily the same as judgments about the desirable representation of parties in parliament.
The better option meets the proportionality requirement of Section 46(1)(d) of the Constitution.
The better option changes incentives in a desirable direction. A successful parliamentary career would depend not only on party relationships, but also on the cultivation of a constituency and effective representation of its interests. Report-backs and consultation would be essential. Parties would still select their constituency candidates, but they would have to consider their potential appeal to the relevant constituencies. The diversity of place would be better represented in the National Assembly, and accountability to the electorate would be strengthened, deepening democratic legitimacy.
Charles Simkins, Head of Research, Helen Suzman Foundation.
[1] A compendium can be found at http://www.elections.org.za/content/Elections/Laws-and-Regulations-Elections, reflecting the position on 8 March 2019
[2] Case CCT 110/19 at 128[4]
[3] Case CCT 110/19 at 128[5]
[4] A problem would arise only if the number of National Assembly members contracted in aggregate by more than fifty. This unlikely event could be prevented by an all-round proportional increase in MPs designed to increase the total number of members to 350.
[5] In turn, this would require an appropriate interpretation of the Constitution where it deals with voting in the National Assembly.
[6]A person loses membership of the National Assembly if that person ceases to be a member of the party that nominated that person as a member of the Assembly.
