IRR paper sets out solution for electoral reform quandary
7 November 2022
The IRR is proud to publish today a report by respected independent elections analyst Mike Atkins titled The Road to Electoral Reform.
The paper takes an in-depth look at how South Africa has reached a moment that carries the risk of a democratic and constitutional crisis, the country having narrowly avoided such a crisis in 2021.
The paper takes an in-depth look at how South Africa has come to face this moment that contains within it the makings of a democratic and constitutional crisis, having narrowly avoided such in 2021.
Says Hermann Pretorius, IRR director of communications: “In a constitutional democracy, the health and integrity of the electoral systems cannot be overstated. These systems, at local, provincial, and national level, are the foundations of what it means to be able to have a voice and a vote – a say in the future of our country.
“The IRR is proud to have the opportunity to work with Mike Atkins, a highly respected independent elections analyst. In recent years, especially since the Constitutional Court’s judgment in the New Nation case, it has become clear that a constructive, non-partisan contribution to the debate regarding electoral reform is greatly necessary. The paper published today by the IRR is such a contribution.
“The Road to Electoral Reform sets out the current state of electoral reform in South Africa, and shows the significant shortcomings of the Electoral Amendment Bill, only recently approved by the National Assembly, to be something close to a constitutional disaster. The decisions of various parties and officials are shown to have contributed to the dangerous situation our democracy finds itself in. Finally, and crucially, an alternative legislative foundation is set out that will ensure the health of our constitutional democracy.
“We urge all stakeholders to engage with this paper, with us at the IRR and with Mike Atkins. Electoral reform is surely the greatest opportunity for collaboration across all political and ideological boundaries and differences. We cannot allow our democracy to be hijacked or captured or for us to accidentally stumble into a system that simply will keep power in the hands of the powerful.”
Access The Road to Electoral Reform can be read here or below:
Issued by Hermann Pretorius, IRR director of communications, 7 November 2022
THE ROAD TO ELECTORAL REFORM
Michael Atkins
November 2022
The Road Ahead for Electoral Reform in South Africa
Introduction
With the passage of the Electoral Amendment Bill through Parliament, it is necessary to examine the question of electoral reform in greater depth. We need to understand where we are with the Bill and understand how we got here. Somehow, we need to chart a course through the 2024 national and provincial elections, and then we need to go back to square one and start a rational and meaningful process of electoral reform.
The Electoral Amendment Bill is premised on the logically absurd inclusion of individuals in proportional representation elections, alongside party lists. While there are partial precedents for this, they do not translate at all into our electoral and constitutional framework. All of the complications with the Bill, all of the contradictions, and all of the difficulty in conducting meaningful discussions, have arisen from us ignoring the underlying absurdity.
Unsurprisingly, the logical contradictions translate to outcomes that are unfair either to independent candidates, or to voters. Any such unfairness, or any of the inevitable numerical distortions, render the Bill unconstitutional. There is no other logical outcome. And yet, from the moment the particular electoral system was proposed (in its' original form) within the Ministerial Advisory Committee (MAC), the incipient problems have existed.
Under the injunction of the Constitutional Court, Parliament is pressing ahead, and we are caught, as in a vice-grip, between the absurdity of the Bill, and the imperative of the court, and the holding of the 2024 election. As the Bill is processed by the National Council of Provinces, and is then likely signed by the President, we have to take stock and consider how we avert the implementation of a confusing and unfair electoral system, how we conduct the 2024 elections, and then how we engage in valid and meaningful electoral reform.
We can't merely adopt a wait-and-see approach. As difficult as it is, there is still time to do these things, and still conduct a valid and fair election in 2024 (albeit not necessarily an entirely constitutionally valid one). But there are several tripping points on the road, and it would not take much for us to fall into a trap of our own making. In 2021, we narrowly avoided an electoral and constitutional meltdown. It is inconceivable that we should put ourselves in the frame for the same type of panic again.
But even as we face these electoral and constitutional storms, we also have before us the prospect of real change, and of meaningful electoral reform that many have been crying out for. A good electoral system does not guarantee good outcomes for a country, but a bad electoral system almost certainly does guarantee bad outcomes. It is therefore necessary to look beyond the immediate constitutional crisis (for this is what it is) and consider how we go about the process of electoral reform, and what choices we have available to us.
The Absurd Premise of the Electoral Amendment Bill
It is evident from the debate in the National Assembly, and from several other public statements, that many MPs do not have a good understanding of the actual effects of the Bill, or why civil society has the temerity to oppose what Parliament has decided. It is almost as though the court leaving the "details" of how to remedy the constitutional defect of the Electoral Act to Parliament gives carte-blanche to create any system that MPs feel like. Of course, the irony is missed, that we are here precisely because civil society took to the court a complaint about the Electoral Act previously passed by Parliament. In that case, the court concurred, and required a change.
Put simply, the Bill is premised on the idea that individuals can be included alongside political parties on the existing proportional representation ballots. The MAC’s first option describes itself as being, "the retention of the [existing] electoral system", where "independent candidates should be included".
This is a breathtakingly absurd proposition that has attracted little or no comment. Having individuals on a proportional representation ballot is a contradiction in terms precisely because an individual is limited to occupying a single seat, irrespective of the number of votes obtained. There is no polite way to express this.
From the outset, there was never any other possible outcome, except that this results in numeric distortions that make a mockery of the constitutional requirement that our electoral system results, in general, in proportional representation. And if the patterns of results arising from seat allocation calculations are thus distorted, then it is also logically inevitable that the election cannot be fair.
Logically, it is false to describe the system as the “retention" of the existing system. As soon as some participants in the election cannot be "rewarded" in proportion to the votes received, this becomes a new system, where some mechanism must be found to deal with either the votes that cannot be used, or the excess seats that cannot be awarded to independent candidates.
New Nation Movement NPC, and How We Came to Electoral Reform
The question of electoral reform has been on the table for a long time but has largely been ignored. Proportional representation was seen during our transition to democracy as the fairest system to give effect to the democratic aspirations of a diverse society. The pure proportional representation that we have used in national and provincial elections was also the simplest and safest system to implement in 1994, with very tight time frames and significant logistical constraints, in a fraught context.
It was always held that a more nuanced and more directly representative system should be introduced once our democratic processes were safely entrenched. The Electoral Task Team headed by Frederik van Zyl Slabbert recommended in 2003 that a constituency system be implemented, with 69 multi-member constituencies electing 300 members of the National Assembly. No further action was taken with respect to this report. At the ANC's 2007 National Conference, a Resolution was passed that we should consider moving to a constituency system (balanced as always by proportional representation). In 2017, the High-Level Panel on Key Legislation, headed by Kgalema Motlanthe, proposed among other things that we should move to an electoral system based on a "proportional representation and constituency system for national elections".
It took an unusual and unexpected cause to reignite the process of electoral reform in South Africa. Chantal Revell, the New Nation Movement, and others brought a case to the courts, arguing that our electoral system is unconstitutional to the extent that it does not provide for individuals to contest national and provincial elections, except by virtue of being members of and candidates for political parties.
This was a departure from what had gone before and seemed to take everyone by surprise. Having failed in the High Court, the applicants were granted access to appeal to the Constitutional Court. The Society for the Advancement of the South African Constitution (CASAC) and the Organisation Undoing Tax Abuse (OUTA) were admitted as Amicus Curiae. The case was heard in August 2019, and the ruling was issued on June 10, 2020. The arguments were nuanced, taking into account the rights to freedom of political activity, freedom of association, and the right to stand for election. The 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". The operation of the order was suspended for 24 months to "afford Parliament an opportunity to remedy the defect giving rise to the unconstitutionality".
In the judgment, the court was at pains to point out that the prerogative lies with Parliament as to how the defect was to be cured. Herein lie the seeds of some of the problems we now face. Clearly, the court is correct in law in that it may not prescribe to Parliament in any manner save for ensuring that legislation already passed is consistent with the Constitution. For example, if an Act is deemed to be inconsistent with the Constitution, then the inconsistency can hypothetically be resolved by altering the Constitution (albeit that this would require a two-thirds parliamentary majority and would likely come at a high political cost).
What was not considered (seemingly by anyone) is the question of proportionality. Quite simply, as is explained below, it is not possible to include individuals (independent candidates) in a proportional representation system without having some form of constituency election, balanced by a separate proportional representation system. It is a moot point as to whether the question of requiring constituencies would have had any logical impact on the arguments before, and deliberations of the court. However, it most certainly does have an impact on what follows.
The Road to Drafting the Electoral Amendment Bill
It is important for us to understand something of the process that followed the court ruling, and how this has contributed to the legal quagmire that we are entering.
The parliamentary Portfolio Committee on Home Affairs in meetings held on June 25, 2020, July 21, 2020, and again on August 18, 2020, considered the implications and impact of the Constitutional Court ruling, and discussed in detail the road ahead. A significant amount of work was done at this stage by the Independent Electoral Commission and by parliamentary support staff to set out the implications of the judgment, and what the required actions from Parliament were. The IEC gave a presentation to the Committee in the July 21, 2020 meeting describing the principles underlying different electoral systems.
Of particular note is that in the meeting of August 18, 2020, very detailed timelines were presented to the Committee and to the Minister of Home Affairs for compliance with the court ruling, and for ensuring that implementation of a new system would be possible by the 2024 national and provincial elections. The key fact of this analysis is that for a constituency system to be implemented timeously, work on drafting would need to proceed during the month of September 2020.
As it happened, while there were twelve further Home Affairs Portfolio Committee meetings during the remainder of 2020, none addressed the question of the New Nation judgment, or changes to our electoral system. It must be taken into consideration, though, that the Home Affairs Committee was occupied with the Electoral Laws Amendment Bill, introduced before the court ruling, that contained a number of changes that had to be made in time for the 20201 Local Government Election.
On February 9, 2021, the Minister of Home Affairs gave a presentation to the Portfolio Committee on Home Affairs, stating that the department had studied a variety of electoral systems, and had drafted a Report that was to go to a cabinet Committee, and then to the Cabinet for approval. At the time, the Committee resolved to "give the Executive time to reflect on the amendment of the Electoral Act and to table its proposals". However, also in February 2021, the Minister constituted the Ministerial Advisory Committee to "develop policy options on the electoral system that address the defects of the Electoral Act" and to "recommend possible options to be considered in the South African setting".
The Home Affairs Committee went on to hold a joint one-day "Electoral Reform Workshop" on March 16, together with the National Council of Provinces Select Committee on Security and Justice, where presentations concerning different electoral systems were made. At this workshop, IEC Commissioner (now Chair) Mosotho Moepya spoke about the timelines for electoral systems either with, or without constituencies.
If constituencies were to be implemented, Parliament would have had to decide by October 2021 on the system to be implemented. this was to allow for changes to the Municipal Demarcation Board, the demarcation of constituencies, and for the preparation of the IEC's own systems to reflect the new electoral system, and the newly demarcated constituencies.
The MAC carried out its work during March and April 2021 and presented their Report to the Minister on June 9, 2021. The MAC could not agree on a single proposal, and the Report presented two options. The so-called "Majority" option was a hybrid constituency (first-past-the- post) and proportional representation model, referred to as a mixed-member proportional representation system. This system would have 200 constituencies for the National Assembly and would be structured in a manner similar to our current local government elections. The "Minority" option was a simplistic insertion into the existing electoral system of independent candidates, on the same ballots at political parties, with a small caveat that independent candidates would stand only within their province (referred to as a region).
In October 2021, the Minister commissioned Senior Counsel Steven Budlender and a team to draft the Electoral Amendment Bill, based on the MAC Option 1 (the "minimalist" or “minority” option). The MAC Report, the Bill, and its' explanatory Memorandum were presented to Cabinet for approval on November 24, 2021. The Bill was thereafter introduced to Parliament on January 10, 2022.
It does not help to speculate over motive, but by the time the Minister commissioned the drafting of the Bill, it was already too late to implement the constituency system proposed in the MAC Option 2 in time for the 2024 elections. In that sense, the so-called "minimalist" Option 1 was a convenient "way out". It is not clear why there should have been several deliberations and presentations on electoral systems, after the IEC did this in July 2020. In the joint electoral reform workshop held in March 2021, when discussing the deadline of October 2021 to decide on a system, there was no concern, or declaration that it was already too late to implement constituencies. We presume that, had action been taken directly after the MAC Report was presented in June 2020, it would still have been possible to implement constituencies within the required time periods. Certainly, had the Minister constituted the MAC in September 2020, and had it then reported before the end of the same year, it would have been possible to choose either of the two options with enough time for implementation ahead of the 2024 elections.
It makes no sense that the Minister should have allowed sixteen months to pass, with ongoing presentations, research, workshops, and public submissions, only to then turn round and choose a simplistic "inserting" of independent candidates into the existing system, seemingly because that very delay then meant that it was not feasible to implement a constituency system in time for the 2024 election. Even if the system chosen was rational and fair, this would be an entirely unsatisfactory process by which to determine how elections would function for years or decades to come.
The Electoral Amendment Bill Before Parliament
The Bill that was introduced to Parliament was intended to be a direct embodiment of the "minimalist" Option 1 of the MAC, which described itself as the retention of the existing system, with the insertion of independent candidates. In this sense, the original Bill was poorly drafted with respect to an important part of the seat allocation calculations. For the overall proportional representation calculations to determine the final seat distribution in the National Assembly, the term "compensatory seats" was wrongly interpreted to mean a separate calculation for 200 seats, as opposed to a calculation for all 400 seats, to balance out any disproportional outcomes in the regional seat calculations. Although was corrected later, following input from the IEC, the original error would have granted the ANC a bonus of 4 to 8 seats in any election (this was verified by testing the calculation on the past four national elections).
The original Bill proposed a single ballot for the National Assembly, and a single ballot for each provincial legislature. Independent candidates for the National Assembly could gain votes only from one region, with a vote threshold of double that applying to political parties. Had independent candidates been able to draw votes from the whole country, then we would have been faced with excessively long ballot papers, and with greater distortions of proportionality. In the original Bill, if an independent candidate failed to meet the vote threshold (quota), they were not allowed to "contest" seats allocated via remainders.
In terms of fairness, and of the numeric distortions, the Bill originally presented to Parliament was horrendous.
Following a period of written and oral submissions to the Portfolio Committee on Home Affairs, public hearings were conducted around the country. There were some complaints that these were not well advertised, although perhaps the most unhelpful characteristic was that those presiding over the meetings allowed multiple participants at each venue to read from the same script (in some places, the same statement was read more than twenty times).
Although the public hearings may have ticked the boxes required for public participation, there is one significant sense in which they were defective, given that this is about our electoral system. When electoral reform is undertaken in normal circumstances, the public has before them at least two choices to debate. typically, the public engages in a debate between maintaining the status quo and adopting a proposed new system. In our case, we have to change our electoral system, and we were presented with only a single option for doing this. We were "consulted" only on the details of how the new system could be implemented, rather than on the merits of the system itself.
From March to September, the Portfolio Committee on Home Affairs deliberated on the Bill, with a new round of written submissions being solicited during September. During that period, the Independent Electoral Commission (IEC) made several substantive proposals, correcting some of the worst faults in the Bill.
A particular change, crucial to the entire fiasco with the Bill, was presented will little fanfare, and almost no comment. In their presentation to the Committee on May 13, the IEC proposed that a second ballot be added for the National Assembly. The motivation initially was a vague granting to voters the “right to make political choices”.
In their presentation of June 20, the IEC gave the more substantive (and correct) motivation that this was required to meet the constitutional requirement of proportionality. Essentially, the inclusion of the second ballot created constituencies (multi-member, consisting of whole provinces) where independent candidates could stand, with a separate proportional representation ballot, where only parties may participate.
Structurally, this creates the possibility that the National Assembly can be constituted in a proportionally valid manner in terms of the Bill. The logical implication has been entirely ignored, however. If it is necessary to have two separate ballots in order for National Assembly elections to meet the constitutional requirement of proportionality (it is), then it follows that it must also be true that elections for provincial legislatures require separate constituency and proportional representation ballots. The deliberations in the Committee did not consider this implication at all.
During September and October, two contentious issues occupied the Committee. In the end, the Bill allows independent candidates to contest more than one region for the National Assembly, as well as the election for a single provincial legislature, without considering the perspective of voters in distinct elections all voting for the same person when they can only occupy a seat in a single legislature. The second major issue was the inclusion of a hard requirement for the number of signatures than an independent candidate must provide in order to confirm candidature.
This was reduced from a number equalling the original 50% of the Quota to 30% and finally, 20%. Of concern is that having received advice that 30% of the Quota might be unfair (and hence constitutionally invalid), the Committee adopted the figure of 20% without engaging in any discussion of how this would be substantively different from 30%.
In response to the public submissions in September, the IEC made a number of technical corrections to the Bill, and particularly to the wording to amend Schedule 1A of the Act, which contains the descriptions of how the various seat allocation calculations are carried out. Despite this, two clear logic faults remain. The "quota" referred to in the signatures requirement for independent candidates is generally understood to be the number of votes a party needs under the current system to obtain a seat in the National Assembly. However, in the calculations in the Bill, the quota is a figure nearly double that, being drawn from the calculations within the regions. As astonishing as it may seem, in terms of the Bill, the numbers of signatures required for an independent candidate to contest the National Assembly election is 90 percent higher than the MPs who approved the Bill think that it is.
The Bill was taken to the National Assembly on October 2020, in a sitting that was brought forward by a week for this particular purpose. It was supported by the ANC, the EFF, PAC, and NFP, and opposed by the DA, IFP, FF+, ACDP, and COPE, with GOOD and the AIC abstaining.
What was notable in the National Assembly debate is that there was almost nothing said in favour of the Bill. Mostly its proponents raised defences against criticism. Much was said about the public participation process, and the impression was created that the court did not give sufficient time for a more comprehensive reform of the electoral system.
Of interest is the contradiction in the comments made by the Minister, and by ANC MP, Brandon Pillay concerning the future prospects for future electoral reform. Minister Motsoaledi said "... there is also much criticism that the Bill will regulate ... the elections forever. The Constitution empowers Parliament ... to prescribe an electoral system. And as such, future discussions on this matter may still ensue." However, Brandon Pillay later said, "in an engagement with the Portfolio Committee, the MAC raised that the chosen electoral system should remain in place for the next few decades. Longetivity [sic] is therefore crucial.".
What is Actually Wrong With The Bill?
In the National Assembly debate on October 20, the Minister of Home Affairs, Aaron Motsoaledi attempted to deflect criticism of the Bill. He said that civil society wrongly implied the New Nation judgment required far-reaching electoral reform, rather than the mere "inclusion" of independent candidates. He also stated (without substantiation) that, "much of the criticism is levelled against earlier versions of the Bill". In fact, in the letter sent to MPs by several civil society shortly before the National Assembly vote, it was stated plainly that "you are being requested to consider and vote on a Bill that is unfair and unjust to independent candidates. It is also a Bill that will significantly disenfranchise voters".
The key is that any such criticism of legislation passed by Parliament must be decided on the merits – there is no fault in raising such criticism, even if it ultimately proves to be unfounded. The question, then, is what are the defects of the Electoral Amendment Bill?
The contradiction described above, of individuals being on the same proportional representation ballot as political parties, lies at the heart of the matter. Most of the particular faults are direct outcomes of this core anomaly.
The basic problem is that political parties will gain a seat share that is greater than their share of the votes. This is because, collectively, independent candidates will inherently gain somewhat more votes than their eventual share of the seats. In a proportional representation system, votes being discarded raises the share of the new total that the other parties have. As it happens, this is done in the Bill using a recalculation that is further biased in favour of the largest parties.
The following is a summary of those outcomes:
1. The Bill sets very high barriers to entry for independent candidates contesting the National Assembly election. The threshold of votes (quota) required by independents to gain a seat range from about 68 000 in the Northern Cape to about 92 000 in Gauteng (based on voting data from the 2019 national election). The effective number of votes required by political parties per seat is about 44 000 in terms of the 2019 results. The reason for this is that only 200 seats are contested by independent candidates, with the full complement of votes, inherently raising the quota.
2. The proportional representation calculations for seats in the National Assembly uses the regional ballots (with votes for independent candidates removed) added to the proportional representation ballots. The numerical distortions inherent in the regional ballots are then partially carried over into the PR calculations, effectively giving some bonus seats to the largest parties, at the expense of the smallest parties.
Although a similar calculation is carried out with the proportional representation calculations in local government elections, the other differences between the systems mean that this does not translate well to the National Assembly. In fact, there is a case to be made that the inclusion of Ward ballots at local government level in PR calculations is itself unfair, and possibly also unconstitutional.
3. The inclusion of independent candidates alongside political parties on a single ballot for provincial legislatures creates an inherent distortion to proportionality. The recalculation mechanism used to adjust seats has an in-built numerical bias that favours the largest party or parties.
It is an inescapable fact that, if an independent candidate gains votes equalling any multiple of the quota of votes for a seat, then the effect of each ballot cast in favour of that independent candidate is inversely distorted. Thus, if the candidate wins twice as many votes as required for a seat, then only half of each of those ballots contributes towards the election of the candidate, while the other half effectively contributes towards seats gained by the largest parties.
4. Because independent candidates are allowed to contest more than one regional election for the National Assembly, as well as a single province, then their voters in all but one of those elections are effectively disenfranchised. To spell it out, if a prominent candidate wins seats in more than one election and can obviously only take up only one of those seats, then the voters in the other elections have been stripped of their votes.
They are literally removed from calculations, and the place is filled in a recalculation that is biased in favour of the largest parties. The key is that if voters in the election where they do not take up a seat knew that they could or would be elected elsewhere, then they may have supported another independent candidate or party that would then have gained a seat that they now do not get.
5. The recalculation used for filling vacancies arising from an independent vacating their seats in a legislature is numerically biased in favour of the largest party or parties.
6. The signature requirement for independent candidates is very high, being set at 20% of the quota from the previous comparable election. For the National Assembly, this will be more than 8 000 signatures (providing that the wording is corrected to accommodate the true intent of Parliament. Political parties are required, upon initial registration by the IEC, to supply 1 000 signatures. An important detail is that independent candidates are required to submit the names and ID Numbers of the signatories to the IEC in electronic form, to allow for automated verification that they are all registered voters.
7. In section 31B of the Bill, the requirement for the submission of signatures says that the number must be at least as many at "20% of the quota" in the previous comparable election. For the National Assembly, this quota is between about 68 000 and 92 000 votes, whereas the Members of Parliament are under the impression that it is about 44 000.
8. In Item 21 of the text of Schedule 1A, it states that the recalculation mentioned in either Item 7 or Item 12 must be carried out in a particular circumstance. However, those items, while specifying the circumstances (and hence the referring items) in which the recalculation may be invoked, do not mention Item 21. When a similar error was pointed out in relation to Item 22 and Items 7 and 12, the corrections made were to duplicate text from those items in Item 22, rather than simply adding the necessary reference into Items 7 and 12.
9. The recalculation method, because it biases in favour of parties with more quota seats, has an anomaly where the filling of a vacancy in a provincial legislature caused by an independent candidate instead taking a seat in the National Assembly can remove a seat from a smaller party and award it to a larger party. This can happen up to two days after the results are announced. The equivalent anomaly arising from filling a vacancy at a later stage was "corrected" by a further recalculation, following a submission made to the IEC to this effect.
By avoiding a constituency system, the Bill inherently creates the probability of long ballot papers, as the potential exists for many independent candidates to stand in each election. The reaction to this likelihood has been to raise the barriers to entry for independent candidates with the very high signatures requirement.
The argument has been made in a number of different contexts that no system is perfectly proportional, and so objections with respect to proportionality should not worry us. The critique of the Bill is based on a very clear analysis that shows the distortions of proportionality are built into the system, based on the discarding of votes, and on the particular recalculations that have been used to deal with those discarded votes. Those distortions are also clearly indicative of numerical bias that renders the elections unfair.
If the distortions from proportionality are generalised, then it is not logically possible to declare that the system results, in general, in proportional representation.
What Happens Next
With the passing of the Electoral Amendment Bill in the National Assembly on October 20, it is now before the National Council of Provinces (NCOP). The NCOP has called for further public submissions concerning the Bill, with a deadline of November 9. These submissions are not strictly required, and it seems that there is little to add to the submissions made to the Home Affairs Portfolio Committee during September. It may thus be seen as an attempt to forestall criticism of the process of public participation in relation to the passing of the Bill.
Public submissions will have to be collated and reviewed by the select Committee of the NCOP. After deliberation, the Bill is likely to be presented to the NCOP itself for a vote. This is likely to occur only in the last week of November. There is always the logical possibility that the NCOP will make changes to the Bill and refer it back to the National Assembly for ratification. In general, these would tend to be more technical changes, and there is little expectation of substantive changes.
Once Parliament has finally approved the Bill, it will be sent to the President for signing. The court has given Parliament until December 10 to remedy the constitutional defects of the Electoral Act. The President, if he has concerns over the constitutionality of the Bill may refer it back to Parliament for these concerns (which must be stipulated) to be addressed. Should there be no changes made by Parliament, or should the President be unsatisfied with changes made, then he has the prerogative to refer the Bill to the Constitutional Court. Civil society groups are already calling on the President, should Parliament pass the Bill, to refer it back, and then to the Constitutional Court. This is the simplest and most efficient route for the constitutionality of the Bill to be tested.
There is some speculation that the President may apply for a short extension of the Constitutional Court deadline, in order for him to apply his mind to the merits of the Bill. This would not be a problem in itself, but each delay increases the pressure on the deadlines for the 2024 election.
Why Has this Gone so Wrong?
The absurdity of individuals contesting proportional representation elections, pointed out above, is at the heart of the whole situation. We are in the midst of an existential crisis precisely because this absurdity has been entertained and expounded upon in pseudo-rational terms. We needed somebody to speak up at the outset, and to state this simple truth.
The inclusion of the so-called, "Option 1" of the MAC Report is the ground zero of the current crisis. This was insisted upon by Norman du Plessis (former COO and Deputy-Chief Electoral Officer of the IEC), Michael Sutcliffe (former head of the Municipal Demarcation Board), and Pansy Tlakula (former Chief Electoral Officer, and later Chairperson of the IEC). It is hard to imagine a collection of three South Africans more qualified to deliberate on electoral systems, and what is possible within our context. And yet, what they have proposed simply does not make sense.
The MAC Option 1 purports to be a minimalist option. It purports to be a minor modification of the existing electoral system, with the only change being the "insertion" of independent candidates. But it is neither of these things. It is a new electoral system, where independent candidates are treated, temporarily, as political parties. It is difficult to know what motivated the advancement of MAC Option 1, except perhaps that time was running out to implement a constituency system, as the situation ultimately required. We have seen how, as deliberations on the Bill progressed, Parliament has had to deal with some of the logical outcomes of the contradictions inherent in this arrangement.
An electoral system is precisely that – a system. The interconnections between all parts of a system mean that it is not possible to make a change to one part of a system, or to the inputs, without having a corresponding effect elsewhere and on the outputs. As Parliament deliberated on the Bill, and as the IEC and others proposed changes, no analysis was carried out with respect to the consequences of those changes elsewhere in the system. An adjustment to the seat allocation system automatically creates changes to balance of how seats are allocated. Allowing independent candidates to contest multiple elections means that there are distortions when they win seats that cannot be taken up.
The drafting of the original Bill was carried out without a clear understanding of how the current seat allocation mechanisms work. Much of the legal advice during the entire process has seemed to entirely ignore the question of proportionality. In fact, very little numerical analysis has been carried out, even while criticisms have been waved away on the grounds that no electoral system is perfectly proportional.
As changes have been made, such as allowing independent candidates to contest more than one election, there has been no logical or numerical analysis of the implications of those changes. There has been almost no discussion to evaluate particular aspects of the Bill against the different constitutional criteria that apply. In particular, there has been no analysis of the numerical effect of each vote cast. As Valli Moosa put so succinctly, everyone must be able to cast votes that are of equal worth.
A particular thread has run through all of the processes and deliberations surrounding the Bill – on the part of all of the bodies and entities engaging officially with this Bill, there has been an unhealthy deference to the Executive. A failure to take initiative, a failure to ask difficult questions, an unwillingness to contradict or question the decisions taken by the Minister. We have a number of bodies, such as the State Law Adviser, Parliamentary Legal Services, the Department of Home Affairs, and the IEC, who have a duty to inform this process on the basis of their mandate to serve the country, the Constitution and the legislative process, above any task of facilitating the wishes of the Executive.
What Happens for the 2024 Election?
We are already in a constitutional crisis. It is now not possible to hold a constitutionally compliant election in 2024. In fact, this was true before the end of 2021. The Electoral Act is unconstitutional, and the Electoral Amendment Bill has no prospect of being constitutional. The only valid solution, that of implementing a hybrid constituency system, is not possible before the 2024 election.
If the President signs the Bill, then the next step is almost certainly that a number of civil society bodies will challenge the validity in the Constitutional Court. Importantly, it will be necessary to apply to the Constitutional Court for direct access. If this is not granted, then it would be necessary for the challenge to be heard first in the High Court, which would create catastrophic delays.
The core constitutional problems are the violation of the constitutional requirement of proportionality (ss 46 and 105), together with the right to equality before the law (s 9), and the right to free and fair elections (s 19). It will also be shown that the Bill does not give sufficient expression to the New Nation ruling that independent candidates must be able to contest elections in a meaningful manner.
If the court finds that the amendments to the Electoral Act are unconstitutional, wholly or in part, then a very difficult decision is required with respect to the conduct of the 2024 election.
The court will have to condone some form of unconstitutionality for 2024. It would not be proper to presume what the final ruling would be, but there are a few logical scenarios:
1. The amendments to the Bill could be ruled unconstitutional in their entirety, and the 2024 election could be held in terms of the existing system, as expressed in the current Electoral Act. This would clearly then exclude independent candidates from the 2024 election;
2. Parts of the current amendment Bill could be ruled unconstitutional, and those parts struck down. the 2024 election would then be carried out partly in terms of the existing system, and partly in terms of the new system; and
3. In striking down some of the more problematic amendments, the court might "write in" a few targeted changes to allow at least a partial involvement of independent candidates, with some improvements to how the Bill is currently structured.
None of these outcomes is a happy one. It seems inconceivable that, with the court having ruled in June 2020, we find ourselves still unable to give effect to the rights of independent candidates in 2024. However, the court is particularly reluctant to "write in" any law. In the original judgment, there was a very clear statement by the court that the determination of an electoral system lay entirely within the purview of Parliament.
However, without some changes to the Bill, it is almost certain that elections can be shown to be not fair, even if the changes are implemented only in part.
One potential outcome could be that the election for the National Assembly is conducted in terms of the new system, but with a few key adjustments to the calculations. The provincial legislatures are much more difficult (given that there is only a single ballot), and so it might be ruled that independent candidates are not permitted to contest provincial elections.
In this arrangement, certain minor areas of unfairness with the elections for the National Assembly would be condoned, as would the unconstitutionality of independent candidates being unable to contest elections for the provincial legislatures.
Essentially, we already have a conflict between the rights of individuals to contest national and provincial elections as independents, and the right of citizens to fair elections. We cannot say at this point what the precise outcome will be.
A practical consideration that adds a great of pressure to all concerned is that, once the Bill is signed by the President, the IEC has the immediate duty to rebuild a significant portion of their IT systems, and redesign the processes and workflows to accommodate the changes. This is lengthy process, taking months to complete, and to verify. If the new electoral system is challenged during that process, and if it is struck down, then much of that new work will be in vain. If we get a resolution of the constitutional conundrums during 2023, then it is probably likely that it will be possible to adapt. The existing IT systems will be in place. However, if the process of challenging the new system is resolved only during 2024, then we may face the real risk that it is not possible to hold an election within the required period.
Calls for Meaningful Electoral Reform
As awareness of the Electoral Amendment Bill has increased, and as civil society bodies have contemplated the implications, there has been a growing clamour for more meaningful electoral reform. The Zondo Report spoke of the problems of Members of Parliament not holding the Executive to account, given that they are "beholden to their party bosses" by virtue of the electoral system. Although the question accountability of Members of Parliament to the electorate was not a particular consideration on the New Nation judgment, many who are speaking up now are raising the question of accountability.
In the National Assembly debate on the Electoral Amendment Bill on October 20, Minister Aaron Motsoaledi observed that critics of the Bill were conflating the separate issues of the Constitutional Court ruling being about independent candidates, and the question of whether MPs should be elected to Parliament via constituencies. Technically, the Minister is correct, but for two important reasons, his statement does not help. Firstly, as already discussed, it turns out that constituencies are the only means to include independent candidates within the framework of a system of proportional representation. Secondly, he avoids mention of the fact that he consciously chose to avoid the electoral system that addressed considerations of accountability, preferring a system that did not.
We must of necessity go back to the electoral reform drawing board. We must also face up to an inevitable loss of public trust in the legitimacy of the processes surrounding electoral reform. We also cannot ignore the growing discontent with South Africa's recent experience of governance and perceived "failures" of democratic processes. As we get down once more to the process of electoral reform, it is perhaps incumbent on all of us to engage more vigorously. There are two very important reasons for this:
We are still in the process of extricating ourselves from the mess caused by a member of the Executive arm of government choosing an electoral system for us and pushing it through Parliament (with the attendant layers of irony). As responsible citizens, we simply cannot stand by and allow the Executive to repeat this process (even if the system chosen happens to be fair, and constitutionally compliant). The court gave Parliament the duty of rectifying the constitutional defect in the Electoral Act, and Parliament is OUR representative body, to whom the Executive must account.
Restoring public trust in the process of electoral reform, and hence in the outcomes, necessitates far greater public involvement in that process. Had a valid electoral system been implemented following the New Nation ruling, there would perhaps have been no problem with muted public involvement in the process. But public perceptions over the validity of electoral processes are almost as important as the processes themselves.
Therefore, as we move forward, we should be giving serious thought to electoral systems that give independent candidates their rightful place, while being constitutionally valid, and producing fair outcomes. We should also be taking into account social context, our experience of voting over the past 25 years, and the practical and logistical implications.
Valid Electoral Systems That Include Independent Candidates
An electoral system is not merely an administrative matter. The form and structure of the system, and the processes involved, make a profound difference to outcomes, and ultimately to perceptions of elections, and hence of democracy itself. We dare not take another misstep on the road to electoral reform.
The Constitutional Court gave Parliament twenty-four months to remedy the constitutional defect in the Electoral Act but was at pains to make clear that it was up to Parliament to decide how this could be done. Much has been made of this to deflect criticism of the Electoral Amendment Bill, missing the obvious point that Parliament is free only to make laws that are consistent with the Constitution.
From our experience in looking at the effects of the Electoral Amendment Bill, we know that we need to take great care with the structure of ballots, and the seat allocation calculations. We MUST ensure that the required "in general, proportional representation" is maintained, and that elections are fair from the perspective of both the participants, and the voters themselves. We need to be alert to the possibility of unintended consequences. having ensured that we consider only valid electoral systems, we can then start discussing
What was perhaps missing in the court's understanding, and indeed in much of what has followed, is that the only means to accommodate individuals within a proportional representation system is a hybrid of some form of constituency elections, balanced by an overall proportional representation ballot.
Although there are variations and some alternatives available, there are two primary forms of constituency elections that can be combined with proportional representation to form a valid system incorporating independent candidates:
The first is a single-member constituency system, as proposed by the Ministerial Advisory Committee, where 200 constituencies are demarcated, and the balance of seats awarded to reach totals determined by a separate proportional representation ballot. This two-ballot system would apply for the National Assembly election, as well as for the provincial legislature elections. The system proposed is similar to that employed for the local government election, with one significant difference being that the constituency votes (Ward ballot, in the case of local government) would not count as part of the total proportional representation tallies.
The other viable option is the multi-member constituency system, where 300 constituency seats are divided into between 60 and 70 constituencies, each having between three and seven seats. Within each constituency, a proportional representation election would determine the members to represent the constituency, where independent candidates may stand alongside parties. In order to retain proportionality, the system would need to have two National Assembly and two provincial legislature ballots. This system was initially proposed, albeit without independent candidates, by the 2003 Electoral Task Team, headed by Frederik van Zyl Slabbert, and reiterated by the High-Level Panel chaired by Kgalema Motlanthe in 2017, reviewing key legislation.
A possible variation of the multi-member constituency system would be an open-list system, where all of the candidates from each of the parties, as well as independent candidates, is shown on the ballot paper. Voters would vote for a particular candidate, giving more direct influence over the selection of Members of Parliament. An alternate system already proposed is a variation of the multi-member constituency system, where votes cast for independent candidates are transferable. This was introduced by Mosouia Lekota to Parliament in 2020 but has not found general appeal. Although the proposals have their advantages, the complexity involved, and the lack of general appeal make it difficult to spend time on in the current fraught situation.
The Ministerial Advisory Committee set out principles to guide our consideration of electoral systems:
1. Inclusiveness (national unity);
2. Fairness;
3. Simplicity;
4. Accountability;
5. Gender equality;
6. Proportionality;
7. Effective participation of independents;
8. Genuine choice;
9. Effectiveness; and
10. Legitimacy.
Note: for any who may feel that gender equality should not be a criterion at this point, it is necessary to take into account that in many societies, women still tend not to fare well in direct (constituency) elections, which leads to a preponderance of men among candidates, and hence among elected representatives. This is a fact that we cannot ignore in designing electoral systems, particularly with our history of patriarchy. This is not a radical or subversive ideological position, but rather a simple observation about society that we cannot avoid. It is obviously easier to ensure a greater participation of women where party lists systems are used, as policies and rules can lead to greater involvement of women, without diminishing voter support.
To complement that, we can add the constitutional imperatives for elections:
1. In general proportional representation (ss 46 and 105);
2. Fair elections (s 19); and
3. Equality before the law (s 9).
An initial case is made, therefore, for the two simplest and most common forms of constituency elections that can be used in a hybrid system, with overarching proportional representation. These are presented as an initial discussion, to be explored further.
In any form of constituency system, the constituency results, comprising a portion (often half) of the total seats are calculated in their own right. In a distinct calculation, seat allocation is carried out for parties only, using the full set of seats, less the number of constituency seats that have been won by independent candidates. The difference for each party between the seats won in constituency elections and the total seats determined by the proportional representation election is then filled from party lists. Typically, the largest party or parties are over-represented in constituency elections, but with a fairly allocated number of seats in the over PR election, a balance is reached with the party list seats.
The Case for Multi-member Constituencies
Multi-member constituencies would be regions or districts with enough voters to elect between three and seven members of the National Assembly. For ease and simplicity, existing metro municipality and district council boundaries would be used. Metros would have between one and three constituencies each, while District Councils would, in general, constitute a multi-member constituency. In a few cases District Councils would be combined, due to sparser populations. Constituencies would not cross provincial or metro boundaries, and only cross District Council boundaries where they need to be combined.
The same constituency boundaries would be used for provincial legislature elections, although the numbers of MPLs elected in each constituency would vary, due to the relative differences in the numbers of members of provincial legislatures This would happen more in the smaller provinces, due to the minimum size of a provincial legislature being 30 members. The numbers of registered voters in each constituency would be reasonably balanced, although there would be some variations. Having distinct proportional representation ballots to balance out overall representivity in each election would remove any potential problems arising from variations in sizes of constituency and would eliminate any concerns over gerrymandering.
The primary advantage of multi-member constituencies is that each district /constituency would have MPs and MPLs that reflect the overall voter choices in the particular area. In each constituency, it is near certain that more than one party would be represented, which would create a greater local connection between voters, and their elected Members of Parliament.
Given that in each of the multi-member constituencies, elections are conducted in terms of party lists, this system makes it relatively easy to include greater numbers of women in the political process.
The demarcation of multi-member constituencies is a relatively straightforward process, employing the principles described above.
The Case for Single-member Constituencies
The country would be divided into 200 or more individual constituencies. No constituency would cross provincial or metro municipality boundaries. Outside of the metros, constituencies would mostly be contained within District Councils, with a few exceptions requiring the combining of adjacent District Councils to form the pool for constituencies.
There would be some complications for provincial legislatures arising from the differing relative sizes of the legislatures. In the main, constituency boundaries for provincial legislatures could be kept the same as for the National Assembly, but certain careful adjustments would be needed. The particular adjustments would also take into account possible changes to the composition of provincial legislatures, and the formulas used to determine the numbers of National Assembly seats to be allocated to each province.
The most immediate perceived benefit of single-member constituencies would be a much closer connection between Members of Parliament and the electorate. While constituency elections are clearly not a panacea and would not guarantee the much sought-after "accountability" among MPs, a single constituency system creates the possibility of this happening, while the current party list system does not.
A second benefit would be that the system would mostly mirror that of local government and would thus be familiar to voters.
Where the MAC proposal (correctly) differs from the system used at local government is that the constituency ballots will not be added to the proportional representation ballots to calculate the overall representation in the respective legislatures. This is an exceptionally important distinction, pointing to the potential need to revise the current local government system. If constituency ballots do not count as part of the proportional representation totals, then political parties do not need to field candidates in every constituency (as they do not lose votes from not fielding a candidate). While some larger parties may field candidates in all or most constituencies, as a matter of principle, it would free small- and medium-sized parties from the burden of campaigning in constituency elections where they have no realistic prospect of winning. Importantly, this would mean that ballots would be much shorter in constituency elections. Elections with a few local candidates would be more meaningful and provide a greater incentive and better prospects for strong local independent candidates to contest.
But What About Chantal Revell?
The whole process of electoral reform has come about because of Chantal Revell, the New Nations Movement NPC and others, together with those who have supported their cause through the courts. It is therefore important for us to consider how the current Electoral Amendment Bill would affect her, but more importantly how a valid constituency system would work for someone in her position.
This is a very difficult question. Chantal Revel and the New Nation Movement have brought out the question of different types of representation. Essentially, there is a tension in how the size of electoral districts affects how one chooses to contest elections.
Constituency elections are, by nature, more localised. As an independent candidate, one would contest a constituency on the basis of having relatively concentrated support in a confined geographical district. In single-member constituencies, it is therefore obviously necessary to obtain more votes than any other candidate. in a multi-member constituency (with between three and seven candidates), support is drawn from a wider pool, and it is not necessary to secure an outright majority.
The Electoral Amendment Bill, however, has stretched the definition of constituencies, to make the entire province, technically speaking, a multi-member constituency. Superficially, this arrangement would benefit Chantal Revell, as she would likely draw support from across a wide area. For her, a province would be the ideal geographic district from which to attract support.
However, the dilemma that Chantal faces is that if she were to stand for election in terms of the Bill as it stands, the stark reality is that each vote cast for her as an independent candidate will contribute to additional seats to be won by the largest party in that election. This is a violation of the heart of the Constitutional Court ruling that sought to give greater expression to the right to make political choicer, and to contest elections, along with the rights to freedom of association and to fair elections.
Once we move to a more viable constituency system, it is inevitable that these constituencies will be more localised than whole provinces. Chantal Revell will then be faced with the choice of contesting a constituency election as an independent candidate, or of forming a political party to be able to draw support from whole province, or the whole country.
Beyond the right of individuals to stand for election as independent candidates, we will all ultimately benefit from the process of electoral reform. It could well be that Chantal Revell, having taken this fight through the courts for several years, might not be able to take full advantage of the rights that she has won for all of us.
The Road Ahead
It would be a brave person who predicts how the course of electoral reform will move forward from where we are now.
At the moment, Minister Aaron Motsoaledi, and the ANC are caught in a trap of their own making. Understandably, having committed themselves, almost certainly without appreciating the unintended effects of the Electoral Amendment Bill, they are proceeding with the Bill in its current form. It seems inconceivable that they would now suddenly contradict all of their confident pronouncements on the Bill and withdraw it. This would entail an admission to the court that Parliament has failed entirely in giving effect to the New Nation ruling. This would involve a significant loss of face, even though it may be the best outcome for the country as a whole.
It is difficult to speculate as to the thinking of the Minister and of the ANC at this point. While they have pronounced their confidence in the validity and fairness of the Bill, there was the curious incident of the ANC sending Snuki Zikalala to the Ahmed Kathrada Annual Lecture during October. He made the astonishing statement that the ANC did not support the Bill, and that the Minister was open to discussions with civil society on the matter. This was followed rapidly the next day with a denial issued by the Minister and the Department of Home Affairs. One possible avenue for the ANC, at this stage, might be that the President refers the Bill back to Parliament, and then to the Constitutional Court, specifying his constitutional concerns.
The court will need to rule on the validity of the Bill, and to find a way to conduct the 2024 elections. Whatever the outcomes for 2024, it is inconceivable that any version of this current Bill will be employed thereafter.
What this means is that we have to start thinking about what comes after that. Fundamentally, the problem with the process thus far is that it was driven by the Executive. This problem has been magnified by the fact that the Minister has been poorly advised in respect of the consequences of the Bill, and the question of proportionality. Parliament waited for, and then followed the lead of the Minister. Seemingly, they have deferred to the Minister in all of the deliberations.
We need to begin again. But this time, we need to do it right. The Minister treated the inclusion of independent candidates as a technical, or administrative change, but this is not the case. We need the process to be more broad based and informed by all that we have learned up to now.
An electoral system lies at the heart of any democracy. It is part of the DNA of any society. As such, it is essential that society is actively engaged in the process of choosing a new system. In one sense, the debacle with the Electoral Amendment Bill has served to bring this fact to the fore. Until now, there has not been much awareness of the changes required to our electoral system. It may be that the distortions and unfairness of the Bill will drive a far more meaningful public participation.
As much as we are still occupied with the problem of the 2024 election, we need to begin a national conversation concerning what come after that. As citizens and voters, we must have a sense of ownership of our elections. Regardless of political philosophy, all who value the primacy of democratic expression should stand together and seek out the best electoral system.
It is therefore heartening to see the awareness that is growing, and the active participation of a broad range of civil society bodies. It is hard to think of any national issue, with the possible exception of the HIV/AIDS treatment fiasco, where such a diverse gathering has stood in one accord. On this foundation, we can build for the future, as we come together on the basis of shared democratic values.
Appendix - A Draft Bill
The key problem with the Electoral Amendment Bill is that it is based on an electoral system that cannot be consistent with the Constitution. The essence of that system is described in the text that is intended to replace Schedule 1A of the Electoral Act. This contains the seat allocation mechanisms that underpin the chosen system. While there are some provisions in the body of the Bill that reflect the type of electoral system to be used, the heart lies in Schedule 1A.
What is presented here is a proposed draft text to replace Schedule 1A, to implement an electoral system based on single-member constituencies, as articulated in the MAC Option 2. This is a first draft, intended to stimulate discussion. Clearly, there may be omissions or other aspects that require improvement.
Notes:
1. The included text addresses the allocation of seats, but not the incidental matters relating to maintenance of lists between elections.
2. The draft of Schedule 1A leaves unresolved the question of the number constituencies for the National Assembly and the number for provincial legislatures not being equal in all cases.
3. The draft assumes that the number of seats available in any of the elections is an even number.
4. This system would not allow any candidate to contest an election for more than one constituency, whether for the National Assembly or for provincial legislatures.
5. Unlike local government elections, the proportional representation seat allocations take into account only votes cast in the proportional reprsentation ballots.
6. In the event that a party obtains more seats in an election than the number of names they supply to the Commission, the party would be given two days to supplement the list, rather than forfeiting those seats immediately, as is the case with the current Act.
7. In the unlikely event that a party fails to provide supplementary names as described above, a recalculation would be carried out by adding the number of forfeited seats to the set of remainder seats to be allocated in each case.
8. In the National Assembly, remainder seats are currently allocated on the basis of surplus votes, to a maximum of five, and then in order of the highest average votes per seat on seats already allocated (this favours smaller parties with seats already allocated over parties with lower surplus votes, whether they have seats or not). With an increase in the number of parties participating in elections, the number of remainder seats increases. This paper proposes that the seats awarded on the basis of highest surplus be set to at half of the total remainder seats, to create a greater balance.
9. A possible variation to the seat allocation for provincial legislatures could be the inclusion of highest average calculations for remainder seats, on the basis described for the National Assembly. This would prevent the allocation of remainder seats being based on very low numbers of surplus votes.
Schedule 1A
SYSTEM OF REPRESENTATION IN NATIONAL ASSEMBLY AND PROVINCIAL LEGISLATURES
National Assembly
1. Votes cast for the National Assembly shall be on two ballots, as follows:
(a) Voters shall vote for a candidate nominated to contest the constituency in which the voter is registered. This shall be known as the constituency ballot; and
(b) Voters shall vote for a party on the proportional representation ballot.
2. The seats in the National Assembly are as determined in terms of section 46 of the Constitution and item 1 of Schedule 3 and are allocated as follows:
(a) half of the seats are filled by independent candidates and candidates from parties contesting the constituencies. These shall be referred to as constituency seats; and
(b) the remaining seats are filled by candidates from lists of candidates of parties and these shall be referred to compensatory seats.
3. (1) The Commission must prepare lists of candidates for each of the constituency seats, comprising:
(a) Candidates nominated by parties for that constituency; and
(b) Independent candidates contesting that constituency.
(2) No candidate may contest more than one constituency, whether in the National Assembly or a provincial legislature.
4. (1) Registered parties contesting an election of the National Assembly must nominate candidates on a list of candidates to fill compensatory seats.
(2) The list of candidates submitted by a party must not contain more names than the number of compensatory seats in the National Assembly, and each such list must denote the fixed order of preference of the names as the party may determine.
(3) A candidate nominated by a party for a constituency election may also be included on that party’s list of candidates, providing that, if the candidate wins the constituency election, they will be omitted from the candidate list in the allocation of compensatory seats.
5. The seats in the National Assembly must be allocated to parties contesting an election as follows:
(a) The total seats to be allocated to parties is calculated by subtracting the number of seats won by independent candidates in constituency elections from the total number of seats in the National Assembly.
(b) A quota of votes per seat must be determined by dividing the total number of valid votes cast in the proportional representation ballot referred to in Item 1(b) by the total seats calculated in paragraph (a) plus one, and the result plus one, disregarding fractions, is the quota of votes per seat.
(c) The number of quota seats to be allocated to each party is calculated by dividing the proportional representation votes cast in favour of the party by the quota of votes per seat determined in terms of paragraph (a), disregarding fractions.
(d) If the total number of quota seats allocated in terms of paragraph (b) is less than the number of seats to be allocated to parties calculated in paragraph (a), then that difference is known as the number of remainder seats, and these seats are allocated in terms of paragraphs
(e) and (f).
(e) The surplus votes for parties competes for remainder seats, ranked from highest surplus to lowest, to a maximum of the lowest whole number of seats equal to at least half of the number of remainder seats.
(f) Any additional remainder seats are allocated in decreasing order of the average number of votes per quota seat allocated.
(g) The total seats allocated to each party is the sum of the quota seats and remainder seats allocated to that party in terms of paragraphs (c) and (d) to (f).
6. The compensatory seats for each party referred to in Item 2(b) are filled by subtracting the number of constituency seats won by that party in terms of Item 2(a) from the total seats allocated in terms of Item 5(g).
7. (1) If a party has submitted a list of candidates for the National Assembly containing fewer names than the number of seats allocated in terms of Item 5(g), then the party must provide to the Commission in the prescribed manner a supplementary list of names within two days of the results being announced, without altering the order or composition of the original list of names supplied.
(2) If a party fails to provide the supplementary list of names within two days, then the party forfeits a number of seats equal to the deficit.
(3) In the event of a forfeiture of seats in terms of subitem (2), the allocation of remainder seats conducted in terms of Item 5(d) to Item 5(f) would be repeated, but with the number of forfeited seats added to the number of remainder seats.
Provincial Legislatures
8. Votes cast for each provincial legislature shall be on two ballots, as follows:
(a) Voters shall vote for a candidate nominated to contest the constituency in which the voter is registered. This shall be known as the constituency ballot; and
(b) Voters shall vote for a party on the proportional representation ballot.
9. The seats in each provincial legislature are as determined in terms of section 105 of the Constitution and item 3(1) of Schedule 3 and are allocated as follows:
(a) half of the seats are filled by independent candidates and candidates from parties contesting the constituencies. These shall be referred to as constituency seats; and
(b) the remaining seats are filled by candidates from lists of candidates of parties and these shall be referred to compensatory seats.
10. (1) The Commission must prepare lists of candidates for each of the constituency seats, comprising:
(a) Candidates nominated by parties for that constituency; and
(b) Independent candidates contesting that constituency.
(2) No candidate may contest more than one constituency, whether in the National Assembly or a provincial legislature.
11. (1) Registered parties contesting an election of a provincial legislature must nominate candidates on a list of candidates to fill compensatory seats.
(2) The list of candidates submitted by a party must not contain more names than the number of compensatory seats in the provincial legislature, and each such list must denote the fixed order of preference of the names as the party may determine.
(3) A candidate nominated by a party for a constituency election may also be included on that party’s list of candidates, providing that, if the candidate wins the constituency election, they will be omitted from the candidate list in the allocation of compensatory seats.
12. The seats in each provincial legislature must be allocated to parties contesting an election as follows:
(a) The total seats to be allocated to parties is calculated by subtracting the number of seats won by independent candidates in constituency elections from the total number of seats in the provincial legislature.
(b) A quota of votes per seat must be determined by dividing the total number of valid votes cast in the proportional representation ballot referred to in Item 8(b) by the total seats calculated in paragraph (a) plus one, and the result plus one, disregarding fractions, is the quota of votes per seat.
(c) The number of quota seats to be allocated to each party is calculated by dividing the proportional representation votes cast in favour of the party by the quota of votes per seat determined in terms of paragraph (a), disregarding fractions.
(d) If the total number of quota seats allocated in terms of paragraph (b) is less than the number of seats to be allocated to parties calculated in paragraph (a), then that difference is known as the number of remainder seats, and these seats are allocated in terms of paragraphs (e).
(e) The surplus votes for parties competes for remainder seats, ranked from highest surplus to lowest.
(f) The total seats allocated to each party is the sum of the quota seats and remainder seats allocated to that party in terms of paragraphs (c) and (d) to (e).
13. The compensatory seats for each party referred to in Item 9(b) are filled by subtracting the number of constituency seats won by that party in terms of Item 9(a) from the total seats allocated in terms of Item 12(f).
14. (1) If a party has submitted a list of candidates for the National Assembly containing fewer names than the number of seats allocated in terms of Item 12(f), then the party must provide to the Commission in the prescribed manner a supplementary list of names within two days of the results being announced, without altering the order or composition of the original list of names supplied.
(2) If a party fails to provide the supplementary list of names within two days, then the party forfeits a number of seats equal to the deficit.
(3) In the event of a forfeiture of seats in terms of subitem (2), the allocation of remainder seats conducted in terms of Item 12(d) and Item 12(e) would be repeated, but with the number of forfeited seats added to the number of remainder seats.
15. (1) After the counting of votes has been concluded, the number of representatives of each party has been determined and the election result has been declared in terms of section 190 of the Constitution, the Commission must, within two days after such declaration, designate from each list of candidates, the representatives of each party in the National Assembly or provincial legislature.
(2) Following the designation in terms of subitem (1), if a candidate's name appears on a list for both the National Assembly and a provincial legislature (if an election of the Assembly and a provincial legislature is held at the same time), and such candidate is due for designation as a representative in more than one case, the party which submitted such lists must, within two days after the said declaration, indicate to the Commission from which list such candidate will be designated or in which legislature the candidate will serve, as the case may be, in which event the candidate's name must be deleted from the other lists.
(3) If a party fails to indicate to the Commission from which list a candidate will be designated or in which legislature a candidate will serve, such candidate's name must be deleted from all the lists.
(4) The Commission must forthwith publish the list of names of representatives in the legislature or legislatures.
Issued by the IRR, 4 November 2022
