Home NewsLocal News ‘Mixed Member Proportional electoral system manipulated’

‘Mixed Member Proportional electoral system manipulated’

by Lesotho Times

THE opposition is livid about the allocation of seats. Their anger has heightened following the announcement that the Southern African development Community appointed moderator, Ketumile Masire, was abandoning his mission.

The government is adamant that it did not do anything wrong.

As part of our search for answers this week we run a lightly edited version of an article that was written in June 2008 by Professor Jorgen Elklit, a respected Danish expert on electoral issue.

Elklit is one of the experts that the parties had agreed should come to Lesotho to explain the Mixed Member Proportional (MMP) model as part of efforts to resolve the dispute. The government has however said the experts will not be allowed into Lesotho.

This article appeared in the EISA’s Journal of African Elections — Special Issue: Elections and Democracy in Lesotho. The journal was published in June 2008. 


LESOTHO took an innovative step forward in 2001, when the introduction of the MMP electoral system became a central part of the political and consensual solutions to the upheaval following the 1998 elections.

MMP combines proportional representation on a national scale with single-member constituencies and the solution was seen by many as a promising step forward. The system was incorporated in Lesotho’s Constitution in 2001 and was used for the first time in the 2002 elections, where it clearly delivered on its promises.

In 2007, however, the picture was very different, primarily because the IEC had accepted the participation in the election of political parties which had formed informal alliances aimed at circumventing the 2001 constitutional amendment. The main problem was that the memorandum of understanding of one of alliances was accepted by the IEC, despite the fact that the intention was clearly to circumvent the Constitution.

To the consternation of the other parties, the arrangement gave the alliance an additional 20 seats. The fact that the alliance did not directly violate the electoral law and was accepted by the IEC has resulted in an extremely complicated political and legal impasse.

Recent decades have seen a welcome revitalisation of political as well as academic interest in electoral systems.

There is a renewed interest in the nexus between electoral systems and their effects on election results, political inclusivity, and political legitimacy.

One consequence of this is — at least in some countries and under some circumstances — a willingness to test new electoral systems, which might lead to expected or unexpected results different from those previously seen, because of contextual and institutional differences between the situations in which the systems are being implemented.

However, a change from one electoral system to another is not the only way to increase a party’s chances of electoral success.

Electoral engineers and their political masters might also, at least in certain situations, be tempted to manipulate the electoral system already in place to their own benefit, no matter what the consequences might prove to be for political inclusivity or legitimacy, not to speak of the assessment of such manipulation by external observers, development partners, or history.

The temptation to exploit a given electoral system to one’s immediate benefit is certainly not new, the best known example probably being the manipulation of constituency delimitation (so-called gerrymandering) — particularly when the electoral system is the single-member plurality (first-past-the-post —FPTP system), which has been practised since the early 19th century. Proportional representation (PR) systems have also been misused, one clear example being a major Danish party’s successful exploitation in 1947 of a loophole in the country’s two-tier electoral system, as it was at the time.

This paper investigates the interesting situation which arose during the 2007 general election in Lesotho, where — at least in the eyes of some observers — the recently introduced mixed member proportional (MMP) electoral system was manipulated to such a degree by the Lesotho Congress for Democracy (LCD), the incumbent party, that the only possible answer was a strong reaction both inside and outside the National Assembly.

As a former British protectorate Lesotho inherited the FPTP system at inde­pendence in 1966. The turbulent political history of the mountain kingdom, marked, in the main, by abruptly changing political allegiances and military governments, is key to understanding the development of the Basotho political mentality, but need not be explored here, as our purpose is different.

The 1998 elections for the 80 seats in the National Assembly triggered a long sequence of dramatic events because the incumbent party, the LCD, was able to win 79 of the 80 seats although it received only slightly more than 60 percent of the total vote.

Many in the 11 opposition parties (some of them quite small) cried foul, but the election was actually relatively clean. Therefore, it must be accepted that this election was one of those rare occasions where one party won at least relative majorities in almost all constituencies.

Post-election violence nevertheless soon escalated, public buildings in Maseru — and some private ones as well — were burned down, and the government began to question the loyalty of the army and the police force. South Africa and Botswana intervened on behalf of SADC, a military and diplomatic intervention which ensured the restoration of peace.

As part of the resultant political agreement (which had the full backing of the international community) a new political institution in Lesotho, the Interim Political Authority (IPA), was established, mandated to propose institutional changes (constitutional, legislative, or other), which might eventually end the political impasse.

The IPA consisted of two members from each of the 12 political parties that participated in the 1998 elections. The inspiration for this format undoubtedly came from the South African negotiation fora of the early 1990s and was apparently suggested by a member of the South African delegation to the settlement talks.

However, identical solutions only make sense when the problems are identical, or nearly so, which was not the case here. As a consequence, a complicated situation soon arose when the 22 opposition representatives (some of them from minuscule parties) opted for a solution, first suggested by a German political scientist, which was to introduce an electoral system similar to the German MMP system, combining single-member constituencies with an overarching PR element.

The two members of the ruling LCD objected strongly and a political stalemate developed when the majority of the members of the IPA — mandated to propose constitutional and other amendments — insisted on the MMP solution, while the LCD government (with its 79:1 majority in the National Assembly) benefited from the fact that all IPA proposals would eventually have to be formalised according to existing legislative and constitutional requirements.

It soon became clear that it would not be possible to hold fresh elections in May 2000, as scheduled in the international agreement brokered primarily by SADC with the support of the United Nations Development Programme and the Commonwealth Secretariat.

The IPA had quite a blurred idea of the practical and logistical consequences of its own decision when the author of this paper was drawn into the picture in late 1999.

The overwhelming majority was nevertheless adamant that it had made a worthwhile decision, while the LCD government was equally adamant that it could always block all IPA proposals, as they would have to be enacted by Parliament.

The LCD’s preferred option was not the MMP system, but the parallel (or mixed member majority — MMM) system, where only a fraction of the seats are allocated by PR, which means that it is not a genuine PR system.

The MMP system is, however, a genuine PR system, as all seats are included in the conscious attempt to reach a proportional result through the use of a strong compensatory mechanism.

It was only during the second half of 2001 that a compromise was eventually reached, maintaining the IPA’s preferred MMP option, but with a seat combination reflecting directly the LCD’s proposal for a parallel system (80 single-member constituency seats and 40 PR seats).

The political compromise was then implemented through a constitutional amendment, in which MMP is mentioned specifically, and by relevant changes to the electoral law, primarily in the form of a new schedule (known as Schedule 5).

With these changes in place the Independent Electoral Commission (IEC) could finalise its preparations for the next National Assembly election, which took place in May 2002. All in all, the election went well and most noteworthy, with a result that reflected the electorate’s overall partisan distribution much more effectively than had been the case in any previous election in Lesotho’s troubled political history.

This outcome clearly demonstrates how a constructive exchange of ideas and suggestions of solutions between local power structures, donors, and foreign advisors may contribute to the joint development of useful solutions.

As a couple of candidates had died prior to the 2002 election two fresh constituency elections were conducted later in 2002, leaving the LCD with a clear 79:41 parliamentary majority.

It soon became clear, however, that the governing party had not really bought into the idea of a more consensual political style and a consensus-oriented parliamentary system, which is the natural complement to a PR electoral system, especially as this type of electoral system usually does not go well with the traditional Westminster-style parliamentary system.

The LCD government / party leadership continually tried to hamper and annoy opposition parties and MPs elected to compensatory seats.

Furthermore, various electoral system adjustments, which would probably have decreased the prospects of the opposition in the following election, were also suggested.

In a word, the old political power game was still being played, even though the specific proposals were eventually dropped.

Lesotho has a five-year electoral cycle, so the next National Assembly election was expected in mid-2007, either before or after the period when winter-weather conditions in the mountain districts cause logistical problems.

However, it was no big surprise that Prime Minister Pakalitha Mosisili, as head of government, called an early election, to be conducted in February, as that was likely to inconvenience his former colleague in government and long-time LCD stalwart, Tom Thabane, who was trying to prepare for the up-coming election at the head of a new party, the All Basotho Congress (ABC), a splinter-group from the LCD.

Anxious to be returned with more than the slim majority Mosisili com­manded on the eve of the election after the defection by Thabane and his followers, the LCD had developed a strategy which, to all intents and purposes, would annihilate the seat allocation effects of the MMP electoral system, which the party had accepted and voted for in the National Assembly as part of the 2001 constitutional and electoral law amendment compromise.

The basic idea was to take in as a partner a minor political party and, for this purpose, the National Independent Party (NIP), was chosen.

The NIP had done reasonably well in the 2002 general election (to some observers’ surprise), when the party won five compensatory seats.

On 15 January 2007 the two parties signed a memorandum of understanding on strategic partnership and co­operation for the 2007 general elections’, as the front page reads.

The core of the agreement (Art 3) spelled out that the LCD would only compete for the single-member constituencies while the NIP would contest only the compensatory seats.

However, and most remarkably, the NIP party list should be compiled by the executive committees of the two parties according to a clear-cut formula, which gave the NIP the five first positions on the list, the LCD the following six candidates (who would also run in single-member constituencies), the LCD the following four for some ‘ordinary’ party-list candidates, then five NIP candidates followed by ten LCD candidates, and thereafter, alternately, one from the NIP, one from the LCD, and so on.

The article also states that the followers of the two parties as well as the general electorate should be sensitised to vote for the LCD on the constituency ballot and for the NIP on the party ballot.

The IEC had, in early December 2006, declined to take a clear position on the 

legality of an informal agreement between two registered political parties, merely reminding the LCD in writing that only registered parties may lawfully endorse candidates for election.

So the LCD / NIP Memorandum of Under­standing (MoU), developed and endorsed by the two parties, became the basis for subsequently submitting to the IEC 80 LCD constituency nominations as well as a NIP party list with 50 names on it.

This was possible for two reasons — the electoral law does not request a party to compete for both constituency and compensatory seats (as it also does not in other MMP systems) and there is nothing in the current electoral law which specifically forbids a candidate from one party also to stand for another party (which goes against the practice in electoral systems in most other countries).

 The importance of the agreement became even more evident when Mosisili was selected as the first of the LCD candidates on the NIP party list (number 6 on the list), with other top positions on the list also going to high-ranking LCD members.

The National Assembly election on 17 February 2007 yielded more or less predictable results: the LCD took 61 constituencies and the ABC 17.  One of the two remaining constituencies went to a well-established political personality, Kelebone Maope, a former deputy prime minister in the LCD government, now the leader of the Lesotho People’s Congress (LPC), a small political party which had merged with two others to form the Alliance of Congress Parties (ACP), while the election in the last constituency was postponed because of the death of one of the candidates.

The constituency was later won by the LCD.

As the LCD and the ABC did not run party lists they could and should not be included in the allocation of the 40 compensatory (PR) seats (which are, of course, based on the eligible parties’ overall performance in relation to the 119 seats available for allocation).

This created considerable confusion among many participants in Basotho politics, who had difficulty understanding — and accepting — that the IEC had to allocate the compensatory seats according to the rules in the Electoral Act (Schedule 5), and that computations had to be based only on the votes cast for those parties which had submitted party lists.

The first round of seat allocation yielded, as one would have expected, an odd result, as all ten parties with party lists became entitled to one or more of the 119 seats, which, in strict compliance with Schedule 5, had to be the basis for seat allocation during this first round of computations.

However, as only 41 seats were actually available for this purpose (41 being the sum of 40 compensatory seats and one constituency seat gained by a party with a party list, namely the ACP) a second round of computations had to be performed (again in full compliance with Schedule 5).

This increased the electoral quota by a factor of almost 3, and now the 40 compensatory seats were easily allocated to the nine parties entitled to them.

The NIP, the LCD’s alliance partner, had won 52 percent of all party votes cast and the Lesotho Workers’ Party (LWP), the ABC’s alliance partner, had won 24 per cent.

Consequently these two parties were allocated 21 and 10 compensatory seats, respectively. The remaining nine compensatory seats were scattered among seven parties, six of them getting one seat each, and one — the previously powerful Basotho National Party (BNP) — three.

As a hypothetical exercise, a private computation was subsequently con­ducted based on the realistic assumption that the NIP party vote could be used as a reliable estimate of the party vote a combined LCD /NIP would have been able to achieve.

Similarly, the LWP party vote was used as an equally realistic estimate of what the ABC/LWP alliance would have gained had the two parties stood under one banner.

The election results in constituencies (constituency votes as well as party votes, available on the IEC website) make these assumptions realistic.

This hypothetical calculation demonstrated that the LCD / NIP arrangement had secured an additional 20 seats for the two parties — had they run together they would probably only have garnered 62 seats (61 constituency seats + 1 com­pensatory seat) instead of their current allocation of 82 seats (LCD 61 constituency seats + 21 NIP compensatory seats — of which 10 went to NIP candidates and 11 to LCD candidates). The computations also showed that the ABC/ LWP had actually lost two seats because of this circumvention of the 2001 agreement, as the hypothetical calculation gave them 29 seats (17 from constituencies and 12 compensatory) compared to their current 27 seats (17 constituency + 10 com­pensatory seats).

The private calculations also demonstrated that all other parties (as well as the ABC / LWP alliance) would have obtained more compensatory seats than they actually did.

The seat allocation was subsequently challenged by one of the parties with only one seat, the Marematlou Freedom Party (MFP), and the High Court of Lesotho is, at the time of writing, looking into the matter, which is quite complicated, legally as well as politically.

Thereafter, the mediation by Sir Ketumile Masire is expected to continue and, hopefully, to reach a conclusion acceptable to all.

The legal and constitutional problem is obviously that Lesotho in 2001 formally amended its Constitution to include specific provisions for an MMP electoral system which is, by definition, a system with a well defined compensatory element.

It is clear that the LCD /NIP MoU — with the specifications of the party list positions to which each party was entitled — was a deliberate circumvention of the Constitution, as amended in 2001.

The reason for stating this in no uncertain terms is that the MoU would — if permitted by the IEC — secure for the LCD a number of compensatory seats over and above its (full) complement of seats won in the constituencies.

The fourth amendment to the Constitution states explicitly that the principle of proportional representation must be applied in respect of the National Assembly as a whole.

However, the IEC’s acceptance of the mixed LCD/NIP (and ABC/ LWP) lists as NIP and LWP party lists, respectively, has effectively disallowed seat allocation according to the principles and intentions of the current Lesotho Constitution.

The consequences of this combination of the deplorable acceptance of party lists with candidates standing for other parties in the constituencies and the subsequent allocation of compensatory seats adhering, to the letter, to Schedule 5 (which for good reasons was never challenged within the IEC) is that the 

electoral system in Lesotho has to a considerable extent — and in a manner probably unintended by the IEC — been changed entirely, from the MMP system of 2001 to a parallel system (or mixed member majority), that is, exactly the system which was rejected in 2001 as part of the over-all settlement and the subsequent constitutional and electoral law amendments referred to above.

So the claim that the Constitution of Lesotho (as amended) has been violated and circumvented, more by design than by default, is well founded.

Given that the IEC’s acceptance of the LCD /NIP memorandum of understanding was a major blunder, overlooking, as it did, that the consequence (and, indeed the intention) of this informal party alliance would be a serious violation of the Constitution, what is the outlook for the future?

The IEC focused on the regrettable oversight in the drafting of the amendment to the electoral law which omitted any explicit prohibition on candidates standing for two different parties in the same election and came to the conclusion that as it was not forbidden, it was acceptable.

The result of this political debacle is that it is very difficult to argue that the actual allocation of seats in February 2007 was improper. On the contrary, the seats were allocated meticulously, according to the letter of the law and the IEC’s acceptance of the LCD / NIP MoU.

Therefore, it is easy to understand that the LCD and the NIP — and their members of the National Assembly — feel strongly that they are fully entitled to the seats they hold.

It is equally difficult, however, to argue that the LCD / NIP MoU was not a conscious attempt to violate the Constitution (as amended in 2001), which the IEC ought to have realised.

The IEC should never have allowed the two parties to proceed with their informal alliance, which so obviously violates the Fourth Constitutional Amendment (Art 3). For this reason, it is easy to understand those aggrieved political parties – such as the MFP – who feel equally strongly that they have been cheated of seats in the National Assembly.

There is no easy way out of this impasse as both parties have a strong point. The only way forward appears — at least at the time of writing — to be, first, to close the loophole that appears to allow candidates to stand for more than one party in the same election.

Second, there must be a decision about whether parties should be allowed to field only constituency candidates or only submit a party list.

It is a difficult decision and there are no such requirements in other MMP systems such as those in Germany or New Zealand, because that is against the general principles behind the MMP system.

 It should be mentioned, however, that a recent MMP election in Albania (2005) failed to deliver a fair and proportional result for similar reasons.

The two alliance partners in Albania, however, did not submit a written MoU to the electoral authorities for early approval, relying only on the parties’ information campaign to supporters and voters as the channel for communicating their circumvention of the system – a strategy which, potentially, could be used in other countries as well.

Third, fresh elections should be called as soon as the legislation is in place, as there is no way in which a solid and acceptable political solution can be found.

Fresh elections would also allow the voters a chance to deliver their verdict on this complicated constitutional and legal case, which could then, hopefully, be put to rest.

You may also like