7.1 Introduction
Disputes arise in every election. An effective electoral dispute resolution (EDR) mechanism is required if an electoral system is to function properly. The election management body (EMB) needs to have in place certain principles and procedures to enable it to deal with disputes as they occur. The procedures should be known, rule-based and predictable. They should be designed and maintained to anticipate, manage and resolve election disputes. EDR should operate before, during, after and also between elections and become an essential element of good electoral management practice.
This chapter deals with various post-election procedures crucial to the integrity of any electoral process. It addresses post-election audits of the electoral process, post-election dispute management and the allocation of nominated seats. It also touches on the assumption of office by those elected, internal evaluation of the EMB and the custody of election materials.
7.2 Post-election audits and evaluations
Audits are standard practice in both the public and private sectors. They provide government departments and private businesses with an opportunity to identify mistakes and set benchmarks for future performance. There is no reason elections should be exempt from this standard accounting practice. Post-election audits provide election management bodies (EMBs) with the opportunity to improve on their performance and voters with a justified sense of confidence in election results. They make elections as transparent as possible by identifying shortcomings in the process and verifying the accuracy of vote counting. They can also point to areas of election fraud or rigging. Audits are an effective tool for building public confidence in election outcomes because they can detect human errors and help correct them.
While the ECK has not conducted proper post-election audits in the past, it has for some time now maintained a practice of holding an evaluation meeting after every election to assess how it was conducted and learn lessons to apply in the future. This is in keeping with good organisational practice generally, and for EMBs in particular, but does raise certain issues. Partly because these meetings are attended only by ECK commissioners and staff, any objective review of performance is hardly possible. These meetings produce evaluation reports that tend to gloss over major lacunae and in some cases are nothing more than self-congratulatory exercises. Given that EMBs guard their space and their independence jealously, rigorous self-examination could persuade the ECK to inject good practice into post-election evaluation exercises.
No EMB is perfect, but an EMB that learns from past mistakes and changes in response avoids their repetition and constantly improves the quality of its work. For instance, objective discussions of public expectations at County Hall in the 2002 general elections and at the KICC Media Centre in the 2005 referendum made it clear that, for the ECK to slake the public thirst for rapid and accurate results, it would have to put in place the required technology to improve results transmission. However, in the face of this, a seemingly inward-looking institution continued asserting that “only the ECK had authority to announce results” – forgetting that results, once announced at a polling station and in the public domain, will be seized upon by media more than eager to publicise them without waiting for the ECK.
In order to be useful, an evaluation has to be as objective as possible. Internal reviews are not necessarily frowned upon. In circumstances where the institution has as sensitive a mandate as the ECK’s, internal reviews may well be advisable. However, even in this case, there is merit in bringing in objective external actors (such as election management experts or representatives of other EMBs) who can facilitate open discussion of the issues under review.
A practice that may be useful is a post-election audit of the process of conducting elections, including a review of results in a specific number of constituencies such as has been carried out by IREC (discussed in detail in chapter 6). In some countries (for instance, the United States) these audits are conducted before results are made public. They may even provide reasons for recounts.
In Kenya, it would be important to have post-election audit mechanisms that enable the detection of anomalies at polling station level. As has been mentioned in other sections, voter turnout in the 2007 general elections was suspiciously high in a number of constituencies and it is necessary to use methods that allow audits at that level. This means that the electoral process should take into account security features and strategies designed to assist in detecting anomalies. For instance, it has been suggested that one possible stratagem would be to supply voter lists to polling stations without including the voters’ national ID numbers. When a voter presents his/her document, a clerk would enter the ID number both in a space left for such purpose in the voter list and on the ballot counterfoil. If a polling station is audited, the ID numbers on the list/counterfoil will be checked against those in the ECK’s possession, and irregularly added voters could thus be detected.
The issue of auditing, however, involves complex IT and other technical considerations, including sample design. While IREC could not undertake the full technical analysis required for such an exercise, it would recommend strongly that the ECK explore this and alternative approaches to audit.
Generally, internal audit mechanisms can provide valuable insights into how the process ran; the degree of compliance with the policies of the EMB and election law and the integrity of the results overall. Audits could be conducted either by the ECK itself or by a contractor, analysed for presentation at a retreat convened for the purpose and made public afterwards. Discussions would aim at identifying areas of weakness, recognising strengths, learning lessons and agreeing on internal reforms of election procedures as well as external reforms on which the ECK could advise, say, Parliament, with a view to improving future elections.
Findings
While the ECK has not conducted proper post-election audits in the past, it has for some time now held post-election evaluation meetings to assess performance and to learn lessons to be applied in future elections. A post-election audit of the election process could inter alia review results in a number of constituencies, remembering that it is important to use audit mechanisms that will permit the detection of anomalies right down to polling station level.
7.3 Post-election dispute settlement
During the 2007general election period in Kenya, a material contributor to the tension at KICC, broadcast live to the country, was the absence of an effective electoral dispute resolution (EDR) mechanism to resolve the mounting challenges to the integrity of the results from Kibaki strongholds. The response by ECK Chairman Kivuitu and Minister Martha Karua, directing challengers to the courts, merely served to exacerbate matters. ODM representatives, adverting to the appointment of five new judges a few days earlier, made plain their distrust of the judiciary and insisted on their challenges being resolved there and then, if necessary delaying the announcement of the final result.
There is indeed no satisfactory mechanism to deal with such disputes. Sections 10 and 44 of the Constitution, read with sections 19 to 23 and 28 to 30 of the National Assembly and Presidential Elections Act, dealing with challenges to presidential and parliamentary results, make provision for determination of challenges only after the result has been announced. Although there is provision (in the practice rules contemplated by section 23 of the Act) for expediting the judicial process in the case of election petitions, resolution of a dispute may eventuate only months or even years later, especially if there is an appeal. Challenges to civic election results are also possible only after the event (see section 16 of the Local Government Act).
Flexibility and expedition: It is unwise in principle to oblige disgruntled candidates and their supporters, convinced as they are that they have been wronged, to wait until after the prize has been awarded to their opponents and only then to start litigation – which could drag on indefinitely. If a quick and appropriate remedy is unattainable it predisposes to the kind of political crisis that occurred here. The blanket referral of such disputes to litigation makes it impossible to find and implement quick, sensible and fair solutions. Thus, for instance, if the mistake in Kirinyaga Central had been discovered on, say, 30 December 2007, after the form 16 had been delivered at the national tally centre and the result had become technically final, the matter could have been resolved there and then instead of taking its course through the courts. For this to be possible, there has to be a tribunal qualified and legally empowered to resolve such disputes quickly, fairly and practically, rather than on the basis of strict legality.
Competence: Many Commonwealth countries, such as Kenya, adhere to the quaint fiction that judges are generalists who can grasp and deal with any matter, however esoteric, provided it is competently argued. In the case of electoral disputes this attitude needs to be re-examined. The principles and practice of electoral administration have developed exponentially over the last two decades and a substantial body of international learning has been produced. All of this bears on dispute resolution and ideally requires specialised judicial attention. Because electoral disputes usually demand rapid resolution and do not allow time for extensive legal research by the adjudicating tribunal, familiarity with electoral law and practice is therefore a highly desirable attribute of such a tribunal.
A special electoral dispute resolution court: EMBs are a unique feature of modern democracies. They function, uniquely, at the interface between politics and law. The essence of an election is political, not legal. It is a political process driven by political forces for a political purpose: the allocation of political power within a given polity. Therefore, although an election must always be governed by law, the law should never be allowed to entangle an election in the inevitable niceties and rules, the procedures and precedents, the predilection for detailed analysis, for thorough debate and mature reflection that are the proud hallmarks of the judicial process. Electoral dispute adjudication requires flexibility and pragmatism, an eye to the political exigencies, sometimes even at the cost of strict legalism. Furthermore, and most importantly, in a highly politicised society such as Kenya, where judicial impartiality has already been doubted, the judiciary should not be unnecessarily exposed to the risk of being politicised, or being seen to be politicised, by its involvement in political disputes.
The type of electoral dispute resolution court: There is no universally accepted mechanism. Some states, especially older democracies in Europe, have left the task in the hands of administrative agencies within the government. Others, including Kenya, have consigned electoral adjudication in general to the ordinary judiciary, sometimes under adapted procedural rules. Countries such as Mexico, where there has been radical electoral reform comparatively recently, have established a special electoral tribunal with extensive power and exclusive jurisdiction in relation to elections. This may be costly but has the great benefit of speed and special expertise. Then again, South Africa opted for a special electoral court with exclusive and final jurisdiction consisting of three senior members of the judiciary who ordinarily continue with their routine work but are on standby to deal with electoral matters on an urgent and prioritised basis. Expedition has thus been ensured with the prospect of some degree of specialist knowledge.
The lowest level of competence rule: It is often not easy to decide who is to perform a particular adjudicatory task in terms of an EDR system, nor is there necessarily a one-solution-fits-all answer. An adaptation of a well-known principle of good management dictates that electoral disputes should be determined at the lowest appropriate point in the chain of authority, there, then and finally.
Adequate enforcement remedies: A crucially important feature of effective EDR is to ensure that every dispute is resolved in a manner that affords adequate redress and/or imposes appropriate penalties. Obviously a dispute is to be resolved in time for a remedy to be of value. One of the reasons for opting for extra-judicial EDR is because it is in the very nature of an election that, in order for relief to be effective, a decision often has to be given there and then. One of the most problematic aspects of current Kenyan EDR (which may have contributed to the eruption of violence) is that issues that could and should be addressed and possibly resolved by the ECK are consigned to the judiciary for determination in the fullness of time. The remedy must not only be timely, it must be fitting in other respects. The ordinary remedies at the disposal of the judiciary are limited. An EDR tribunal should have many more – and more varied – remedial powers than a court ordinarily has. It should also be armed with a much wider and more severe array of penalties than those contained in paragraph 8 of the Electoral Code of Conduct (Fourth Schedule to the National Assembly and Presidential Elections Act).
Consequences of failed remedies: Failure on the part of an EMB to afford an individual effective relief not only does an injustice to the individual, it also devalues the EDR system as a whole, eroding confidence in the EMB itself. Loss of legitimacy by any adjudicator is extremely serious, striking at the viability of the very adjudication process. In the case of an EMB, such loss of confidence in its EDR system is particularly grave. It functions in an arena where power is at stake, state power and control over national resources. Its loss of credibility has implications ranging much wider than the particular dispute or disputes the resolution of which (or, worse, the non-resolution of which) initiated the conflict.
Government’s role in establishing electoral integrity: It is trite that the force of a set of rules and the integrity of the system they aim to enforce depend in large measure on the certainty in the eyes of the public at large that transgressions will be investigated, prosecuted and appropriately penalised. In the case of EDR rules, this general principle applies with particular force, for it is usually public figures who are the transgressors or the beneficiaries of such wrongful conduct.
7.4 Swearing-in and assumption of office
It is standard practice to lay down in electoral legislation a period of some length between elections and winners’ assumption of office. In the well known case of the United States, the presidential election takes place on the Tuesday following the first Monday in November and the new president is inaugurated on 20 January the following year. Mexico has an even longer period: the 2006 elections took place on the first Sunday in July and President Calderón sworn on only on 1 December.
There are two reasons for this practice:
1) The first, which has been extensively discussed in the previous chapters, is to allow sufficient time for the solution of all pending electoral disputes and for the verification of election results. The period was used for such purpose in both countries mentioned above.
2) The second reason is that there should be a peaceful and orderly transition of power. For instance, a new president may have to select a cabinet, and the new members of the cabinet should be adequately briefed by their predecessors.
7.5 Custody of election materials
A successful post-election audit and analysis requires that all used election materials be safely stored after the elections until any analysis is concluded. In the Kenyan case this would include all materials in ballot boxes (including ballots, counting sheets, voter lists used during the election and all election result forms [16A, 17A, 16, 8 and 9]), the ECK master voter register with all voters’ information and checklists of all materials sent to polling stations, giving the serial numbers. To the list of materials should be added details of ECK election-day staff with information as to where they worked and of party agents with information about the parties they represented and where they were deployed.
Following the post-election analysis, certain documentation should be archived, where possible in electronic format.
7.6 Allocation of nominated seats (MPs and Councillors)
Nominated MPs: Nominated seats have existed in Kenya since before Independence in 1963. According to section 33 of the Constitution, the National Assembly includes twelve members “nominated by the parliamentary parties according to the proportion of every parliamentary party in the National Assembly” (section 33(3)). Section 33(4) stipulates that the proportions “shall be determined by the Electoral Commission after every general election”. Furthermore, section 33(1) indicates that these nominated members are “to represent special interests” and following section 33(5) the Electoral Commission ensures “observance of the principle of gender equality in the nominations”.
The constitutional disposition should have been complemented with a regulatory framework at a lower level. However, the Act on Presidential and Parliamentary elections is silent on the subject, and the ECK has never issued any regulation or statement concerning the process for the allocation of the nominated seats. The ambiguity is twofold. First, it is not clear whether the “nominated” seats are the equivalent of what is usually known as “reserved” seats which some legislatures establish to ensure representation of minority groups. Second, the formula for allocating the seats is not indicated either in the Constitution or in subsidiary legislation.
Even in the absence of specific regulations, the ECK has proceeded to allocate the seats. The subject is extensively discussed in annex 7.A, which shows that the ECK has not been consistent in the allocation of seats in the last three general elections, and that the allocation made in 2007 does not correspond to any of the usual formulas employed for the allocation of seats under proportional representation (Largest Remainders/Quota: Hare, Hagenbach-Bischoff, Droop or Imperiali; or Highest Averages/Divisor: D’Hondt, Sainte-Laguë, Imperiali or Danish). The ECK informed IREC of the formula used, based on the “best decimal rates” but, as shown in the above mentioned annex, even when that formula is used, the resulting allocation of seats does not coincide with the allocation decided by the ECK. The fact that different formulas allocate nominated seats differently makes it essential that the ECK specify the formula in advance of the elections, after consultation with all relevant stakeholders.
Although the Constitution does not define “special interests” as contemplated by section 33(1), the High Court, in the Il Chamus case, decided that they include those interests which have not been taken care of by the election process and which are vital to the effectiveness of the democratic elections in terms of adequate representation for all in a democracy. The ECK was presented with two choices. First, it should take into account the claim by the Il Chamus for representation when redrawing boundaries, in what the High Court considered would be a “reasonable” departure from population equality. Second, the court stated that “minority groups and all the other minority interests including the Il Chamus do constitute a special interest for the purpose of nomination”. According to the court, the ECK is constitutionally empowered to vet party nominations to ensure compliance with the special interest criterion and gender equality before transmitting names for appointment to the President.
The court’s decision placed the ECK in an impossible situation. Where should the line be drawn in terms of what constitutes a “minority interest”? Do the Dahalo who live along the Kenyan coast constitute such? Do the Suba, a community of 30,000, many of whom live on the islands of Rusinga and Mfangano on Lake Victoria, qualify? Even in the case of Il Chamus, identified by the court as a clearly defined minority group deserving representation, which party is obliged to include a member of that group in their list of nominated candidates? Do disabled people constitute a minority interest as a whole, or should blind people be considered a separate group? Do people with AIDS constitute an identifiable group with special minority interests? How can the ECK decide on the distribution of the twelve nominated seats to minority interests and then vet the lists? As a matter of fact, the ECK did not introduce into the vetting process any consideration of “special interests” with the exception of ensuring gender balance.
Bomas and Wako/Kilifi Constitutional drafts take different approaches, each with its own problems. The Bomas draft maintains the “first-past-the-post” (FPTP) or Westminster system for an undefined number of constituencies (their number to be defined by law), defines a different set of constituencies reserved for women (districts shall become special single-member constituencies for this purpose) and introduces fourteen members elected by marginalised groups (the seven groups mentioned include women, persons with disabilities, the youth, ethnic and other minorities, older members of society, trade unions and geographically marginalised communities). The elections for these groups “shall be by electoral colleges of the respective marginalised groups as provided by legislation”. The electoral management, if such proposal had been adopted, would have been particularly complex. Most voters would have had three votes in parliamentary elections: one for the normal constituency, one for the special women district constituencies and one as a member of a particular electoral college. Only middle-aged men without disabilities, not belonging to trade unions, ethnic or other minorities or to a geographically marginalised community would vote twice. It seems that these practical considerations were not taken into account at the time the Bomas draft was discussed and approved.
The Wako/Kilifi draft also maintains FPTP for an undefined number of constituencies (their number to be defined by law). Women would be elected from specially designed constituencies.
The nominated members would be divided into two categories: people with disabilities (one-third of them women) and special interests, of which only the youth and workers are mentioned. The number of positions for each group would be equivalent to 5% of the total membership of Parliament. There would be two additional groups. First, there would be additional members, “as may be required to ensure that no more than two-thirds of all the members of the Parliament are of the same gender”. However, these would be nominated by political parties in proportion to the votes received in each election. Furthermore, the President may appoint not more than 20% of his ministers from persons who are not members of Parliament (but with similar qualifications) and they will automatically become ex officio members of Parliament. As can be seen, the Wako/Kilifi draft is as complicated in terms of electoral management as the Bomas draft.
The case of the Bomas and the Wako/Kilifi drafts has been analysed in detail, in spite of the fact that neither of them is presently relevant, to point out the problems and the extremes that may be reached in the attempts to ensure the representation of special interests.
Before discussing the main issues and alternatives, it is important to clarify a few issues. First, the distinction between appointment and nomination: the non-elected members of Parliament might be appointed by the President or by Parliament, but the important issue is who nominates them. Second, the issue of non-elected members is closely related to the requirement that ministers and assistant ministers should be members of Parliament. This should not necessarily be the case, particularly in the case of a mixed system that combines features of a parliamentary system with a strong Presidency. Third, representation can be achieved without necessarily having the right to vote. The District of Columbia in the United States has representatives in Congress who do not have the right to vote, but can bring to the attention of the House issues that are relevant to the District.
The system as it existed before the Inter-Parties Parliamentary Group (IPPG) accord was of direct nomination by the President, at his pleasure. This approach, similar to that of Zimbabwe, is designed to reinforce the presidential majority in Parliament, or to compensate for losses at elections. It is one component of the approach taken by the Wako/Kilifi draft, which provides for making non-elected ministers (up to 20% of cabinet) members of Parliament. This approach flies in the face of elementary tenets of democracy and should be rejected, except where a non-elected minister has a voice but no vote.
The IPPG introduced a significant modification: the nominated seats would have to be distributed in proportion to the number of seats held in Parliament, and should represent “special interests”. This left two questions open: the formula to be used in the allocation (fully discussed in an annexed document) and the identification of the “special interests” to be considered in the nominations by political parties. The High Court has intimated that the ECK should vet candidates proposed for the nominated seats using the “special interest” criterion but, as has been pointed out above, this amounts to an impossible task. The High Court approach presents some significant problems. For instance, the fact that a party, perhaps under pressure from the ECK, appoints a person with some disability does not make that person a representative of the interests of disabled people generally. The same is true for the High Court judgment: if a party selects as a nominated member one of its own members who happens to belong to the Il Chamus ethnic group, that alone will not make that person a legitimate representative of the Il Chamus community.
The concept of reserved seats is introduced by the Bomas and Wako/Kilifi drafts in different ways. The concept is more clearly defined in the Bomas draft which would reserve fourteen seats for persons elected by marginalised groups: women, persons with disabilities, the youth, ethnic and other minorities, older members of society, trade unions and geographically marginalised communities. But the problems seem to be greater than those presented by the traditional nominated seats. First, the list in Bomas includes groups that cannot be considered marginalised. Women are under-represented in Parliament at present, but the Bomas draft contemplates the election of one woman per district – a total of some 150. Where is then the need for an elected two additional women as representatives of a “marginalised group”? The argument raised by the Il Chamus people was “that the likelihood (of having one of them elected) is so infinitesimal as to amount to an effective denial of the right of representation.” This is certainly not the case with older people – President Kibaki, a septuagenarian, is still in his prime. The idea that trade unions, usually powerful political players, are a “marginalised group” is simply risible.
Reserved seats for groups of people elected by their own pose a serious danger to democracy in its wider sense. In the case of a hung parliament, special interest representatives can cast the decisive votes, giving those groups a power similar to that held by the ultra-religious parties in Israel during certain periods – not an advisable example. One possibility that might be considered is to provide some “marginalised groups” (selected according to better criteria than those employed in the Bomas discussions) with non-voting seats in Parliament. This would ensure representation without introducing distortions in the democratic process.
Nomination of councillors: Section 26 of the Local Government Act (Cap. 265) provides that a municipal council shall include councillors elected for each electoral area by the electorate thereof, councillors nominated by the Minister to represent the Government or any special interests as the Minister may by order determine, and in certain circumstances a councillor appointed by a contiguous county council. There is a further provision that generally the number of non-elected councillors shall not exceed one-third of the number of elected councillors.
Section 29(7) provides that the Minister may, in respect of any county or town council, nominate the chairman from among the members of the council or persons qualified to be members of the council. Section 39 provides for the number of councillors for a county or town council and section 46 for an urban council.
The Act states that “the criteria and principles for the appointment of nominated members of the National Assembly under section 33 of the Constitution shall mutatis mutandis apply to the nomination of councillors” (section 26(2)). Section 33 of the Constitution specifies that those to be appointed shall be persons qualified to be elected as members of the National Assembly [section 33(2)] and shall be “nominated by the parliamentary parties according to the proportion of every party in the National Assembly, taking into account the principle of gender equality” [section 33(3)].)
Following every general election the proportion is determined by the ECK which then informs the political parties concerned of the numbers they are entitled to nominate: in response each party supplies the ECK with its list of nominees and the relevant information on each. The ECK then checks:
* the proportionate number of councillor(s) for nomination
o whether the person proposed qualifies to be elected as a councillor (e.g. is a registered voter of that local authority)
o • observance of gender balance and special interest
* that the nominee is able to read and write
If the ECK is satisfied with the information provided, it forwards the names to the Minister for Local Government for gazetting.
Sections 27(2), 40(1) and 47(3) provide that the term of office of a councillor shall be five years. The sections give the Minister the power to terminate the nomination of a councillor by notice in writing delivered to the councillor.
The role of the Minister in the nomination of councillors is limited to publishing the names of those proposed by the political parties. The Minister has no power under the Act to substitute others for party nominees or to nominate councillors of his own choice. Because the law empowers the Minister to terminate the nomination of a councillor, it seems that he may do this but has to revert to the ECK and eventually to the parties for the nomination of replacement councillors.
Following the 2007 election, the ECK obtained names for nomination from the political parties of the relevant local authorities and forwarded them to the former Minister for Local Government. He substituted his own nominees for some of the names he had received. Under section 29(7), the Minister is allowed to nominate the chairman of a county or town council from among the members of the council or persons qualified to be members of the council. This has been misinterpreted to mean that he may nominate another set of councillors whom he can then make chairmen of county or town council.
The Chairman of the Electoral Commission thus wrote to the Minister after he had gazetted the names of the nominated councillors, advising him that the law requires him to gazette only the names of the party nominees and does not permit him to remove the names submitted to him by the ECK or substitute other names. The Chairman requested the Minister to correct the error.
The councillors whose nominations had been gazetted in breach of the law subsequently had their nominations revoked by the new Minister who took over after the coalition agreement. The revocation was effected with due regard to the law in an attempt to correct the error. (It should be noted that the nominations that had caused the hullaballoo were not the nominations that the Minister is permitted to make of persons to represent the Government or special interests.)
The new Minister purportedly invoked his powers under sections 27(2), 40(1) and 47(1) to revoke the nominations of the councillors who had been nominated without due regard to the law. He did not, however, revoke the nomination of persons appointed as mayors or chairmen of town and county councils because, under section 29(7), the Minister can nominate persons qualified to be members of the council. The previous Minister’s nominations in this respect were not interfered with. The law as it now stands does not specify the circumstances under which the nomination of a councillor can be terminated.
The new Minister’s attitude was that, since the nomination of the councillors was illegal and void ab initio, he could revoke the nominations. The Court of Appeal of Kenya held, however, in Taib
A. Taib versus the Minister for Local Government (Civil Appeal 107 of 2006 (2007) eKLR) that the Minister has no power to revoke the nomination of a councillor once so nominated.
The Local Government Act vests the power of nomination or appointment in the Minister for Local Government but subject to the provisions of the Constitution. These provisions are confusing. The law allows the Minister to appoint persons to represent the Government in the local government and also empowers the local authorities to nominate some members. These, however, have to come from the list presented by political parties with representation in the relevant local authority. How one can select appointees of the parties to represent the Government is hard to know. Neither is it clear how the local authority itself, e.g. a county council, would appoint members to the municipal council.
The other grey area is the number of appointees and the interests they have to represent. The one-third of local government councillors who are nominated ought to represent special interests and also promote gender equality. As in the case of nominated members of Parliament, the ECK has not been very strict on this aspect. The political parties have continued to defy the law and the Constitution with impunity and the ECK has not made efforts to enforce these provisions.
7.7 Analysis of post-election procedures
A recurring feature of IREC’s public meetings around the country, and a topic touched on by more than one expert, was the perception of unseemly haste that pervaded the swearing-in of President Kibaki the evening of 30 December 2007. Such perceptions are extremely important. Elections are inherently divisive and the inauguration of the head of state offers a golden opportunity for a public display of united patriotism, unity in diversity at the commencement of the new term of office of the president and Parliament. It is therefore imperative that adequate time be allowed for the resolution of legal challenges and for political transitions to be harmonious. The current situation, where there seems to be doubt about the effect of the interaction between the provisions of sections 7 and 9 of the Constitution when a president is reelected, can and should be resolved in conjunction with related aspects of the post-election period and procedures. Of these the most important, substantively, is that Kenya combines two inherently inconsistent constitutional courses: retaining the British parliamentary prerogative of the President as the political head of the executive to dissolve Parliament, on the one hand, with a programmed handover of office by the President in his capacity as head of state. The constitutional review process might well wish to consider whether the time has not come to have a fixed term of office for both President and Parliament, allowing for a predetermined electoral calendar, from notice of dissolution of Parliament through to inauguration, with adequate periods of time for all the requisite formal and administrative steps to be routinely scheduled and dealt with. These could then be fixed in the Constitution, allowing, for instance, at least a month between the elections and the formal assumption of office for all disputes to be resolved. But whether or not so radical a change is adopted, the constitutional uncertainty needs urgent attention.
Recommendations
Recommendation on post-election audits and evaluations:
* IREC recommends that the ECK institutionalise the practice of post-election audits and evaluations, not only by continuing to hold them but also by improving the quality of objective data and other information (as well as actors) that feed into the evaluation exercise. This will enable the ECK to be a continuously learning EMB, whose processes are objectively reviewed and improved upon after every major electoral exercise, and at other instances as the EMB deems fit.
* IREC records a recommendation by a member that post-election audits be conducted by an independent auditor and made public.
* In order to facilitate the conduct of such audits, IREC recommends that the ECK publish on its website the results for all polling stations (forms 16A).
Recommendation on post-election dispute settlement:
• IREC recommends that in the course of the constitutional reform debate and in reviewing the legal framework for elections, provision should be made for the establishment of an appropriately composed and empowered special electoral dispute resolution court.
Recommendations on swearing-in and assumption of office:
* IREC recommends a transition period between a successful candidate’s election and his swearing-in.
* IREC recommends that the period provide sufficient time to resolve most electoral disputes and to verify election results.
Recommendations on custody of election materials:
* IREC recommends that the ECK develop procedures for safe storage of election materials until any post-election analysis has been completed.
* IREC recommends that the ECK store certain relevant election materials (such as the election results) in electronic format.
Recommendations on allocation of nominated seats:
o If it is decided to maintain the nominated seats, IREC recommends that an electoral formula (D’Hondt or Largest Remainders with a Hare quota offer two alternative
o options) be specified. In terms of the use of the nominated seats to support underrepresented groups or special interests, it would be preferable to devote all of them to improve the representation of women in Parliament. The number of nominated seats might be the same as at present, or Parliament might decide on an increase. Kenya might also explore alternative approaches (such as those adopted in Uganda and Rwanda) to increase the representation of women in Parliament.
* In the case of persons with disabilities and ethnic minorities, IREC recommends the introduction of reserved seats. Rather than preparing special rolls it might be better to ask organisations representing those groups to submit candidacies, with the final selection to be conducted by Parliament. Reserved seats should have a voice but no vote in the deliberations of Parliament.
* IREC recommends that, since there is no necessity for allocating reserved seats to women (who would be better covered by nominated seats and selection by political parties), to young people (some MPs are reasonably young), to other minorities (the concept might be somewhat difficult to operationalise), to trade unions (certainly not a marginalised group) or to geographically marginalised communities (another concept difficult to operationalise), this not be done
* IREC recommends that consideration be given to leaving the President a margin for appointing ministers who are not MPs. This would eliminate pressures on the use of nominated seats. Ministers who are not MPs might participate in Parliament with voice but without vote.
* IREC recommends the proper regulation of the matter of nominated councilors. It should be left to the ECK, applying a predetermined formula, to make the nominations from a list presented by the political parties. A minister who is a politician (and will always belong on a particular side of the political divide) is not expected always to act fairly/equitably in the allocation/nominations, as was demonstrated after the 2007 elections.
* IREC recommends that the Constitution or legislation clearly define what special interests or groups should benefit from nominations.
Recommendation on post-election procedures
* IREC recommends an amendment to sections 7 and 9 of the Constitution to eliminate all doubt as to the expiry of the President’s term of office.
* IREC recommends that consideration be given to amending the Constitution to provide for fixed terms of office for the President and Parliament together with concomitant provisions for an electoral timetable.
