Saturday, January 29, 2011

Kenya’s Electoral System: Failed Thinking in the New Constitution


Sometimes in 2010, Najib Balala, one of the leading luminaries of the Orange Democratic Movement (ODM) apologized to President Kibaki for Moslems' rejection of Kibaki’s candidature in 2007. One of the questions raised by Balala’s ‘apology’ to Kibaki  is the extent to which a vote in the first-past-the post electoral system used in Kenya in 2007 is a true expression of a voter’s real preference. Put differently, does a system that assigns equal weight to every vote without measuring the intensity of a voter’s preference generate truly democratic outcomes?

This matter is relevant to the Kenyan scenario. In 2007, three strong candidates-Kibaki, Kalonzo and Raila- vied for the presidency, and between them 99% of the votes cast were divided. At the end of this heated election, thanks partly to an inefficient and compromised electoral body but also due to the fact that only a few thousand votes separated Raila and Kibaki, it was near impossible to determine an outright winner. Could Kenya have avoided the violence that ensued by the adoption of a different electoral system? The answer to this question will, in my view, explain Balala’s apology.

It will be recalled that in the run-up to the much maligned elections, proposals to reform the electoral benchmarks were tabled by the Paul Muite-led Parliamentary Committee on Justice and Constitutional Affairs. Essentially; these proposals would have required that a president elect secures a supermajority. Apart from garnering a simple majority victory and at least 25% of the votes in five provinces as currently mandated, the proposals required a president to be additionally voted for by at least 50% plus 1 of the voters.  Absent this threshold, a run-off between the two best performers in the general elections would ensue. This is one approach which has worked in emerging democracies including Ghana and Sierra Leone. More established democracies including France also pursue this system. This system is however arguably expensive considering that a fresh election would have to be organized. Further, it could exacerbate existing tensions to manic levels especially in the transitional period between the initial election and the run off. In an atmosphere polluted by the politics of ethnic balkanization such as our own, there is no guarantee that a run-off outcome would not suffer similar credibility crisis as did the 2007 elections. Moreover, institutional failures both at the electoral body and within the broader security and justice sectors could undermine the run-off.

A more efficient system can however be created; one in which the run-off scenario rather than actual is merely simulated. This can happen if Kenya adopts an alternative vote system which gives a voter the opportunity to rank, not only their number one candidate, but also their number two, three and so fourth candidates. In a context in which the votes are too close to call or where none of the candidates receives a supermajority,  the electoral body would simply simulate a run off by carrying out an analysis of not just the number of votes received but also their intensity. In the 2007 elections for instance, the electoral body will have considered how many of Kibaki’s voters had Raila as their number two preferred candidate; how many Raila candidates had Kibaki as their number two preference and how Kalonzo’s voters viewed the two. Indeed, the current coalition arrangement is intuitively a fair representation of such an analysis, albeit outside a coherent constitutional framework. By choosing to support Kibaki over Raila in the tussle for the presidency, Kalonzo’s posture suggested that his supporters ranked Kibaki higher than Raila, and would have voted for him in the event of a run off. Hence one way of rationalizing Balala’s assertion is to argue that even though Muslims largely voted for Raila in 2007, Kibaki was their second close preference, and could have received their backing in the event of a run-off. Admittedly, the alternative vote electoral system is also not fool proof, but at least it does capture the question of intensity of voter preferences, which bears on the quality of the democratic choice. Unfortunately the new constitution adopted in August 2010, adopted the 50% plus 1 rule which although better than what pre-existed it, is costly in monetary terms as well as in relation to instability during the period between the first and runoff elections.

Balala may be a political turncoat or a politician with an axe to grind, but his recent apology to the president on behalf of Moslems suggests that Kenya should think seriously about its electoral system. Unfortunately, the new constitution did not go beyond the obvious dichotomies of choices between a presidential and parliamentary system. Instead, one would have hoped that the search for options for deepening democracy should additionally focus on electoral arrangements that will most answer to a country that is so diverse and politically polarised. It is now for the judiciary to interpret the new constitutional provisions using a more rational lense.

In the end, the pure majority rule that now obtains cannot be reconciled with the idea of maximizing social welfare because it ignores the preferences of minorities. It is no accident that long lived democratic systems deviate from this unalloyed version of majority rule by having several layers of representation and by requiring, in numerous settings, supermajority or near consensus.