The power of elections

What the election skeptics underestimate is the fact that there can be no democracy without elections, and that formally correct elections are vital for the legitimacy of any democratic regime. Staffan Lindberg (2006) underpins this argument in his detailed study of elections in Africa from 1989 to 2003. On the basis of systematically analysed data from 232 multi-party elections in 44 countries, he draws far-reaching and well-based conclusions about the “power of elections” (as the original title of his study read):
– Elections are not the culminating points of democratic regime changes; rather, they promote the liberalisation of regimes.
– Elections are self-reinforcing in the sense of becoming better each time they are held. Statistics show that regime collapse becomes rather unlikely after three multi party elections.
– Elections help civil and political liberties to become more firmly institutionalised.
Lindberg rightly speaks of elections having a “self-improving quality”. What is often dismissed as “mere procedure” actually does have an intrinsic value and quality.

A weak point, however, is Lindberg’s conclusion that better elections improve the quality of democracy per se. This monocausal argument may well be a fallacy. For one thing, it ignores the years between elections in which the democratic character of a regime needs to be proven. For another, quality and consolidation of democracy do not depend on elections alone, as many other factors matter too.

The significance of Lindberg’s work extends well beyond its immediate topic. He shows that politics in Africa cannot be described only in terms of arcane informal practices, as prominent colleagues like Patrick Chabal, Jean-Françoise Bayart and even Göran Hyden still do. Even in Africa, formal institutions have begun to play a role.

Little inclination for reform

Lindberg’s position is indirectly supported by Christof Hartmann (2007), who examines electoral reforms in Africa in a book, which I co-edited (2007). Hartmann finds that such reforms tend to be rare and that a fundamental switch from first-past-the-post to proportional representation was only made in four countries: Namibia, South Africa, Sierra Leone and Liberia.

This finding comes as a surprise. After all, the many regime changes from single to multi-party systems in Africa seem to suggest that electoral systems changed as well – not least, since this is often something opposition parties demand.

Nonetheless, the desire to embrace reform is limited; and Hartmann provides two main reasons:
– widespread ignorance of possible alternative systems (especially under the intense time pressure of a transition process) and
– marked uncertainty about electoral reform because no one involved can foresee the impacts on the balance of power.

Even opposition parties that originally call for reform often change their mind after discussing the details and coming to believe that the existing voting system will work for them once they come to power. It is telling that radical changes of election systems occurred in countries which had time to check alternatives and – with the exception of South Africa – were under intense international pressure.

All this should be a warning to those who see the switch to proportional representation as a quick and easy solution to representation problems in multi ethnic societies. What they generally underestimate anyway is that simple majority voting is largely confined to Anglophone countries.

Violent politics

In the same book, two essays by Andreas Mehler and Liisa Laakso focus on a barely researched but frequently observed phenomenon: violence in and around elections. Mehler’s explorative contribution shows that the major perpetrators of violence in the context of elections are generally political parties trying to influence the outcome. Moreover, such use of violence is by no means confined to revolutionary “anti system” parties or “normal” opposition parties. Rather, it is also a tool used by parties in government; and the latter typically have an advantage as they wield state powers in a partisan way.

Liisa Laakso analyses the targeted use of violence by government and opposition parties, assessing the cases of Kenya (1992, 1997, 2002), Tanzania (2000) and Zimbabwe (2002, 2005). She finds that electoral violence rarely occurred spontaneously. In most cases it was planned and orchestrated; and parties’ youth organisations often played crucial roles. The events surrounding recent elections in Kenya and Zimbabwe clearly confirm Laakso’s observations.

Comparing systems

Dieter Nohlen’s “Wahlrecht und Parteiensystem” (“Electoral Systems and Party Systems”), now in its fifth edition, is built on a much broader base of corroborated findings. It is the ultimate textbook on the subject, unmatched even internationally in terms of systematic coverage and global empirical scope. Nohlen provides essential information about the technical details of voting systems and their political implications. He shows how new systems are again and again engineered by combination of various familiar elements. Nohlen does not focus exclusively on the Western world but addresses specific African, Asian and Latin American issues in intraregional comparisons.

The discussion of the political implications of the various voting systems is instructive, revealing that many generally accepted effects of particular systems are not borne out by historical/empirical evidence. Even in its most radical form, for example, a first-past-the-post voting system (a plurality voting system in constituencies with only one member of parliament) does not necessarily prevent fragmentation of the party system. Conversely, it is possible for systems of proportional representation to have only two or three political parties. In different societal and politico institutional settings, the purported pros and cons of electoral systems can have totally different impacts.

Nohlen’s core thesis is that the impact of an electoral system depends to a large extent on the societal context. How an electoral system is rated depends on the criteria used for assessment. Is greater importance attached to representing sociopolitical diversity accurately, or does it matter more to easily achieve a stable majority?

In contrast to prominent scholars of party and voting systems like Giovanni Sartori and Arend Lijphart, Nohlen shows that there is no such thing as the best voting system. What matters is that the electoral system should suit the societal context. For consultants active in the field, this implies focusing on the actual environment and taking into account as many of the various interlinked factors as possible.

The key consideration, therefore, should not be the popular demand that any electoral system should be simple – a call that is not directed at countries with low education standards only. Indeed, every voting system is designed to help to resolve complex political problems. To do so, it needs to serve a number of functions, which can often be guaranteed only by laboriously putting together a variety of elements of electoral law. This is precisely why new voting systems keep being invented; and there are by now many examples that prove that, even in supposedly “underdeveloped” countries, voters cope with complicated voting systems.