Getting voters to turn out for the ongoing biometric voter registration (BVR) exercise is the puzzle before the Zimbabwe Electoral Commission (Zec), political parties, civil society organisations (CSOs), development partners (like the United Nations Development Programme (UNDP)) and citizens at large.
Innocent Batsani Ncube,Scholar
According to figures released by Zec, the district-based registration phase that began with “pomp and fanfare” at State House, netted only 40 000 registrants in almost one month. The “blitz” has also started on a low note and the signs of voter registration apathy loom large. This puts Zimbabwe’s ability to hold credible elections that will address legitimacy issues at risk.
This article seeks to contribute to the discussion towards solving this puzzle through providing a commentary undergirded by credible empirical research inferences. It is not prescriptive, but following an explanatory approach to articulate from an empirical perspective why and how voters turn out.
Specifically, five voter turnout models, namely, rational choice, mobilisation, psychological, political-institutional and socialisation models, will be discussed from both a conceptual and practical Zimbabwe perspective. The writer has drawn these models from an excellent 2012 meta- analysis study conducted by Kaat Smets and Carolien van Ham on micro-level or individual level motivation for voter turnout.
Their “analysis of analyses” type study involved reviewing 90 empirical studies of individual level voter turnout in national elections published in top 10 journals in the world from the period between 2000 and 2010. The limitation of using this best practice is that the studies are mostly from established democracies and an argument of context may arise. This, though, is easily refuted because there are standard electoral practices the world over — what is good for other countries in terms of fulfilling democratic participation should indeed be good for Zimbabwe as well. This is the principle that flows in this article. The importance of this study in the evidence-based argumentation I will be making cannot be over-emphasised.
The rational voter
We will begin by analysing rationale choice voters; who are by far the widely recognised both in practice and in theoretical postulations. This type of voter is a person who when confronted with a decision that affects their interests does a cost-benefit analysis and is interested in maximising gains.
For this rational voter, the rewards of voting or in this case registering to vote, is a function of the benefits of voting and the perceived probability of casting a decisive vote. Hence, a rational voter will register when the benefits and perceived probability of casting a decisive vote are larger than the costs. If the costs outweigh the benefits, it then follows that the rational voter would decide to stay at home.
Three key findings from Smets and Ham’s study are instructive for the Zimbabwean case. One, higher costs of voting are found to decrease the possibility of turnout. Two, contrary to popular belief, the national economic situation as well as the evaluation of one’s economic situation fail to predict turnout in most studies. Three, Positive evaluations of parties do seem to boost turnout as it has a statistical success significance rate between 67% and 82%.
The mobilisation model is rooted on the idea that parties, candidates, interest groups and social movements mobilise prospective voters to take part in the political process. These social networks significantly reduce the cost of participation by providing information about parties, candidates and the electoral process.
Associational life is seen as an important catalyst in consolidating values and this is presumed an important mobilising effect. Smets and Ham’s findings on this element highlight the test of effectiveness of non-partisan and partisan mobilisation.
Before I mention their findings, it is important to highlight that in Zimbabwe, more emphasis in terms of funding and programming has been concentrated on non-partisan voter education and mobilisation. Parties usually do their own mobilisation programmes ad hoc in a manner that does not attract support from the big development agencies that support CSOs in this work.
Firstly, Smets and Ham find that partisan and non-partisan efforts boost voter turnout, but are mediated by a more general inclination to vote and, hence not expected to affect all prospective voters in the same manner. The crucial inference for Zimbabwe is that in their aggregated analysis they find that success rate is higher (70%) for partisan mobilisation than for non-partisan mobilisation (56%). Food for thought.
The model stresses the role of attitudes and psychological predispositions such as political interests, partisanship and political efficacy. This is the most difficult yet an important angle of understanding the individual level decision-making of a voter.
The movement of the brain cylinders, the mental ferment and firing internal decisional pistons is truly a very personalised phenomena. These different cognitive elements are projected to act as arsenal for reducing costs of voting and increasing turnout.
The different ideological tastes are expected to increase the inherent benefits from the act of casting a ballot or registering to vote. It is assumed that there is a positive relationship between politically active/interested citizens that are involved and their confidence level in their ability to have influence on the macro-political system.
Taking it a step further, personality characteristics are important cues in explaining the extent to which people engage in selfless disposition such as the act of voting as a civic responsibility. Smets and Ham do indeed find a high success rate (72%-85%) in establishing a link between party identification and political interest.
One important myth that is dispelled in the meta analysis is that satisfaction with democracy increases turnout — the study finds this to be statistically insignificant.
This model sees turnout as a consequence of the political and institutional context in which citizens live. By far the most focussed on model by CSOs and the conventional electoral technical assistance framework provided by UNDP, among others. Two key variables were used by Smets and Ham to analyse this model. The first is closeness and concurrence of elections. For this article I will just focus on the concurrent election part — which Zimbabweans are familiar with — the harmonised elections.
Harmonised elections have been assumed to contribute to an increase in turnout (this is mainly because of the increased party mobilisation, sustained campaigning, and concentrated media spotlight). The study finding does not confirm this hypothesis. They find that holding concurrent elections does not influence turnout.
The second component is the legal framework — a commonly used term in electoral management and advocacy — Smets and Ham refer to this as voter facilitation rules. The legal framework is thought to influence turnout. In Zimbabwe it has acquired legendary status because of its centrality to the electoral contests. The political opposition is known for demanding electoral reform. Smets and Ham refer to vote facilitating rules as institutional measures to motivate and mobilise people.
Examples that are relevant to Zimbabwe include holiday or weekend voting, the placement of special polling booths (for example in and around shopping centres and spreading elections/ registration over a certain period. The purpose of this is to significantly reduce the cost of voting. The current narrative in Zimbabwe includes demand for further voter facilitation with the latest prospective incentive being the possibility of Zec registration officials serving as commissioners of oath for authenticating proof-of-residence affidavits.
The results of the meta-analysis are less conclusive regarding the voter facilitation rules with about half of the tests and studies falling into the category “success”and the other half in the category “failure”. However, the good news is that the average effect size is positive and significant.
Socialisation model and conclusion
I will conclude this article by discussing the socialisation model. This is mainly because of its strong appeal and potential long-term solution to our turnout quagmire.
It is held that citizens form the basis of their political attitudes and behaviors during the impressionable or rather formative years between the childhood and adulthood period. A number of socialising agents that include the family, peers, school, mass media and the political context play a facilitatory role to the political learning trajectory of a voter. Political discussion has been emphasised by social exchange theories as important in persuading people to participate in politics. The logic being that discussion leads to higher levels of interest and political knowledge emphasising social norms inclusive of turnout in elections, additionally due to proximity, this may also induce norm-conforming behaviour. Results of Smets and Ham’s meta-study are that half of the tests have a positive impact.
The road to voter mobilisation is arduous, it needs all key stakeholders to reflect on these models and learn from and test good practice. We are a young democracy, which should continuously evolve its theory of change. Recourse to tested knowledge and solid evidence should guide our electoral policymaking and advocacy. This will make us solve the elusive voter turnout puzzle.
Batsani Ncube is a Chevening Scholar reading Elections, Campaigns and Democracy at the Democracy and Elections Research Centre, Royal Holloway, University of London. These are his personal views. — firstname.lastname@example.org