To approve or not to approve: this is not the question

Authors: José Carlos R. Alcantud and Annick Laruelle

José Carlos R. Alcantud is a professor of Economics at the department of economics and economic history and a member of the multidisciplinary business institute (IME) of the University of Salamanca

Annick Laruelle is an Ikerbasque research professor at the department of Foundations of Economic Analysis I of the University of the Basque Country (UPV/EHU).

Voice and exit are often alternative ways of exerting influence, but with regard to voting the exit option spells no influence; only voice can have an effect[…]

A. Lijphart ¹

In polls many citizens express some dissatisfaction with politicians. Usual ways to voice this dissatisfaction in elections are absenteeism, spoiled or blank vote, or voting for a political newcomer or a fringe candidate (Mickey Mouse option). No legitimate and explicit negative option is generally offered to electors. In consequence the opinion of unsatisfied citizens does not affect the result of elections.

Most of the times blank votes are not tallied into final results and in that case they are spoilt votes. The objective of the recently created “Movimiento Ciudadano por el Voto en Blanco Computable” is to compute the blank votes and to convert the seats that would correspond to protest votes into empty seats. This Spanish movement intends to make dissatisfaction count. However their proposal treats equally blank votes cast by electors that reject the political system, or that are dissatisfied with the candidates, or even with the current electoral system.

To better identify blank votes that express dissatisfaction with the choice of the candidates, the NOTA (“None Of The Above”) organization proposes to give voters the ability to withhold consent (as an expression of their legitimate consent in an election). They suggest to add to the list of candidates a “None of the Above; For a New Election” option. In case this option received the larger number of votes, no candidate would be elected and a follow-up by-election with new candidates would be held.

More generally the following questions are worth being addressed: why our democracies do not allow expressing disagreement in an explicit way in elections? Why can we only vote in favor of a single candidate (or a list of candidates) and not against a candidate?

Some historical examples tell us that there exist (or existed) electoral systems where negative vote also counted. The Greek institution of Ostracism permitted citizens to veto politicians (and to exile them). In the electoral systems introduced during the Perestroika voters were asked to cross off the names of the candidates against whom they wished to vote. A similar rule is used in some Chinese village elections. In the State of Nevada the “None of the above candidates” option has been implemented.

Similarly, why voters are asked to vote for one single candidate? Why can’t voters express their opinion on every candidate? If one candidate will be the head of the government of all of us, shouldn’t all of us express our opinion on this candidate (thus on all of them since he or she is not known beforehand)? This is what proposes the “approval voting method”, which is used in some scientific societies like the American Mathematical Society, the Mathematical Association of America, the American Statistical Association, the Institute of Electrical and Electronic Engineers, among others. Each voter is asked to state which candidates he could stand to see win; such candidates receive his approval. The candidate with the highest number of approvals is elected.

A natural extension of the approval voting method appears when the option of disapproval is added in order to obtain what can be referred to as “dis&approval voting” ²: for every candidate the voter can cast a positive vote (if the candidate is considered as appropriate), a negative vote (if the voter is against the candidate) or a null vote (if the voter does want to abstain on the candidate). The candidate who obtains the highest difference between the number of positive votes and the number of negative votes is elected.

What are the advantages or this system? ³ If voters have more options they should be more interested in participating in elections. In particular voters who are unsatisfied with one or several candidates should have the possibility to express such opinion and this should increase turnout. And this is an important benefit because electoral participation is at the heart of the democracy. This system is more likely to give the small candidates their proper due: the common dilemma of the “useful vote” (a person voting for an option perceived as having a greater chance of winning over her/his preferred option) would disappear. The voters could give a positive vote to both the weak and strong candidates. Finally this system should favour “compromise” candidates. To be elected a candidate will have to convince most citizens, not only her/his own supporters.

Is this a utopia? Is this system difficult to understand? In the last French presidential elections (2002, 2007 and 2012) frame field experiments were conducted ^{4 5 6 7}. Electors were proposed to partake in an experiment on the very day of the elections. They were asked to vote again with alternative systems (instead of the traditional runoff system). The experiment was well received by the population: 80% of those who were asked agreed to participate. The alternative systems, including approval voting and dis&approval voting, received a positive evaluation. In particular the dis&approval voting system was reckoned to be very natural.

In Social Choice theory it is well known that the voting system matters. The choice of candidates depends largely on the electoral system. This was illustrated in the 2007 experiment: François Bayrou received more approvals than the two candidates who arrived at the final round-off of the elections, to wit, Nicolas Sakorzy and Segolène Royal.

More experiments will be necessary to test the practical properties of the systems. How will voters react to a new system? Will political parties present more consensual candidates? The theoretical properties of these systems are also worth investigating in order to argue on their virtues and vices.