Projektstatus: Idea Release In use

FairElection

FairElection creates a tool for political organisations to select candidates according to their self-chosen criteria of diversity or the diversity of the Swiss population. Everyone can use the tool to simulate the results of a past election by modifying these same diversity criteria.

#Round 1 

In a nutshell

The project “FairElection” aims to create a tool for participation within Swiss politics. The tool offers two functionalities, corresponding to its two target audiences.  

For the general public, “FairElection” offers a tool for political simulation. By integrating existing data from past federal and cantonal elections, it allows everyone to “play” with certain representation criteria (e.g. age, gender, …) in order to visualise what the elective bodies would look like if other criteria had been taken into account. 

For political parties and organisations, “FairElection” offers a tool and a method to develop a process for internal (“primary”) elections or, more generally, to conduct a discussion around the desired representation within the party. Members, supporters and/or the general public can thus choose their criteria for representation and then use them for an internal election. 

The problem we want to solve 

On the evening of the election, a large portion of the population do not recognize themselves in those elected. This impression of a lack of representation has destructive potential to democracy, by calling into question the bonds of trust necessary for it to function properly. Political parties are often seen as carrying a heavy responsibility in this situation. Candidate lists are not sufficiently representative and the nomination processes within parties, sometimes complex, are presented as the ideal culprits.  

Furthermore, at all political levels, representation biases continue to favour certain groups of people (in particular certain genders, ages, and socio-professional levels). Within the parties, candidates are far from having the same opportunities when it comes to drawing up electoral lists, but also during the formal election. The question of fair representation of an elected body – legislative, executive – should be at the centre of public debate: what representation do we as citizens want? 

Our idea in more detail 

The FairElection tool will support political parties and organisations by monitoring internal elections and drawing up lists of candidates. The method works in two steps. First, members, supporters and/or the general public can choose their criteria for representation. What criteria will such a list have to meet? Parity, representation of the generations, place of residence, level of education? 

A first vote is organised to choose these representation criteria. Secondly, these chosen criteria are applied during an election. In a second vote, voters choose their favourite candidates. The freedom of choice of voters is not limited. They themselves do not have to create the perfect list that meets the criteria. It is the FairElection algorithm that ensures that the criteria chosen in the first vote are applied to the election result. The group of winning candidates will therefore be the one that satisfies the chosen criteria while respecting the democratic choice of the members.  

This way, the winning group of candidates will be satisfying the criteria chosen by the user, while respecting the democratic choice of members. The algorithm provides a mathematical guarantee that the winning group is the one that obtains the most votes while respecting the criteria. 

The mathematical method was developed together with the EPFL for the primary of the “Appel Citoyen” movement in Valais in 2018. This approach has been the subject of several academic contributions, underlining its character as a technological novelty with a high potential for disruption. This short movie explains the way the algorithm works.

This tool offers a solution to the following challenges: 

  • It offers parties a tool to bring transparency and trust into the selection process of candidates. The discussion on criteria of representation is distinct from the nominal choice of candidates, which allows for quality participation: first the criteria, second the people. 
  • It offers party members, party supporters and the general public (if the party holds open primaries), the opportunity to choose a fair representation and determine the desired criteria. The discussion on representation becomes a public discussion, thus contributing to stronger legitimacy of candidate lists and party elections. 

By making the implementation of representation choices easier and more transparent, the tool contributes to equal opportunities in politics. Candidates from under-represented groups have a better chance of being nominated as candidates.  

For the general public, FairElection also includes a political simulation tool. By integrating existing data from past federal and cantonal elections (and communal elections where possible), it allows everyone to “play” with certain criteria, to imagine what the legislative bodies could look like, if other criteria had been considered. For example, citizens could run a political simulation for the 2019 federal elections, and ask themselves who would have been elected, if a gender parity criterion or an age criterion had been applied. The tool FairElection shows the face of this fictional parliament. Citizens can immediately visualise the effect of a criterion based on real data. 

This second dimension of the tool is a response to two challenges mentioned above: 

  • The sense of inevitability diminishes. Thanks to political simulations – for example of an “alternative” federal parliament – the general public becomes aware that different representation is indeed possible. 
  • The discussion about distance between the population and candidates becomes more objective. It makes it possible to highlight which sections of the population are not represented in the election process. 

The team 

The following people make up the team supporting this project: