Every Vote Counts

An student issue analysis for MIS41240 at University College Dublin.

Written by Erik Obendörfer, Summer 2019. Dublin.

“Every vote counts” – a statement that is often made in the run-up to elections to encourage people to vote because, after all, just a few votes can be the difference between victory and defeat. However, there are alternative ways of interpreting these three words. It also implies that every vote was actually considered, was correctly assigned, and consequently contributed to the election result in the intended way. But there is no guarantee for that; we are compelled to trust the people and machines involved in the vote counting. In recent years, researchers have been concerned with the question of how elections can become more verifiable, and finally bring discussions about human error and fraud to an end. While this discussion predominately focuses on elections of the parliament, they often neglect voting that takes place in the parliament.

Voting behaviour is not communicated in a transparent way to the public

This neglect is peculiar, given that the public only votes for the parliament every so often but parliament votes for the public every day. In the case of the European Parliament which is currently on top of our agenda given the upcoming election, such voting concerns the regulation of the air we breathe, the food we eat and the money we spend. Taking into account the significance of the voting that takes place in the European Parliament, the public should have the right to retrace the voting behaviour of lawmakers. They were the ones who put trust in them and made them their representatives in the parliament.

One way to learn more about the election behaviour of one' s representatives are protocols that are created for each voting. However, this solution is not consistently convincing considering that these protocols can be changed in hindsight, although the corrections do not affect the outcome. It can be argued that the EU parliament’s protocol does not give an honest representation of the voting that takes place. For example, in March 2019 ten members of the European Parliament (MEP) stated that they accidentally voted against an amendment of the controversially discussed articles 11 and 13 (also referred to as “link tax” and “upload filter”). In such a case, MEPs have the chance to change their decision in the protocol for recording the voting. Consequently, it now says in the protocol that these ten lawmakers approved the amendment. However, these changes are of rather symbolic (critics may say misleading) character because it does not affect the outcome of the voting. It aims to enable MEPs to express their real intention, but at the same time to prevent them from being put under pressure and changing their vote afterwards for this reason.

In this specific case, the end result would have been different if the MEP had voted correctly in the first place. In other words, there would have been a majority for the amendment of the draft law concerning such important areas of our lives. This type of instance occurs regularly for reasons such as delayed translations, pressing the wrong button, or just being distracted. Critics such as Magnus Andersson, leader of the Swedish Pirate Party, take it a step further and question whether it actually was an accidental mistake. Andersson expects some of the MEPs voted as they intended and describes the subsequent changes to the protocol as “a way to get away with how they voted”.

The general issue here is that voting behaviour is not communicated in a transparent way to the public. Numerous solutions have been proposed to solve this issue. One proposed solution is that one could simply change the rules concerning the protocol. Given the significance of the issue at hand and the breach of trust that it entails, a solution that offers complete transparency without relying on the previously faulty protocol may be preferred. As such, adequate solutions could be the introduction of a traditional database or a blockchain-based voting system. While both options could ensure long-term tracking, the latter option seems to have more potential regarding immutability of the information and will be elaborated on in the following. A prominent example for the use of blockchain is Bitcoin, and it will therefore serve as a starting point to briefly explain what a blockchain is and how a voting system based on it might look like.

In Bitcoin, transactions are recorded on a distributed ledger, validated by miners and then grouped into blocks. Each block is chained to the one before and after it, creating a secure sequence of blocks, the so-called blockchain. Using this particular technology enables long-term tracking and unalterable record of transactions as well as pseudonymisation. Considering the characteristics of blockchain, it seems reasonable to apply aspects in the context of parliament voting. Along with that, several critical design challenges need to be addressed, mainly with regard to transparency and the type of distributed ledger.

In the Bitcoin ledger, for example, identities are pseudonymous, and only transactions are entirely transparent, while the proposed voting register would have to work with both, transparent identities and transparent transactions. Furthermore, another difference to the Bitcoin blockchain would be to ensure that only elected representatives can cast their vote. Bitcoin is an example for the implementation of a permissionless blockchain because, theoretically, anyone can join the network, and thus make transactions and/or validate the ledger. For the proposed blockchain-based voting, a permissioned and public system seems more suitable, meaning that only elected MEP can make transactions (vote) that would be verified by an EU central authority. Further, decisions should be verifiable for the public, and therefore accessible on the blockchain (see “White Paper: Blockchain Beyond the Hype”). While this is the main feature of the proposed solution, a lot of other design principles of this solution to non-transparent voting remain open to discussion. The initial idea clearly demonstrates that cryptographic tools do not necessarily lead to increased secrecy but can help in the establishment of open, transparent societies by empowering the people with long-term tracking of their representative’s election behaviour.