E-voting with blockchain?
Voting is an intentionally high-friction endeavor. Elections are held at in-person polling centers, governed by strict voter ID laws, and designed to minimize fraud. This causes long queues, and voters may need to wait long hours to exercise their right and obligation as citizens. Elections are often executed using decidedly low-tech methods, such as paper ballots, which can be susceptible to user error and (accidental or deliberate) miscounts. Results can also take long to announce as it is very time consuming to count all ballots one by one. Moreover, elections can be costly. For example, in France, the presidential and legislative elections of 2012 cost 342M €, and around 40% of all expenses related to the national elections come from paper: campaign material production and mailing, ballots, transportation, etc.
Last, voting turnover is not very high in many countries (about 50% in the US), and it may be partially caused to the inconvenience of the voting process. If people were able to vote from their homes with the use of their smartphone, could we reach a voting turnover of +90%? Governments should naturally ensure the integrity and authenticity of the results. But how could they do that?
It is clear that efficiency is not the main point of the voting process, but security is.
Bringing voting into the digital age, by developing a technological solution that ensures security, transparency and support for new devices, seems that would solve the limitations of the current voting system. Can blockchain cover these technical necessities?
Privacy is a main requirement as votes should be anonymous to prevent coercion. On the other hand, e-voting should provide some sort of public verifiability, because otherwise, the provider of the e-voting solution — or someone who managed to compromise it — might be able to change votes at will. Voting from a digital device removes the need to physically travel to a polling station, which benefits minorities who have trouble accessing them. It also offers modern voters a method of participating which fits their everyday life. Finally, significant cost reductions come from the digitization of paper and manual processes, disappearance of physical polling stations to maintain and secure, as well as automation of auditing mechanisms.
Challenges
Cybersecurity experts oppose online voting of any kind because, they feel, it’s fundamentally insecure. Although a number of countries have embraced the practice, in 2015 a team of cryptographers, computer scientists, and political scientists looked closely at the prospect of internet voting in the US and concluded that it was not yet technically feasible. Protecting connected devices against hacking is hard enough, and, even if that could be achieved, developing an online system that preserves all the attributes we expect from democratic elections would be incredibly difficult to accomplish. Blockchain-powered e-voting introduces new security vulnerabilities and securing the vote count against fraud is more easily, simply, and securely done with other approaches.
Where are we at?
There is clearly some development to be done in order to run national elections with blockchain and allow all citizens to vote electronically with their smartphones. However, there are currently smaller projects carried by blockchain start-ups that are enabling e-voting with blockchain. Voatz, a Boston-based start-up, created an app for the state of West Virginia using blockchain encryption so that people can vote remotely and securely if a polling place or dependable mail services are unavailable. The Voatz system uses biometric authentication to identify individual users before allowing them to mark an electronic ballot, and the votes are then recorded in a private blockchain. The start-up says that in a general election pilot, its system would use eight “verified validating nodes,” or computers (all controlled by the company) that algorithmically check that the data is valid before adding it to the chain.
The Swiss startup Agora carried out counting in two districts for the national elections of Sierra Leone in March 2018. After the voting, a team of accredited observers from different locations manually entered approximately 400,000 ballots into Agora’s blockchain system. This test was considered a partial deployment of a blockchain, as the elections were only verified by blockchain, not blockchain powered. Agora described Sierra Leone’s elections as a “use case” rather than a “full implementation” of e-voting. This test also showed how blockchain e-voting can increase the speed with which votes are tallied, as Agora reported that it published election results on its website five days before the official manual counts ended.
Conclusion
Blockchain seems to be able provide the necessary to technology to have secure, transparent, and accessible national elections. However, it is not yet ready as security cannot be guaranteed: developing an online system that preserves all the attributes expected from democratic elections is incredibly difficult to accomplish.
Sources:
http://www.senat.fr/rap/r15-123/r15-123_mono.html
Do you need blockchain — Wüst & Gervais (2017)
https://slate.com/technology/2018/09/west-virginia-blockchain-voting-app-midterm-elections.html
