Voting security has become a significant issue in many countries. Some have argued blockchain voting systems could solve election security problems and reinvigorate democratic governance. In this article, I explore the problems with traditional voting methods, the engineering of the blockchain, its benefits, and current work being done on blockchain voting. I will then suggest the best, most realistic implementation method for blockchain voting. To conclude, I address challenges to the widespread adoption of blockchain voting methods, in particular, why some governments might not be in any hurry to embrace the change.
The Limits of Current Voting Systems Threaten Democracy
Democracy is a system of governance whereby those who are governed participate in the process, typically through elected representatives. For democracy to function, fair and secure voting methods must be in place. While traditional paper-based ballot voting is accessible and inexpensive, it has two significant problems:
- It is not scalable. Paper-based voting is best suited to smaller local forms of government. There are significant challenges with ensuring accuracy when the method is used on a large scale.
- Increasingly, paper-based voting is combined with electronic voting machines, which are much less secure, introduce multiple security vulnerabilities, and make election manipulation easier.
Some argue blockchain technology can solve these problems. First, because it is massively scalable, and second because it is tamper-proof. Is this true? Could blockchain technology offer a safe and reliable voting method suited to the needs of modern democracy?
The Engineering Behind Blockchain Voting
To understand blockchain voting requires understanding how the blockchain works. While the hype surrounding Bitcoin seems to have waned, governments and businesses are becoming more interested in its underlying technology: the blockchain. It has been claimed the invention of the blockchain is as important as the creation of the Internet.
At its core, the blockchain is a distributed database that maintains a secure and ever-growing ledger of records (usually transactions) known as blocks:
“Picture a spreadsheet that is duplicated thousands of times across a network of computers. Then imagine that this network is designed to regularly update this spreadsheet and you have a basic understanding of the blockchain.”
A network of nodes manages this database (the blockchain). The nodes are merely computers connected to a network that have agreed to process the validity of transactions on the blockchain. Once a node validates a transaction, it adds it to a chronological group known as a block, which is then added to the blockchain. Valid transactions are grouped and added to the database in a block, one after the other, hence the name blockchain. Data on the network is public and can be accessed and checked by anyone. This method of data storage is incredibly versatile. The type of transactions stored in the database can be anything, cryptocurrency transactions like bitcoin, health records, land titles, supply chain records, copyright and royalties, or votes.
The Blockchain is Transparent and Incorruptible
To understand why the blockchain is tamper-proof with requires going a bit deeper into the technology. When the first block of the chain is added, it is marked with a hash function. As the second block is added, it is also marked with a hash function that contains part of the first block’s hash function. Therefore, when a node submits a new block to the chain, if the node has changed any of the database transactions included within the previous blocks, the hash function of that block would also need to be changed. When that altered block is added to the blockchain, all other nodes will realize its hash function is incorrect (and that a change must have been made on previous blocks) and the update will be rejected (for more detail on this process see here).
This fundamental aspect of blockchain is what makes the technology tamper-proof and secure. Thus, Blockchain technology allows for transparent and incorruptible data that does not have a single point of failure and cannot be controlled by a single entity, making it potentially an ideal platform for digital voting.
Current Blockchain Voting Applications
A few different blockchain voting systems have been proposed.
The startup Follow My Vote is also trying to implement a blockchain voting solution that involves remote electronic voting. With Follow My Vote, the voter has to install a “voting booth” on their computer, tablet or smartphone. The voter also needs to verify his/her identity by submitting legal documents to a so-called “Identity Identifier”, which presumably will have been approved by the entity holding the election. Once a voter’s identity is verified, he/she can request a ballot and vote digitally. One interesting thing about the Follow My Vote system is it allows voters to change their vote right up to the end of election day. Once the polls close, the most recent vote counted.
Follow My Vote’s approach of e-voting is flawed if it is intended to be used by governments for political elections: authentication is challenging and easily compromised, voters’ personal computers could be compromised, voter intimidation is more likely remotely, auditing is more difficult remotely, and, most alarmingly, the developers found that even a “minimally resourced adversary could launch a Denial of Service attack on an entire neighbourhood”. This is why most voting cybersecurity experts favour a paper-based audit trail.
In short, there are too many security vulnerabilities for e-voting to be used for political elections. However, it could still be used for private elections with less at stake.
Votebook, which was created by NYU Master’s students, is also trying to implement a blockchain voting solution. However, its implementation is very different from Follow My Vote. The significant difference lies with how votes are cast. The Votebook system does not use remote voting because of the security vulnerabilities. It uses a paper-based audit trail.
Votebook’s system involves voting machines that look like traditional voting booths. However, they are also computers acting as nodes on a blockchain. On an election day, the Votebook voting machines collect votes as they are cast and organize them into a “block.” The block is broadcast on the network and every other node checks the validity of its components using a public key to decrypt the hash and check for a match. The receiving nodes then verify the hash of the previous block in the database to make sure no tampering occurred.
With the Votebook system, the voting experience is almost identical to the current voting experience, while being significantly more secure and reliable than current electronic voting options.
VoteWatcher is a branch of Blockchain Technologies Corporation (BTC). It describes itself as a blockchain voting system not just for government elections but all types of voting. To date, VoteWatcher has been used at eleven different voting events and has counted up 5,000 votes.
Similar to Votebook, the voter experience with VoteWatcher is similar to traditional voting in that it involves a paper ballot. However, in addition to boxes to check, the ballot contains three QR codes— one is a blockchain address, one is a ballot ID, and one is an election ID. Once a voter marks the ballot it is scanned by an onsite machine, which sends the vote to the proper candidate’s unique address (similar to a bitcoin wallet) on an offline blockchain. The machines keep an image of every ballot. Once the election is over, all the machines are connected on a network and start adding the votes to a final online blockchain. Using this method, anyone can check how many votes each candidate acquired in real time using a blockchain explorer.
VoteWatcher uses blockchain technology to create a tamper-proof and public auditing system but paper ballots still back everything. The reason is the same as Votebook’s: e-voting is impossible to secure if done remotely.
A Blended Approach
Based on the three examples above, it would seem a voting system that uses a blockchain as a ledger but still uses paper ballots is the most realistic option for government elections today. Following this logic, Concordia University Professor Jeremy Clark has proposed a blended system that would continue to use paper ballots for voting, while adding an auditing and counting system that uses blockchain technology.
In Clark’s proposal, voters would register to vote in the same way they do today. They would also vote similarly in offline and physically secure voting booths. However, in doing so they would use paper ballots with three QR codes similar to those developed by VoteWatcher. Once the voting period is over, the booths/machines would store all digital information on a DVD or flash-drive before they are put online. The DVD or flash drive and the physical paper ballots would be securely stored in case auditing is necessary. Once online, each machine would become a node in a “permissioned blockchain,” similar to what was proposed by Votebook.
Thus, Clark’s system borrows from both VoteWatcher and Votebook to create a very secure voting method that could be used for governmental elections. It draws on blockchain technology to enable the secure scaling of old-fashioned paper voting. The use of on-site voting minimizes any security risks that could come along with e-voting while creating a transparent, incorruptible voting record with both digital and paper audit and recount features.
Governments Might Not Want More Secure Voting Methods
Despite being a more secure technology, there are obstacles to adoption of blockchain voting for political elections. First, the technology is new and relatively unknown to the general public. Second, blockchain voting requires a robust and reliable internet infrastructure, as well as access to expensive and energy-intensive voting machines. There aren’t many countries around the world that can meet these requirements. Finally, and most importantly, some governments might not want fair and secure elections, since they might lose the ability to manipulate the weaknesses of their current voting systems. For blockchain voting to become more common, governments around the world must be willing to improve the democratic process in their respective countries.
Nonetheless, there are signs that the political adoption of blockchain voting is underway. The first use of blockchain voting technology happened in Denmark in 2014 when the Danish Liberal Alliance used a blockchain voting system to conduct an internal vote. Also, the VoteWatcher system was used at least twice to count votes at Libertarian Party Conventions in the U.S. In 2016, Ukraine and the United States signed a memorandum to develop a blockchain voting system. Russia also announced in 2016 it had developed a blockchain-based proxy voting system. Finally, the tiny Australian Flux party has advocated for the use of blockchain voting in Australian elections.
Based on the difficulty with traditional voting methods and the different blockchain voting solutions that have been developed so far, it seems like the best and most realistic implementation method would be to continue to use a physical ballot and onsite voting while mobilizing blockchain technology to count and keep an accurate record of cast ballots. Beyond the technical challenges of blockchain voting, however, are political ones. Acemoglu and Robinson argue the primary driver of economic development is whether or not a country’s political system is “inclusive.” An inclusive political system has the people’s best interest at its center. Whereas, an “exclusive” political system is centred around the interests of a country’s wealthy elite.
Fair and tamper-free elections are perhaps the primary way a political system can be “inclusive”, and this can be enhanced using a blockchain system. Yet, it is precisely why blockchain voting might encounter resistance in exclusive political systems. There is a ruling elite with a vested interest in the way things currently operate. Perhaps a way around this will be to focus on the implementation of blockchain at very local levels, where it will be easier to accomplish. Over time, perhaps, one of the cumulative effects of enhanced elections at local levels will be to pull the broader political system towards inclusivity.