Considerations for Blockchain-based Voting Systems.

  • Commentary
  • November 16, 2020

As an emerging technology considered to have a wide range of applications, blockchains have been proposed as solutions to the shortcomings of countless legacy systems; these include social media platforms, supply chain management systems, and, of course, financial services. In the current political environment, however, it’s no surprise that chief among these proposals is a blockchain-based voting system. In a recent article for Forbes, John Koetsier discusses this matter with Horizon State CEO Tim Goggin. Building on this interview, this dispatch evaluates at a high-level the potential benefits and shortcomings of blockchain-based voting systems.

Blockchain-based voting systems have long been proposed as a means of increasing voter accessibility, while their immutability and censorship resistance theoretically allows them to preserve ballot and voting integrity in a way that other such proposals cannot. More recently, they have also been considered for their potential to facilitate more accurate and auditable vote counting. In the aforementioned interview, Goggin reaffirms these assertions, stating that accessibility, authentication, and security are the primary value drivers for blockchain voting systems and offers an example of how such benefits would be realized. Imagine, he suggests, a hypothetical system in which a voter’s real-world identity is verified and assigned to a specific blockchain address, accessible only by the owner holding the private keys. A token representing one vote is then sent to the voter’s address and, in turn, the token holder may vote by sending this token to a contract representing their preferred candidate. 

From our perspective, the most obvious benefit of such a system, with regard to accessibility, is that it would enable digital voting in a more secure manner than legacy voting systems, thus allowing users to vote from various locations and removing the need for in-person voting. However, according to Goggin, this system would also increase accessibility by mitigating the risk of physical loss and improper addressing present in mail-in voting systems. This is perhaps misleading, as removing the risk of physical loss does not remove the risk of loss altogether; for example, it is possible that users lose their private keys, or perhaps their tokens could be ‘lost’ should users send them to an invalid contract address. Additionally, the risk of improper addressing due to voters’ moving addresses would not be fully eliminated, as adjustments must be made if the voter moves counties or special districts, making it very possible for a user to receive voting tokens for candidates and measures outside of their current locality as the result of outdated address data.

We also consider the signing of tokenized ballots using private keys to be improvements upon the current system’s authentication mechanism, which relies on, in many cases, merely a signature. The use of digital signatures can remove the risk of signature forging and the risk of legitimate ballot rejection based on lack of an exact signature match. However, to realize these benefits, such a system would require a robust identity primitive. This would presumably require initial identity verification through a driver’s license, social security card, or other physical identifiers, the use of which in traditional voting systems has been widely criticized and litigated for its potential to disenfranchise certain populations. Further, while blockchains’ pseudonymity may mitigate the risk of voter corruption—though not meaningfully more than current systems do—vote-selling by willing voters represents an outsized threat to the purported authentication benefits of blockchains. The remote nature of digital voting makes the actual transfer of tokenized ballots significantly easier than the transfer of mail-in ballots, for example. Further, the auditability of blockchains allows for voters to trustlessly prove they sent tokens to a specified candidate’s contract and, in turn, receive compensation from nefarious third-parties.

Perhaps the most consequential aspect in which blockchain voting systems can improve upon legacy systems is security. The redundancy, cryptography, and resulting immutability likely protects votes from being tampered with post-submission and post-counting. Auditability ensures reliable counting and recounting, without the need for in-person monitoring of vote counting. However, there exist two centralized points of failure wherein blockchains have no ability to improve security. First, in order to facilitate usability, there would presumably need to be an off-chain platform for the submission of votes, which is as corruptible as any other comparable Web2 application. Second, an off-chain database would be required to match real-world identities to on-chain addresses, which could be likewise exploited in order to invalidate certain votes. Further, the extent to which immutability is present in a cryptonetwork is dependent on the number of nodes on the network and who is operating them. Lastly, because this hypothetical system has no ability to independently determine eligibility, it would not be able to ensure that all votes cast are legitimate and not submitted by or on behalf of newly-ineligible voters; it would similarly not be able to ensure that one person receives only one vote, nor that the process of distributing addresses to voters is free from bias and corruption.

Notably, this discussion is not comprehensive in either its evaluation of the proposed hypothetical blockchain voting system nor does it consider alternatively structured systems also based on distributed ledger technology. Indeed, it is almost certain that the merits and limitations of, for example, a mail-in voting-adjacent blockchain system used for simple verification would vary from those discussed above. Rather, during a time in which many appear eager for changes to the current voting system, this dispatch intends to contribute to a more balanced discussion surrounding what can and cannot be made possible by current instantiations of distributed ledger technology. Smith + Crown believes that blockchains have the potential to improve upon legacy voting systems in the future and will continue to monitor this space for advancements that may possibly produce a system free of some of the limitations discussed above.