In financial services, one of the most buzzwordy and divisive topics you can mention is Bitcoin. More generally, this refers to cryptocurrencies, types of currency that operate without any government or central bank controlling the flow information. This both worries investors and is an opportunity to be a part of a huge disruptor in financial services across the world. What I find even more fascinating though, is the tech underlying Bitcoin and other cryptocurrencies, known as blockchain.
While this post isn’t going to delve into the specifics of blockchain and its implementation, a basic explanation is that it’s an electronic ledger system that is distributed among computers across the internet, and is not controlled or housed by any one company or institution. As new entries, or “blocks” are added to the ledger on one machine, the updated ledger is then checked against all other machines that are a part of the distributed system to ensure the most recently updated ledger’s whole chain back to the beginning of time (or the creation of a ledger) matches all the other ledgers, save this one addition. This distributed network allows a high degree of trust between transactional partners, without the partners actually needing to trust each other in real life. The blockchain, as a part of its distributed nature, is incredibly resistant to tampering and hacking, as any hack on one system would be rejected by all other nodes on the system, and the hacked information would not continue in the creation of the blockchain. For a more detailed introduction to blockchain, check out this link. What I want to focus on here though, are the potential implementations we have for blockchain in the political sphere, both for elections and money in politics.
In recent years, the US has seen a marked increase in (or at least coverage of) election fraud and voter disenfranchisement. In the 2018 midterm elections, heavily African-American neighborhoods saw 10% of registered voters purged from voting lists, while surrounding neighborhoods with smaller minority populations saw only a 4% purge from their registered voters list. Droves of people in these neighborhoods were unable to exercise their civic right. In North Carolina, the race for a seat in the US House of Representatives was invalidated after election fraud was discovered surrounding the absentee ballots for that district. In our current system, people’s votes aren’t getting counted when they should be, and that’s on top of a population that ranks 31st out of 34 developed countries for voter turnout.
Surely, there has to be something we can do to improve on this. Many people think that moving our voting system online, where people could perform this civic right from their own homes, could be a solution to some of these problems.
So, in 2010, the Board of Elections decided to run an experiment of what voting online from home might look like, and how secure it could be. To put it bluntly, things went horribly wrong. They partnered with a non-profit to develop a system based on Ruby on Rails, then published the source code, and issued an open invitation for people to hack it. A professor and a few students at the University of Michigan were able to obtain the passwords for every voter in the test system, and then programmed the system’s thank you page play the Michigan Wolverines’ fight song. The breach wasn’t discovered for two full days…and the students got to watch them discover the breach, as they’d hacked into the cameras in the network operations center managing the trial system. I can’t imagine what it would be like to get pranked by these students…
Efforts toward online voting were diminished after that major blunder, but progress has come back in recent years. Companies like Voatz, Votem, and Smartmatic are attempting to bring the power of blockchain, and its inherent security protections, to the election process. So far the tests have gone fairly well, even if the tests haven’t been the most rigorous. The systems implemented by these companies allow a secure chain of votes to be strung together across a distributed system, with it being almost impossible to change a citizen’s vote once cast. The system even allows voters to verify their own votes, providing a QR code that the voter can scan (with another device) to ensure that the votes they cast are the votes recorded in the chain.
All of this is in an effort to make our elections end-to-end (E2E) verifiable, the pinnacle of election processes. This would mean voters know their choices are properly recorded, can verify their vote was counted in the official results, and the public can verify that the results of the election are accurate. The companies mentioned are focused on the first and third components, with the second hopefully coming into view as technology develops.
Not only could blockchain voting help ensure that all votes cast were untampered, but could make the election process easier and less expensive for everyone involved. Currently, the cost of a single vote costs between $7.00 and $25.00, but initial estimates of a blockchain voting system put the cost per vote closer to $0.50. That’s somewhere between 2% — 7% of the current cost of running an election, which means those tax dollars could go towards improving roads, schools, or lifting people out of poverty, whatever you want to vote for.
Blockchain also has the potential to combat the political disinformation war (thanks Russia) that was so prevalent in our last election. Limiting this kind of disinformation can be difficult without infringing on free speech, but a blockchain implementation around media source, ensuring that only media stamped with a unique cryptographic identifier that can be referenced with a blockchain record for verified media outlets could solve this issue. If people can know beyond a doubt where media originates, they could, for example, filter their feeds by only trusted media, significantly hampering disinformation efforts.
The final implementation I see for blockchain in the political process is around political donations. Money in politics is a significant problem in our current system, and we now have the data science to prove it. While political campaigns, candidate committees, political parties, and traditional PACs are required by law to disclose donors and contribution amounts, there are a number of ways, often through political not-for-profit companies and some Super PACs, where contributions can be made and no one will know where the money came from. This is often referred to as Dark Money, and you can read a quick primer on that here. While truly getting money out of politics could take many years, putting systems in place that require these organizations to use a public blockchain for all money received and spent would bring a new level of transparency to this system. Every transaction would be traced and viewable. We could see the real source of where contributions are coming from. We would also no longer need to trust the bookkeeping efforts of the organizations, but instead would only need to trust the blockchain record of funds.
While all of the potential here is fantastic, there are a number of people, especially those in cybersecurity, who are hesitant to incorporate this kind of system. In MIT Technology Review, Matt Blaze was quoted saying that it’s not necessarily that blockchains are bad, but that they introduce new security vulnerabilities, and securing votes is more easily, simply, and securely done with other approaches. It’s true that this tech is new, but I’m not sure I agree with his assessment, as the actual structure of blockchain has yet to have its security compromised. While repositories for Bitcoin have experienced security issues, the blockchain tech itself has shown itself incredibly resilient against hacking attempts.
What I and a number of people do see as problematic is the software implementation the blockchain voting companies utilize. The data transfer from the user’s input in the app to the actual node of the blockchain is ripe for tampering, and millions of phones (which many people want be the primary mode of future voting) are often infected with malware without people knowing. If these were our main avenues for voting, I can only imagine how renewed the efforts would be to bring election tampering malware onto these devices. Much of this can be alleviated by users verifying their votes with the QR code mentioned above, but that requires us to assume people will engage with this extra security step, and given how people approach data security today, I’m not convinced that would be effective. I personally also have reservations about how the implementation of this voting technology would be used at a more massive scale. So far, fingerprint scans, pictures of IDs, and video taken with someone blinking (to prove the camera sees a human and not a photo) have been used to verify a persons identity in pilot programs. What all of these metrics require is access to a smartphone that can run the app, take fingerprints, and has a functioning camera. The problem is that those of lower socio-economic status don’t have access to the same level of tech as the middle class. I worry that widespread implementation of this could re-create the voter disenfranchisement that we experience now, discriminating on income and access to technology.
Despite these reservations, a number of pilot programs have begun for the implementation of these blockchain systems, including one for overseas absentee ballots in West Virginia for the recent midterm elections. Denver, CO will also be testing a blockchain voting system for all overseas voters, active-duty military personnel, and their dependents. If we’re ever going to solve the issues that blockchain elections currently have, we need to start somewhere, and I think it’s worth the effort. While we’re still missing a true stress test to hack or break the system (which we need before this grows much larger), the potential for moving our antiquated voting systems into the age of technology, I think, is worth the effort.