On December 3, a consultative referendum was held in Venezuela regarding the disputed Essequibo territory with the Republic of Guyana. The five questions posed for voting boiled down to being in favor or against Venezuela integrating that territory and rejecting the delimitation established by an 1899 arbitration. According to official data, the “yes” vote overwhelmingly prevailed. While non-governmental media reported a marked low turnout, the National Electoral Council (CNE) stated that just over 10.5 million people had voted, which is roughly 50% of registered voters. Unofficially, it is estimated that just over 2 million voted, mainly government employees and military personnel coerced by the authorities.
In addition to the peculiarities of the “territorial dispute” between Venezuela and Guyana, the participation data draws attention to a less visible detail. In this referendum, an electronic voting system with touch screens was used, recording votes directly and printing receipts that voters then deposited in the ballot boxes. This system has been in use since 2004, replacing the optical vote scanning applied from 1998 to 2003, but wasn’t very reliable. Starting in 2012, the entire electoral management system was fully automated: voting, counting, totalization, allocation, and results dissemination. Why is it then important to highlight the electronic voting system?
This system has been questioned since the 2013 elections because its transparency relies on the electoral authority, responsible for both its implementation and the declaration of results. The CNE has boasted of its stability and security, emphasizing the printout of votes as a verification mechanism, having a backup power system, and transmitting data via satellite or unique phone lines, not over the internet.
Although the opposition has denounced electoral fraud, it referred not to result manipulation but to cuts in voter lists, identity theft of voters and supporters of the ruling party voting more than once. Keeping the electronic voting system had been considered an advantage to gather evidence of act manipulation or vote counting; reverting to the manual system would facilitate electoral fraud, so it was never in question. Even allegations of system hacking or software manipulation have remained mere speculation, despite detected irregularities.
However, the results of the referendum on Essequibo deliver a very clear lesson: electronic voting systems are not immune to electoral manipulation. Fraud is orchestrated by those managing elections and issuing results. Electronic voting systems can become a facade to feign integrity. This had already been detected by the company Smartmatic, which designed the system used by the CNE in 2017, when the Constituent Assembly was elected on July 30. The system had counted 6.5 million votes, but the president of the electoral body, Tibisay Lucena, declared 8 million. In the governor elections in October of that same year in the state of Bolivar, 13 machines printed the records. However, military agents prevented the automatic information transmission, manually altering results and favoring the ruling party candidate while preventing the victory of the opposition candidate, whose representatives had the records issued by the electronic ballot boxes.
In addition to Venezuela, Belgium, Brazil, the United States, Estonia, the Philippines, and India have implemented electronic voting nationwide. Estonia is a pioneer. In 2005, it became the first country to allow internet voting, even using an open-source system available for public scrutiny. In the 2019 elections, 44% of the electorate voted online, using only their identification, an internet connection, and a PIN code. In Brazil, electronic ballot boxes have been in use since 1996, and since 2000, biometric identification has been used. The printing of votes or not has been a subject of legal advances and setbacks in that country, but it has progressively been implemented in all elections to date. In Mexico, there are various electronic ballot projects both at the subnational and national levels, successfully tested by electoral authorities since the beginning of the century. In the 2021 elections, voting from abroad can be done online.
The implementation of electronic voting is not a linear process. It is initially not socially and politically accepted, but it is not a process that must necessarily be incorporated into all democracies. In countries like the Netherlands (since 1965) and the United Kingdom (2000 to 2007), where legislation allowed electronic voting, they had to return to the pencil and paper system due to social rejection and suspicions of fraud. “In the Netherlands, we know how to use paper and pencil. The sky did not fall and we did not go back to the prehistoric era,” read a slogan. In Germany, Norway, Ireland and Finland, this system has been banned after several tests, and mostly citizen organizations—many of them led by systems engineers—have raised their voices and pressured governments not to implement it.
Elections are a complicated process and new technologies add more complexity. The risks of fraud can arise from dispensing with verification ballots and from the complexity of the software used, which depends on specialists who can also be manipulated. Now, any new technology applied to election management processes, such as the innovative blockchain, is feasible to use as it has various everyday applications and offers greater security. The dilemma of new technologies is not their technical feasibility but their political use. Where electronic voting has been rejected, it was not because of the system itself but because of suspicions that it can be manipulated.
Voting is not a technical matter. It is a political act with very broad socio-political consequences. Whether through the internet as we can do today or with fragments of ceramics (ostraka) as the ancient Greeks did, voting yields the same result: the expression of a majority. The introduction of any new technology can make the election management process efficient, but it does not change its political nature and the case of Venezuela is an example of that. Even with electronic ballot boxes, manipulating elections is always possible.
*Translated by Ricardo Aceves from the original in Spanish.