by Oleg Shvaikovsky, President of Nortal LLC, May 7, 2020
Voting is a ritual that has not changed in many countries for decades. People still go to a polling station, hide behind the curtain in a small booth, use a pen to mark their preference, and then put their ballot papers through the little slot on top of the box. But in this new reality of social distancing brought on by the pandemic, voting is now a hazard to public health.
The International Institute for Democracy and Electoral Assistance (IDEA) reports (May 2020) that at least 52 countries and territories across the globe have decided to postpone national and subnational elections due to COVID-19.
IDEA stresses that elections are the opportunity for citizens to either reconfirm or remove and replace an elected representative or government. A decision to postpone an election suspends political rights, and as such, undermines the social contract between a government and its citizens. Essentially, both proceeding with or postponing elections before the COVID-19 pandemic is contained entail risks for decisionmakers. While the latter may be the most feasible and responsible option from a public health perspective, the decision may allow other risks to materialize. Governments need to allow for clear pathways that guide how existing institutions and actors can proceed during the extension period, and when the regular electoral routines will be reinstated.
What these countries can do now is take advantage of the fact that change is in the air.
Estonia had its latest parliamentary election in March 2019; during the same time, a year later, it would already have been highly questionable if the polls would go forward as usual with the coronavirus lurking around the corner. But if any country is prepared to hold an election without people overcrowding the polling stations, it’s Estonia.
In Estonia, 44% of people vote online — which has been an option since 2005. Citizens prefer to vote online not because they can’t go to the polling station (although, that too), but because it’s convenient. Internet voting, or i-Voting, is a system that allows voters to cast their ballots from any internet-connected computer anywhere in the world. Utterly unrelated to the electronic voting systems used elsewhere, which involve costly and problematic machinery, the Estonian solution is simple, elegant and secure.
In Estonia, 91.6% of the population uses the internet regularly and the ID-card penetration is 98.2%; both are prerequisites to holding an i-Vote. Setting these numbers against 63.7% — the percentage of the electorate who voted in the last parliamentary elections — it is clear that if Estonia ever needed to switch to voting online-only, it could probably pull it off even though it has always been just an alternative to traditional voting.
One country battling with holding an election during the COVID-19 pandemic is the United States. The infamous presidential election is already pulling crowds to polling stations to cast their vote in the primaries. The alternative is mail-in ballots. Partly because it is believed mail-in ballots give Democrats an advantage, the Republicans have snubbed the mail-in ballots as ‘recipe for fraud’. Voting fraud in the United States is rare; even so, experts say that the mail-in voting system is more vulnerable to fraud than voting in person.
Exploring voting options that would safeguard people’s health, go easy on the country’s wallet and also work in the long run, the i-Vote is a natural conclusion.
The Economist, however, writes in a recent article that “online voting faces serious, possibly insurmountable obstacles. /…/ the technology remains vulnerable to security breaches and cyber-attacks. Malware can tamper with votes before they reach government servers. Hackers can create mirror versions of an election portal, steal voter credentials, or attack computers that count and store online ballots.”
However, when Estonia held local elections in 2005, and two years later became the first country in the world to hold nationwide elections using this method, it was in sharp focus. And in the following years, more than a dozen security delegations and commissions have overseen i-Voting and studied the system. Not one breach of security has been detected in those 15 years. The source code was made openly available in GitHub, if anyone was interested, or is still interested in finding loopholes.
i-Vote tallying mimics the double envelope scheme used in postal voting. The digital outer envelope is digitally signed using the secure e-ID card. And before counting the vote, the system anonymizes the ballot by removing this digital outer envelope.
Votes are then counted by a computer with the bare minimum capability to ensure non-exposure of the result to any computational alteration. The computer has no internal storage and has no internet or network connection, only a DVD drive and a smart-card reader for reading and storing the result; the RAM disk is used as its processor. As the computer is turned off at the end of the process, all information disappears. (Read more about it from HERE.)
The Economist also argues that there are no agreed-upon auditing measures for voting online. Any suspicion that an outside power has tampered with votes cast online can destroy confidence in the whole electoral process.
The same logic applies to any e-service. One case of identity fraud (happens when someone uses your e-identity to do something without your authorization) could potentially compromise the whole idea of e-ID. Any potential case of fraud with prescriptions could compromise the entire notion of e-prescriptions.
The fact is that we know a lot of election fraud cases (see more: https://www.heritage.org/voterfraud/search), but we don’t yet have a single documented case of fraud in e-elections.
Voter coercion is the only fundamental shortfall of i-Voting. When voting from home (although the same as with mail-in ballots), anyone from your family can try to influence you when you cast your vote. Estonia’s solution was to allow i-Voting during a specific period before the polling stations open. The voter is then allowed to log on and vote as many times as they want during the pre-voting period — meaning you can change your vote when you feel unduly influenced.
The Economist, though, is correct when stating that online voting cannot save this year’s elections. When a country wants to do it right and hold a successful i-Vote, fulfilling certain conditions must happen first.
First — determine the objective. i-Voting has had different purposes in the world. For example, a former Soviet republic wanted to engage liberal citizens living abroad to help vote out its communist government. Although extreme, this example shows i-Voting’s ability to engage different groups that would otherwise not participate in elections — either because they live abroad, it is difficult for them to get to a polling station or just because they can’t be bothered. Right now, the main objective is enabling voting in lockdown. Whether a lockdown of such extent will happen again, you still need to determine which groups you want to direct toward i-Voting. Even though voting online has been possible in Estonia for 15 years, more than half of the voters choose to go to the polling station. i-Voting works well as an alternative, but think twice if you want to make it the only option.
Second — set up the infrastructure. The first i-Voting in Estonia took only a few months to set up, but only because Estonia already had national registers, ID-cards, and the X-Road for the secure transfer of data. If a country was to take the opportunity to secure voting no matter what the future might bring, it would take at least a year to set up the national register and another year to implement a secure online identification system similar to eID (e.g. a Smart-ID, which is also used in Estonia).
Third — possess the political will. With every individual trying to redefine their future, it is time for countries to also look ahead and secure their democracies.
Nortal knows what it takes to get there — which laws to change, which institutions to build, how to develop the system and write the code, and how to implement it. All you need to bring to the table is political will.