[COLUMNIST] Should Malaysia implement e-voting to overcome low voter turnout?

Voter turnout was only at 55% while BN won 43% of these votes, meaning BN won more than two-thirds of the seats by capturing merely less than a quarter of possible votes. If more people had shown up, the outcome could’ve been different.

Therefore, uniting their forces and increasing voter turnout are the two clear paths moving forward for the opposition. Luckily, there are technology solutions for the latter.

Specifically, electronic voting (e-voting) does not simply mean an electronic ballot at a physical poll site, but actual remote voting through the internet.

This has been practiced in Estonia (“i-Voting” as they call it) and according to the Estonian government site, it allows voters to cast their ballots from any internet-connected computer/mobile phone from anywhere in the world.

The relatively simple process increases accessibility, effectively increasing the number of voters. Estonians achieved nearly a half-half split between the share of online voters and physical poll-site voters in the European Parliament elections in 2019.

Furthermore, the Estonian system makes the time slot to cast votes to be 6-days (also known as the pre-voting period), instead of what was reportedly a mere half-day window for some Johor voters. Estonians are also allowed flexibility to change their vote within this 6-day window.

This can overcome many reasons that contributed to the low turnout in the recent Johor state elections such as concerns on Covid-19 transmission, difficulty in making arrangements to return to Johor after taking a long Chinese New Year holiday, restricted youth availability to travel due to education schedule, cost, time and procedural issues related to inter-state travels and especially international travels for citizens abroad, and many other excuses.

How can the authorities allow such circumstances to occur when technology solutions are available? Voter turnout is a matter of people’s future. The future of the nation is literally at stake.

Remote voting through the internet makes the distance to the polling station and other factors irrelevant. Imagine being able to get the other 45% of Johoreans to cast their votes.

That said, it is acknowledged e-voting systems are not without problems. It is important to note that development and use of Estonian i-voting system started in 2005 when technology was relatively nascent compared to what it is now. Expectedly, the system faced various issues and criticisms globally since its inception.

For example, computer science researchers from The Johns Hopkins University (JHU) opined those potential issues surrounding technical security and social sciences could make e-voting infeasible. In 2015, international e-voting experts Open Rights Group (ORG) Election Observers performed an independent evaluation of Estonia’s i-voting system and claimed to have found gaps in procedural and operational security that could open the system to cyberattacks.

As e-voting pioneers, Estonia continuously improves the system and in 2019, the Estonian e-voting task force or working group (said to consist of state officials, critics and experts from universities and research institutes, and developers of the system) concluded a report with 25 recommendations to improve the system, addressing reliability issues and potential risks.

The point here is that Estonia has more than a decade of experience in developing and improving the technology and system that Malaysia can learn from.

In particular, the Estonian system shows how electronic identification technology acts as the foundation for e-voting.

To cast a ballot, Estonians use their government-issued electronic ID to log into the system, which can happen anytime within the designated pre-voting period. To protect voter anonymity, the system has been designed to remove the voter’s identity from the ballot before it is counted by the National Electoral Commission.

Accordingly, the pre-requisite for e-voting in Malaysia would be the National Digital ID (NDID)—a digital identification and authentication to verify a person's identity in the digital world.

Unfortunately, it has been reported that Malaysia’s NDID will only be fully implemented in 2024, and that’s assuming no delays or hiccups are associated with project implementation.

The bill for the NDID implementation was passed on 8 May 2019. Even so, a public consultation report on the NDID framework was only published in August 2020.

The fact that it took more than a year since the passing of the bill to have a report on public consultation means this isn’t a priority. Sure, Covid might have taken center stage within this period but it’s not an excuse that can be used forever.

The passed bill was said to include other works such as detailed landscape study, implementation model, cost analysis, follow-up steps, the benefits, etc. which could take a long time if not prioritised and expedited. Malaysia Digital Economy Corporation’s promotion of the digital ID industry at Expo 2020 Dubai shows how Malaysia sees NDID as being an important enabler of the digital economy, and the fact that Malaysia has local companies that are experts in the digital ID space means we have more than enough capabilities to make it happen.

As reported in Digital News Asia, registered chartered and professional engineer Ir Dr Shawn Tan suggested that the Malaysian government adopt similar measures to that used by Estonia in ensuring full adoption of digital certificates and signatures to accelerate the implementation of the NDID.

After all, as written in the Malaysia Digital Economy Blueprint (MyDigital), “accelerating” the National Digital Identity implementation is specifically mentioned as a thrust.

Malaysia’s NDID should also be developed in such a way that prepares for integration with global digital ID systems such as ID2020.

Specifically, both NDID and e-voting systems should be empowered with secure technologies such as cryptography.

The use of new and emerging technologies in the space of cryptography to bolster e-voting systems was mentioned by officials involved in the said Estonian task force that studied how to improve the e-voting system.

Indian multinational information technology services consulting company, Tata Consultancy Services (TCS), published “Impact of Blockchain on Digital Identity” in which they mentioned that the pillars of a blockchain-empowered digital ID include: security through immutable encryption, integrity through continuously kept and peer-to-peer reconciled data, trustworthiness through decentralisation, consensus-driven data authenticity, verification, and privacy by design.

TCS also noted how even the ID2020 initiative is using blockchain-based solutions to achieve its goals. These characteristics appear to fit nicely, at least theoretically, with the security requirements for electronic voting outlined by experts from the JHU.

Requirements include authenticity to verify voter identity and voter entitlement, voter anonymity to dissociate votes with voter identity, system integrity to prevent re-configuration while in operation, data integrity to prevent tampering of votes and ensure proper recording, and many other requirements which cryptography-empowered NDID and e-voting systems could fulfil.

Another key requirement is cost-effectiveness, whereby JHU experts mentioned that collection systems should be affordable and efficient.

This is a real issue as a parliamentary response on the total expenditure by the Elections Commission (EC) just for the Melaka and Johor state elections was said to be RM141 million, covering the procurement and rental of equipment, vehicles, ICT developments, logistics, and allowances of EC staff.

With e-voting, there could be substantial savings on operational expenses as equipment, logistics issues, and on-site manpower requirements are significantly reduced.

As mentioned by TCS, one of the pillars of a blockchain-empowered digital ID is simplicity, achieved by cutting time and manual interventions. Thus, similar benefits could be achieved for e-voting.

French multinational technology corporation, Thales Group, opined that digital ID stakeholders should prepare for mobile-first solutions, that is, an approach in which web designers start product design for mobile devices first.

Thus, combining the latest mobile-first designs and technologies with the technical experience from Estonia (and other countries such as Canada, Australia, France, and Switzerland), we can imagine a secured and verifiable e-voting through mobile phones and other mobile devices.

Of course, it is easy for people to be doubtful of technologies they don’t understand, and such concerns are likely used as a central narrative by those who have a vested interest in not increasing voter turnout.

Paper ballots and the commission running them may not be tamper-proof and have not been free from criticisms either. So why not try something new?

The need for data and systems decentralisation as a key pillar to build trust in e-voting will also open up the chance for the much-needed structural reformation of EC.

Authorities should conduct trial runs, starting with internal testing, scaling up to use in party elections and state elections to ultimately prepare for federal elections rollout. They can start with eligibility for Malaysians living abroad, gradually expanding this to everyone else.

In Malaysia, successful implementation would mean making full use of automatic voter registrations, bringing out the full potential of younger voters, and increasing overall voter turnout.

---

Ameen Kamal is the head of Science & Technology at EMIR Research, an independent think tank focused on strategic policy recommendations based on rigorous research.