This article was published in Dawn on April 17, 2017 and is available at the following link

https://www.dawn.com/news/1327408/the-case-of-evms

ELECTRONIC voting machines (EVMs) seem to have gained greater currency, and popularity as well, in our electoral discourse over the past few years. The Election Commission of Pakistan (ECP) may also have raised the expectations of political parties and the general public when it presented a few prototypes of the EVM to representatives of political parties, in full media glare, in September 2014 and announced it would introduce EVMs in two years.

Apparently, two key misconceptions exist about EVMs in Pakistan. First, a sizeable segment of public and political leaders believe that EVMs are going to solve all or most of the problems relating to suspected election rigging in Pakistan. The apparent success of EVMs in India has prompted many in Pakistan to give exaggerated credit to EVMs in the conduct of the relatively less controversial Indian elections.

The fact is that EVMs have not been without controversy in most of those countries where it has been used in the past years. Generally, it is believed that the Indian EVM has withstood the test of time. In reality, it has faced many challenges both inside and outside the courts of law. In particular, recent allegations of rigging by manipulating Indian EVMs have raised serious questions about the credibility of the machine. These questions have a special meaning for a country like Pakistan where the level of trust in the conduct of the polls is already low.

Electronic voting machines may not be as marvellous as they are made out to be.

EVMs were first used in India in a few polling stations in the Kerala state election in 1982. A candidate challenged its use and the Supreme Court disallowed the use of EVMs on the ground that election laws were not amended accordingly. Even after the laws were amended, some political parties and politicians continued to voice apprehensions and the government had to appoint the Goswami committee in 1990, whose sub-committee examined EVMs.

In 2000, the Election Commission of India (ECI) had to appoint a respected independent think tank called Centre for the Study of Developing Societies to ascertain the trust in and popularity of EVMs among the general public.

Although a major benefit of EVMs was considered to be the saving of a huge amount of paper, some political parties petitioned the Supreme Court in 2010 for a verifiable paper record of individual votes, in addition to the record in the EVM Sim. The court directed the ECI to introduce the ‘voter verifiable paper audit trail’ in a phased manner. The ECI had to modify the EVM design twice to provide for the VVPAT. Now it has started keeping a paper record of each vote polled in case the result is challenged.

In the recent state elections in India, some prominent politicians, including Delhi’s chief minister Arvind Kejriwal and Mayawati who heads the Bahujan Samaj Party, alleged rigging through the manipulation of EVMs. It was widely reported in the Indian media that one of the machines in Madhya Pradesh printed votes for the BJP even when buttons for other candidates were pressed.

Earlier, a security analysis of the Indian EVMs authored by eight persons including Dr J. Alex Halderman, a professor of computer science at the University of Michigan, concluded in 2010 that “in spite of the machine’s simplicity and minimal software trusted computing base, it is vulnerable to serious attacks that can alter election results and violate the secrecy of the ballot”. Similar issues have been experienced in electronic voting elsewhere, including countries which have a far more developed technological base than Pakistan.

The second misconception is about the possibility of conducting the entire upcoming election through EVMs in Pakistan. It will be instructive to again see the timeline followed in India, which first employed EVMs in 1982 in a few polling stations in a state election. It continued to experiment with EVMs in selected constituencies of various state elections. Goa was the first state in which EVMs were employed in all constituencies of the legislative assembly in 1999.

The same year, the ECI used EVMs in 46 Lok Sabha constituencies in 17 states in which 60 million voters used EVMs. It was not until the 2004 general election that EVMs were used across the country for the Lok Sabha election. It had taken over two decades since the first use of EVMs in selected constituencies for the ECI to allow its use across India for the Lok Sabha.

The ECP very recently ordered 150 EVMs (the requirement for all constituencies is around 300,000) which it plans to use on an experimental basis. The machines are scheduled to be delivered towards the latter half of 2017, after which training of staff will commence. It will not be before the first half of 2018 when the experimental use of EVMs would become possible. If the ECP successfully uses EVMs in selected constituencies during the 2018 general election and subsequent by-elections, and if legal challenges do not delay the process, and if funds are made available in time, the 2023 general election may be the earliest opportunity to use the machines universally.

This would be a rather ambitious goal. Conservative estimates indicate that it will require Rs30 billion to shop for 300,000 EVMs. A careful cost-benefit study needs to be carried out before going all the way to procure EVMs. International trends are, at best, mixed. Many EVM user countries have reverted to traditional paper ballots. At the end of the day, EVMs are expensive and will take at least five to six years to be made properly functional. There is no guarantee that the quality of election would be better with the use of EVMs. We should be patient, realistic and cautious with the use of technology in elections.