EC must clinch debate with a proper VVPAT audit

As the Lok Sabha elections are just around the corner, it is no surprise that the controversy around Electronic Voting Machines (EVM) refuses to die down.

Only a month has passed since the rather anti-climactic event on EVM hacking in London where a number of unsubstantiated claims were made regarding the machines by one self-proclaimed cybersecurity expert, Shah Shuja, a former employee of the Electronics Corporation of India Limited (ECIL).

Besides individual citizens, all political parties have gone public with their doubts about EVMs at some point or the other. Needless to say, those doubts vanish when they win an election conducted using the very same machines!

In 2011, the EC had an opportunity to test and demonstrate the relative strength of the machines over paper ballots when to conduct the by-polls to 12 seats of the Andhra Pradesh assembly which had fallen vacant due to mass resignations in the wake of the Telangana agitation, we had to conduct polls in half the constituencies with ballot papers where the number of candidates in the fray were more than 64 candidates -- the maximum the EVMs at the time could handle.

While the results from the EVMs came out in four hours, those from paper ballots took almost 40 hours. Additionally, thousands of votes cast by paper ballot were, inevitably, invalid. Had they exceeded the margin of victory in any seat, they would have cast doubts about the legitimacy of the election and the winners.

Actually, the debate about EVMs should have ended in 2010 when the decision to adopt a Voter Verifiable Paper Audit Trail was taken at an all-party meeting. The two factories that manufacture the EVMs were tasked with the manufacture of VVPATs also. An independent committee of five IIT professors was requested to monitor all aspects of the process. After a series of trials, two full-day election, simulations were held in five cities with varying climates -- Cherrapunji, Delhi, Leh, Jaisalmer and Thiruvananthapuram -- in 2011 and again in 2012.

After the improvements were incorporated into the design of the machines, they were successfully deployed in 20,000 polling booths randomly chosen. As manufacturing progressed, all booths were covered in all constituencies.

In 2013, the Supreme Court even lauded the EC’s decision to bring in VVPATs and directed the Centre to release adequate funds for the same so that the 2019 general elections could be held entirely with these devices.

Since 2017, all state assembly elections have been held using attached VVPATs, and slips generated by nearly 1,500 of these have been counted. Not a single mismatch has been detected. This further confirms that VVPATs are the best solution, not going back to paper ballots.

In a performance review conducted by the EC in mid-2018, it was found that the VVPATs had an aggregate malfunction rate of 11%. This is more than double the rate of EVM malfunction, which stands at 5%. This can be easily explained. EVMs are solid-state devices analogous to calculators (trouble-free for years) whereas VVPATs are electro-mechanical devices analogous to printers (face problems such drying up of toner, jamming of paper, etc).

One distinction in the EVM debate is crucial: malfunction is not the same as rigging, and equating the two seriously harms voters’ trust in the electoral process. An easy solution to remedy EVM malfunction is that enough machines should be kept as reserves for quick replacement so that there is no disturbance caused when they malfunction for any reason.

Recently, Congress leaders, along with other opposition parties, have demanded that all EVMs must be linked to VVPATs and half of the total slips should be tallied. The first is already being implemented, and the second needs immediate discussion.

In the latest such development, a group of 73 retired bureaucrats and diplomats have written to the EC demanding a complete audit of votes cast in the forthcoming election. The text of the letter states that “there is an imperative need for an additional verifiable physical record of every vote cast, in the form of voter verifiable paper audit trail”.

The retired officers expressed their concerns about decision rules and sample size determination. The letter goes on to highlight major shortcomings in the present system, all of which are related directly or indirectly to the methodology by which the EC chooses the sample size to be tallied per constituency, calling the present method of matching just one EVM per constituency “statistically incorrect”. Other concerns include defective EVMs and lack of “transparency about the results of its VVPAT-based audit for the various Assembly elections held in 2017 and 2018 and the details are not available on its website”.

It is worth noting that nowhere in the letter has going back to paper ballots been proposed.

The letter rightly pointed to the fact that “the real issue today is not about EVMs vs paper ballots, rather it is about EVMs with perfunctory VVPAT audits vs EVMs with proper VVPAT audit.” The former CEC OP Rawat announced that the EC had requested the Indian Statistical Institute (ISI), Kolkata, to recommend the sample size that would be the most scientifically sound to ensure 99.9% public satisfaction. The letter mentions the same towards the end, saying “we appeal to the ECI to go in for a statistically correct sample size that can detect at least one defective EVM with 99.9% reliability, in a suitably defined population.” That should be the real clincher in the debate. It is surprising that we have not heard what the recommendation of the ISI is. A quick decision on this is imperative if the Election Commission would like to clinch the debate, rather than leaving it to the Supreme Court where a case on the issue is pending.

(The writer is a former Chief Election Commissioner of India and author of An Undocumented Wonder: the Making of the Great Indian Election).