EVMs Are Honest: The Proof Lies In The Invalid Votes Of Polls On Ballot Paper 

First things first. I wrote this paper originally in 2003, just before the Madhya Pradesh assembly elections were to happen. Then Congress chief minister Digvijaya Singh had thrown an open challenge (yet again) that he would win against all the odds, just like he had surprised everyone five years back.

But before you go on to read this article, you should understand the basic differences between the paper ballot system and electronic voting machines (EVMs) that are critical to getting the context.

In the paper ballot, the voter is asked to cast his/her vote by marking against his/her selected candidate and put the ballot in a ballot box. By law, there should be only one candidate selected, and there shouldn't be more than one mark on the ballot paper, or any other mark anywhere on the ballot paper.

If any paper had more than one mark anywhere, then that vote was considered “invalid” and taken out of the counting process.

In such a case, during the counting of votes, the power of validating or discarding these invalid vote rested in the hands of returning officers.

In India, the returning officers were usually the district magistrates (DM), and it was a standard practice for political parties in power to place a favourable DM in the areas which were supposed to see a very close contest.

Now during the process of counting, the counting agents of different political parties used to keep a keen eye to dispute opposition's vote as invalid, if they used to see even the slightest of the second mark anywhere on the ballot paper. The favourable DMs being the presiding officer, could accept these invalid votes of the ruling party and reject the same of opposition parties.

The decision of the returning officer was seen as final even in the courts.

Generally speaking, this didn't make much of a difference if any party was winning by a significant margin, but if the contest was close, then it became a matter of grave importance in the power games. And needless to say, the party in power (all the parties) used to misuse this influence.

Even though legally the Election Commission of India (ECI) was supposed to transfer such partisan officers, but it was not an openly known abuse of power, till then. So, this was a sophisticated, and systematic violation by ruling parties, in comparison to the crude form of booth capturing.

Enter EVMs, and the following two flaws of the ballot paper were nullified.

First, the EVM doesn't record invalid votes. You press a button and vote is recorded. There is no scope of pushing two buttons on the EVM.

Secondly, the notorious booth capturing tactic was stopped, simply because each vote in EVM is issued by the controlling unit with the election officer. So the option of simply snatching the entire bunch of ballot papers, stamping them and stuffing them in ballot boxes, which was a very common malpractice of that time, became unavailable.

Alternatively, in the polling booths where opposition parties were expected to have a big number of supporters, the strategically placed officers on duty for the ruling party would sprinkle ink in the ballot box, thus rendering most of those polled votes as invalid when the counting started – the returning officer being the final authority to accept or reject such votes.

Apart from this, the invalid votes had become a tool of ‘managing’ the elections and also a major point of controversies. One can learn from the example of 1998 assembly elections in Madhya Pradesh, where the party in power won a simple majority of 172 in a house of 320 seats.

The surprising factor was that in about 35 of these seats, the margin of victory was less than the total number of invalid votes!

Now with the EVMs, any possibility of such controversy is over.

A countless number of PhD theses can be written if you just go through these systematic malpractices that political parties and powerful candidates in power used to adopt just to win the election over and again. No wonder that till 1998 more than 90 per cent of election verdicts were “pro-incumbent”.

But right after the induction of EVM, more than 90 per cent of election verdicts have been anti-incumbent.

Keeping this in mind, I picked up the 1998 Madhya Pradesh election as a case study.

In 1998, all the assessments were expecting a victory for the Bhartiya Janata Party (BJP) in Madhya Pradesh due to routine anti-incumbency, but it was the Congress that emerged triumphantly.

The state was still undivided, and after all the 320-member assembly seats were declared, the ruling Congress party had won 175 seats and the BJP was a distant second with 118 seats. The Bahujan Samaj Party (BSP) won 11 seats, matching its strength in the outgoing assembly. Several top BJP leaders, including former chief minister Kailash Joshi and Vikram Verma, leader of the opposition in the state assembly, drowned at the Digvijay Singh surprise victory.

More importantly, the Congress Chief Minister had said well in advance, that the elections were swayed by his “management”. And he did prove everyone wrong, by overcoming the huge anti-incumbency against him and pulled off a miracle. Yes, it was nothing short of a miracle.

After being ahead in 220 assembly constituencies just a few months back in the February 1998 Lok Sabha election, most of the BJP's traditional strongholds fell to the Congress. The BJP could win only 45 per cent of the seats in Chhattisgarh (where it expected to win most of the region's 90 seats), 35 per cent of the seats in Madhya Bharat and 43 per cent of the seats in the Mahakaushal region. Only in the Gwalior and Vindhyachal regions did the BJP do better than the Congress.

In Chhattisgarh, the Congress secured 46 of the 90 seats while the BJP was far behind with 33 seats. The Congress captured 124 of the 230 at stake outside the Chhattisgarh region as against the BJP's 103. Old Bhopal was the only region in the state where the BJP secured better results. The BJP won eight of the 12 seats in the state capital and its adjoining areas. The Congress won four, wresting two prestigious seats - Bhopal North and Bhopal South - from the BJP.

The region-wise break-up shows that the Congress won 58 of the total 99 seats in the Madhya Bharat region, against 33 by the BJP, six by the Bahujan Samaj Party and two by others. In the Mahakaushal region, the ruling Congress bagged 45 of the total 76 seats at stake. The BSP failed to open its account in the region. The Congress captured 17 of the 43 seats in the Vindhya region, against 15 by the BJP.

Just one per cent of the vote separated the BJP from the Congress - 40 to 39 per cent - but in Madhya Pradesh, a one per cent swing was enough to ensure victory. The anti-incumbency factor, which the BJP was counting on heavily, did not work in the state. This, political analysts feel, is because the Congress negated it by refusing tickets to many sitting MLAs who did not have a 'positive image'. The BJP, on the other hand, awarded tickets to many of its sitting MLAs .

1998 assembly polls

Number of invalid votes: 4,60,000 (about 2 per cent)

Seats 320; thus 1,500 votes on an average for each assembly seat. There were about 4,000 missing votes as well in the state.

The “Invalid” factor in Digvijaya magic

Remember, in 1998, India was still voting on ballot papers. The ECI had not introduced the EVM. In that context, the factor, which worked for the Congress, was of the invalid votes. Digvijaya Singh managed to win 35 tight seats where the margin of victory was less than the number of rejected or the invalid votes.

For example, take the case of Keskal assembly segment. Here the Congress won by a margin of just 343 votes, but the number of rejected votes or the invalid votes was as high as 2561 after the final recounting and adjustment of these votes. Till the penultimate round, the number of such votes happened to be around 3000, which shows how much the last round of recounting had tilted the result in favour of the Congress candidate. Same was in the case of Dhar.

Another example is Konta assembly segment. Here the Congress won by a margin of just 748 votes, but the number of rejected votes or the invalid votes was as high as 2906. This recounting adjustment was proven in 35 assembly seats where the invalid votes finally decided the winner.

The number of rejected votes or the invalid votes are usually low after the final recounting and adjustment of these votes. Till the penultimate round, the number of such votes happened to be high, which shows how much the last round of recounting tilt the result for an individual candidate.

The number of invalid votes has always been very high in Madhya Pradesh elections. If the trend of such votes is analysed, we find that the in the Nineties when there were in all seven elections (three assembly and four Lok Sabha elections), the 1998 elections had the least number of rejected votes.

This actually indicates that 1998 assembly elections witnessed the best management of such votes in the electoral history of Madhya Pradesh. What was more fascinating is the fact that all these seats were very well scattered and they could not account for a typical geographic cluster, which could have shown any deviations in a computer data analysis at the regional scale.

Hence the losing candidates could not put up a collective objection to this validity of the valid votes or should we say invalidity of the valid votes.

Digvijaya Singh was right. He knew the art of managing the elections well. He did manage to win not one, not two, but a big tally of 35 such seats and that resulted in him becoming the chief minister once again.

Now imagine what would have been the results, if the phenomenon of invalid votes had been absent in the 1998 elections. The opposition would have got a clear majority, and after the formation of Chattisgarh, the lead of opposition in Madhya Pradesh would have been further increased.

What Changed in 2003?

ECI said electronic voting machines (EVMs) would be used in all the 41,000 polling booths in the next assembly elections in the state. So, the option of the invalid votes was unavailable.

Not only that, the voters were required to carry some identification, be it photo identity cards or any other document specified by the ECI. Without photo identity, the voter was not allowed to cast his/her vote, and the EC had the court's permission for this arrangement. This negated the option of increased voter list because the person on the list had to come up with a proper photo ID.

It is important to note, that when in 2003 the EVMs were introduced, the number of invalid votes dropped down to zero. And this resulted in zero manipulation in all these close seats where invalid votes were the key factor. There was no scope left for Digvijaya Singh to manage the election anymore. And needless to say that this resulted in the acute loss for Congress in all those seats, which were in a limbo due to the equation of invalid votes.

The "Invalid Syndrome" is valid even for Lok Sabha

Invalid votes made all the difference in 69 constituencies in 1998 Lok Sabha polls. Blame it on voters' illiteracy or antipathy, or official management, but invalid votes were a factor to be reckoned with in as many as 69 Lok Sabha constituencies, where it surpassed even the winning margin of candidates!

This dubious phenomenon was clearly seen in Rajmahal constituency in Bihar where BJP candidate Bom Marandi defeated his nearest Congress rival by just nine votes, while the number of invalid votes cast was a whopping 10,432.

A prominent victim of invalid votes, which are otherwise scrap-heaps of the wasted ballot paper, was former deputy prime minister and Haryana Lok Dal nominee Devi Lal who lost the election in Rohtak to his Congress rival by 303 votes. The number of invalid votes in this constituency was 8,225.

Another cliff-hanger was witnessed in Tirupattur constituency in Tamil Nadu, where the Dravida Munnetra Kazhagam (DMK) candidate managed to win by a margin of 274 votes. The number of invalid votes polled here was 25,250.

According to analysts, had these invalid votes been properly cast, the outcome could well have been different in all the constituencies.

Of the 69 constituencies where invalid votes played a crucial role, Uttar Pradesh accounts for 11, followed by Maharashtra (eight), Tamil Nadu (seven), Andhra Pradesh and Gujarat (six each), Bihar , Madhya Pradesh and Rajasthan (five each), West Bengal and Karnataka (three each), Assam, Haryana and Kerala (two each) and Goa, Mizoram, Punjab and Andaman and Nicobar (one each).

The two major beneficiaries of invalid votes were the BJP and the Congress. While the BJP romped home in 21 of the 69 constituencies, the Congress did so in 20. Incidentally, the two parties were also the major victims of this phenomenon, with the Congress and the BJP losing in 24 and 18 seats respectively out of the 69.

For instance, the BJP lost the election in Andaman and Nicobar to the Congress by 544 votes while the number of invalid votes here was 1,850. However, the party's lone Muslim candidate, Mukhtar Abbas Naqvi from Rampur, defeated his Congress rival by 4,936 votes, even though the number of invalid votes cast here was 9,510.

A prominent Congress loser was former union minister Santosh Mohan Deb who lost to the BJP candidate in Assam's Silchar constituency by 19,566 votes, while a whopping 25,957 votes were declared invalid.

Similarly, BJP heavyweight Vinay Katiyar had the mortification of losing the Faizabad seat to Mitrasen Yadav of the Samajwadi Party by 7,737 votes; the invalid votes cast was close to 11,000.

In Nandyal, once the constituency of former prime minister P V Narasimha Rao, Telugu Desam Party (TDP) candidate Bhuma Magi Reddy defeated the Congress candidate by 4,650 votes, the number of invalid votes cast stood at 10,287.

The introduction of EVM has been one of the most important changes that have been introduced in the Indian electoral process over the last 50 years.

The EVMs were introduced with the prime objective of cutting down expenses incurred in the procurement and printing of ballot papers and to also cut down on time spent on the counting of ballots.

In addition to this, one of the biggest benefits from the introduction of EVMs has been the fact that the number of invalid votes has been reduced to zero in this format as the voter’s preferences are recorded only once.

Unlike ballot papers, where a considerable number of voters put their stamp near the line dividing two candidates, the choice of the voter is reflected with clarity, eliminating the possibility of manipulation at the counting level.

Apart from this, the invalid votes had become a tool of managing the elections and also a major point of controversies.

One can learn from the example of 1998 assembly elections in Madhya Pradesh, where the party in power won a simple majority courtesy those 35 seats, where the margin of victory was less than the total number of invalid votes.

Now with the EVMs, any possibility of such controversy is over. In the rural areas, booth capturing has been nullified, as the EVMs can’t be captured. Even if captured, it is tough to use them for large-scale bogus voting as only one vote is released at one point in time by the returning officer. While in the traditional paper ballot system, it was easy to capture the entire bundle of ballot papers, mark them en mass and put them easily in the ballot boxes.

Now the controversy on how the tampered EVMs caused the landslide win for BJP in Uttar Pradesh, where these political parties have won the state election over the last 24 months using the same EVMs. The critical question is: Can the EVMs be hacked and manipulated? The clear answer is, no!

There are so many checks and balances not just in the machine, but also in the entire process of ECI that systematic hacking is next to impossible.

May be I would write a separate piece on that entire process. But at this moment, remember the thumb rule: those who are in power try to abuse the system to remain in power. And when any system results in more than 90 per cent anti-incumbent verdicts; it means only one thing: that the people in power are unable to abuse it.

Like they abused the paper ballot, and more than 90 per cent election verdicts back then resulted in a pro-incumbent verdict.

Think about it, and you will get the valid answer from the invalid votes.

This piece was first published in Offprint and has been republished here with permission.