SIR shrinks voter rolls by up to 13.24% in five poll-bound states: what the numbers mean

The headline numbers

The number of voters across Uttar Pradesh, Uttarakhand, Goa, Punjab and Manipur fell by up to 13.24 per cent after the completion of the Special Intensive Revision (SIR) of the voter list, according to data made public by the Election Commission and state chief electoral officers. The five states will go to assembly polls early next year.

Across the five states, the electorate shrank from about 18.71 crore to 16.43 crore after the revision, a net reduction of around 2.29 crore — about 12.2 per cent. That net figure hides a much larger churn underneath it: far more names were removed in the draft rolls, and a substantial number were then restored in the final rolls after claims and objections.

The pattern is not uniform. Uttar Pradesh, which has the biggest electorate among Indian states, alone accounted for nearly 90 per cent of the total drop. At the other end, Punjab and Manipur each lost a little over 6 per cent of their electorates. Uttarakhand and Goa lost about 10 to 11 per cent each. The differences matter because they show that the revision did not hit every state in the same way — and because the restoration rates, the share of draft deletions that were later added back, varied sharply too.

It is also important to be precise about what a fall in the rolls means. A smaller roll does not by itself prove that eligible voters were wrongly removed, and it does not by itself prove that every removal was justified either. Rolls can shrink legitimately when dead voters, people who have permanently moved away, and duplicate entries are taken out. They can also shrink wrongly when living, eligible voters are missed or wrongly classified. The argument over the SIR is, at bottom, an argument about which of those two things explains most of the 2.29 crore.

Uttar Pradesh: where almost the entire fall happened

In Uttar Pradesh, which went for SIR in phase II, the electors went from 15.44 crore before the revision to 12.56 crore in the draft roll, then back up to 13.40 crore in the final rolls. The net loss stood at around 2.04 crore, or 13.24 per cent — the steepest among the five states.

The scale of the churn in Uttar Pradesh is difficult to grasp without the draft-stage figures. The draft removed about 2.89 crore names. About 0.84 crore of those — roughly 29 per cent — were restored in the final roll. In other words, for every ten names struck off the draft, about three came back. The remaining seven account for the net loss.

Because Uttar Pradesh's electorate is so large — its 403-member assembly is the biggest in the country — even a percentage fall that looks moderate translates into enormous absolute numbers. A 13.24 per cent cut in Uttar Pradesh is larger than the entire electorate of most other states. That is why the state dominates every headline about the exercise, and why the political argument about the SIR is fought most fiercely over Uttar Pradesh's rolls.

The state's assembly term ends on 22 May next year, later than the other four states, which gives the Election Commission a longer runway there. But the same rules apply: the Commission can hold polls earlier than six months before a term expires, under the Representation of the People Act, 1950, so the final rolls published after the SIR will be the starting point for the election, subject to the summary revision now under way.

The other four states, one by one

In Punjab, where SIR was held in phase III, the electors went from 2.15 crore to 1.94 crore in the draft, then to 2.01 crore in the final rolls. The net loss was about 0.14 crore, or 6.34 per cent. Punjab had the strongest recovery of the five: about 0.07 crore of the roughly 0.21 crore draft deletions — about 34 per cent — came back to the rolls.

In Uttarakhand, also a phase III state, electors went from 0.80 crore to 0.71 crore, ending at 0.72 crore. The net loss was about 0.08 crore, or 10.07 per cent. But Uttarakhand's restoration rate was strikingly low: only about 0.0025 crore returned, around three per cent of draft deletions. That contrast with Punjab is one of the most important details in the data, because it raises the question of why the claims process restored so many names in one state and so few in another. The published data does not by itself answer that question — it may reflect differences in how many draft deletions were genuinely wrong, differences in how actively voters filed claims, or differences in administration.

In Manipur, where SIR was held in phase III, electors went from 0.209 crore to 0.193 crore, ending at 0.196 crore. The net loss was about 0.013 crore, or 6.33 per cent, similar to Punjab. About 17 per cent of draft deletions were restored.

In Goa, where SIR was held in phase II, the electors went from 0.1185 crore to 0.1085 crore, then fell further to 0.1058 crore. The net loss was nearly 0.013 crore, or 10.76 per cent. Goa is the only one of the five where the final roll is smaller than the draft, so there was no net restoration at all. That makes Goa the exception to the pattern seen elsewhere, where the final roll recovered at least part of the draft-stage deletions.

Why rolls shrink: draft, claims, and final lists

Understanding the SIR numbers requires understanding the three stages. The pre-SIR roll is the starting list. The draft roll is published after house-to-house enumeration, and it is where the large deletions appear. The final roll follows a period for claims and objections, during which voters who believe they were wrongly removed can seek restoration, and officials can correct errors.

The gap between draft and final is therefore a rough measure of how much of the first cut did not survive scrutiny. In Uttar Pradesh, 29 per cent of draft deletions were restored. In Punjab, 34 per cent. In Manipur, 17 per cent. In Uttarakhand, about 3 per cent. In Goa, none — the final list shrank further instead. These are very different outcomes, and they are the reason both sides can point to the same exercise and draw opposite conclusions: one side sees restorations as proof the system corrects itself, the other sees low restoration rates as proof that wrongly removed voters never got back on.

The Election Commission's stated purpose for the SIR is to remove the names of deceased voters, people who have permanently shifted residence, duplicate entries, and non-citizen electors, while ensuring that no eligible citizen is left out. That is the standard the exercise should be judged against. The difficulty for outside observers is that the published headline figures — pre-SIR, draft, and final totals — do not by themselves show how many of the net deletions fall into each of those categories in these five states. Without that breakdown, the argument turns on competing inferences from the totals, from individual cases reported in the press, and from the restoration rates.

One further wrinkle matters for interpretation. The Commission has also ordered a special summary revision in 17 states. The Commission has said this will ensure that left-out voters get a fresh chance to be included, and that young voters who turn 18 on or before 1 January will get an opportunity to be included and to vote in these five states. So the final SIR rolls are not necessarily the last word before polling day — but they are the baseline from which the summary revision starts.

The opposition's case

The opposition's case is that the scale of the deletions is itself evidence of something wrong. At the INDIA bloc's rally at Ramlila Maidan in Delhi on 1 November, Congress leader Rahul Gandhi accused Prime Minister Narendra Modi of stealing votes and demanded the resignation of Chief Election Commissioner Gyanesh Kumar, saying, in his words, that Kumar would be removed from the post and would go to jail.

Gandhi's broader claim, as reported from the rally, is that about 130 million voters — roughly 15 per cent of India's more than 960 million eligible voters — were removed from the voters' list after the 2024 general elections, and that the difference between the ruling alliance and the opposition in vote share was just three per cent. He argued that if the deleted voters were not legitimate, then the 2024 election itself would be called into question. These are the opposition's claims, not established findings, and they are denied by the Election Commission and the ruling Bharatiya Janata Party.

Critics of the exercise have also argued that it has disproportionately affected minorities, particularly Muslims, and other marginalised groups — an allegation the Commission and the BJP deny. Some critics point to procedural objections reportedly raised by the two other election commissioners, and to media reports of discrepancies in the exercise, as reasons to doubt its fairness. Those reports and objections are themselves contested, and the Commission has asserted that its SIR orders were approved unanimously by the three-member panel.

It is worth separating three different opposition arguments that are sometimes run together. The first is about scale: that deletions on this order cannot all be genuine. The second is about process: that enumeration, documentation, and the claims window disadvantaged some voters. The third is about intent: that the exercise was designed to help the ruling party. The first two can in principle be tested against data and individual cases. The third is a political charge that the Commission denies, and it cannot be settled by the roll totals alone.

The Election Commission's defence

The Election Commission has defended the SIR as a routine clean-up of the rolls, aimed at removing the names of deceased, permanently shifted, duplicate and non-citizen electors while ensuring that no eligible citizen is left out. The BJP has similarly denied the allegations of voter fraud and has challenged the opposition to explain why it questions an exercise of a kind that has been carried out repeatedly since the 1950s.

Union minister Kiren Rijiju, responding to the opposition's criticism, asked whether the names of illegal immigrants, dead voters, shifted voters and duplicate entries should be removed or not, and pointed to earlier intensive revisions conducted in 1952-56, 1958-61, 1965-66, 1983-84, 1987-88, 1993, 1995 and 2002-2004, asking whether those exercises were illegal. The argument is that large-scale roll revision is a normal part of electoral administration, not an innovation designed for one election.

The Commission's defenders also point to the restoration figures as evidence that the process works: in Uttar Pradesh alone, about 84 lakh names removed in the draft were restored in the final roll, and in Punjab about a third of draft deletions came back. On this reading, the draft is meant to be a provisional list that is then corrected, and the final roll is the meaningful figure.

The Commission has also taken a procedural step that cuts across the argument: the special summary revision in 17 states, which it says gives left-out voters a fresh chance to enrol, including young voters turning 18 by 1 January. Whether that revision restores significant numbers in the five poll-bound states will be one of the clearest tests of both sides' claims in the coming weeks, because it will show how many people who were left off the final SIR rolls come forward and are accepted.

What it means for the 2027 elections

The five states go to polls early next year, with the terms of the Manipur (60 members), Goa (40), Punjab (117) and Uttarakhand (70) assemblies ending on different dates in March 2027, and Uttar Pradesh's 403-member assembly term ending on 22 May 2027. The rolls that emerge from the SIR and the summary revision will decide who is entitled to vote in those elections.

A smaller roll changes elections in ways that go beyond the headline percentage. In close constituencies, even a small number of wrongly excluded voters can matter, which is why the opposition's focus on individual cases of alleged wrongful deletion is politically potent even if the aggregate numbers are dominated by genuine clean-up. Conversely, a bloated roll with large numbers of dead, shifted or duplicate entries also distorts elections, which is why the Commission's clean-up argument resonates with those who see roll accuracy as a precondition for fair polls.

Three things are worth watching before polling day. First, the outcome of the special summary revision: how many names are added in each of the five states, and whether Uttarakhand's and Goa's unusually low restoration rates change. Second, the courts: SIR-related petitions are before the Supreme Court, and any ruling on the process could affect how the rolls are treated. Third, the state-level breakdowns: as chief electoral officers publish more detailed category-wise data — how many deletions were deaths, shifts, duplicates and other reasons — the argument should become more evidence-based and less a contest of competing totals.

For voters in the five states, the practical point is immediate and non-partisan: anyone who believes they were wrongly left off should check their name on the roll and use the summary revision window to file a claim. Whatever the political argument about the SIR, the rolls that will be used in early 2027 are still, for a short time, open to correction.

Editorial disclosure

This article was researched with AI assistance and reviewed by the publisher before publication. Report factual errors through our contact page.