"Here is proof, from a natural experiment, that SIR carries a fatal design defect, The Indian Express", 30 December 2025
Yogendra Yadav
Here is proof, from a natural experiment, that SIR carries a fatal design defect
Rahul Shastri
Yogendra Yadav
Scientists call it a “natural experiment”. When something happens on its own, not by design, and allows us to compare two objects as if an experiment had been run. The sudden lockdown, for example, allowed us to see how blue our skylines would be if industries and vehicles were to stop. We didn’t stop industries for this comparative study. It just happened.
We stumbled upon something similar a few days ago. Not one but two natural experiments have occurred in the course of the second phase of the SIR. We anticipated one of these — the unusual revision in Assam. “Great, just as we expected”, was our first WhatsApp exchange with each other when the CEO of Assam released the draft voters’ list. The second case was a chance discovery, a news clipping about panchayat elections in UP.
Both these experiments now provide rigorous scientific proof, or as close to it as we can get in real life, that the SIR carries a fatal design defect. The problem with the SIR, as we have kept saying, is not in the idea of an intensive revision of the voters’ list. The problem is with the two unprecedented and unnecessary design elements in the SIR —compulsion of enumeration forms and the requirement of proof of citizenship. These two experiments show that once both these elements are removed, voter list revision does not lead to mass disenfranchisement.
Let us begin the story at the starting point, the SIR in Bihar. The publication of the draft list shocked everyone as 65 lakh names were dropped. Eventually, thanks to the Supreme Court-enforced damage control, the net deletions came to 44 lakh. Everyone had their private theories. Surely, Bihar has too many migrants with duplicate votes. Maybe the original list was fudged and inflated due to administrative inefficiencies in Bihar. Or perhaps Bihar experienced the teething troubles of the new SIR method. Careful analysis would have busted each of these theories. But very few had patience for a nuanced reading, especially after the Bihar election results.
The second phase of the SIR in 12 states and UTs has confirmed that the disenfranchising effect was not specific to Bihar. As the accompanying table shows, the extension of this exercise to over half of India has resulted in the exclusion of 6.56 crore names from the existing voters’ list, slightly higher than our provisional estimate of 6.28 crore in this column. (Rahul Shastri and Yogendra Yadav, IE, December 16). We had also red-flagged that the sword of disenfranchisement is hanging over the head of another 5.3 crore (now around 3 crore after massive “adjustment” in UP) electors who would get a notice to prove their citizenship.
The data from the second phase shows many disturbing patterns. It confirms that the SIR-induced deletion of names has nothing to do with migration. Voters’ lists have shrunk drastically in the in-migration states like Gujarat and Goa as they have in out-migra tion states like West Bengal and UP. States like Gujarat and UP with deflated voters’ lists (fewer electors compared to the adult population) have seen mass deletions as have states like Tamil Nadu with inflated voters’ lists. It also shows that women have borne the brunt of the SIR disenfranchisement. The gender ratio in the voters’ list has declined in every state after the SIR.
That leaves one question: How do you know that a different method would not have resulted in disenfranchisement? We had no answer so far. But now we know, thanks to the natural experiment in Assam.
This was the only state where the ECI junked the SIR method used across the country. Voters’ lists were revised at the same time as others but through a simple and rigorous method of house-to-house physical verification. No enumeration forms were required. No one was asked to prove citizenship (a little odd for the only state that has a serious citizenship issue). And the ECI issued meticulous instructions about carrying out this exercise.
Here is the outcome. Assam started its revision with 2.52 crore persons on its voters’ list. As many as 10.56 lakh names were deleted and 10.55 lakh names added. So the number of votes on the recently declared draft rolls is exactly 2.52 crore. In other words, Assam is the only state so far with zero per cent exclusions. Compare it with other major states in the accompanying graphic, each with anything between 8 to 19 per cent deletions. Clearly, not shifting the onus onto the voter and non-insistence on citizenship papers has led to non-disenfranchisement.
That still leaves some scope for doubt: Are we comparing mangoes with mangoes? Can we see the outcome of these two methods in the same state at the same time?
This freak natural experiment happened in UP. While the ECI was carrying out the SIR in UP, the State Election Commission (SEC-UP), an independent constitutional authority, had already been revising its own voters’ list for the panchayat elections. Yes, two constitutional bodies were undertaking parallel exercises to map the same set of persons, the eligible electors of Uttar Pradesh, at the same time. The only difference was that for the panchayat elections, there was no requirement of enumeration forms or citizenship papers. The SEC-UP released its draft list on December 5.
Now just look at the difference. Uttar Pradesh has an adult population of 16.1 crore as of December 2025. The number of persons on the ECI’s voters’ list before the SIR was already low, 15.4 crore. Post the first phase of the SIR, it has dropped to 12.6 crore. Funnily, the total rural electorate listed by the SEC-UP now stands at 12.7 crore, more than the entire electorate of the state as per the ECI. Add 3.4 crore urban voters (declared by the SEC-UP in 2023) to the rural, and the total electorate for local body elections in UP is 16.1 crore. So, we have two different methods that give us two staggeringly different numbers for the electorate — the SIR shows it at 12.6 crore, non-SIR shows it at 16.1 crore, exactly the same as the state’s projected adult population.
The only possible conclusion is that the problem lies with the SIR. While we wait for social scientists to work carefully on this natural experiment, the basic conclusion stares us in the eye: SIR is a disenfranchising monster let loose on the Indian voters.
____________________________
Yours,
