Here is a simple yet critical question for the ongoing SIR exercise: What proportion of Bihar’s adult residents possess at least one of the 11 documents that the Election Commission of India (ECI) asks of them? This is a vital question as this would decide who can become a voter, first in Bihar and then in the rest of the country.
Finally, we have an official, if bizarre, response from the ECI, filed in the Supreme Court as a counter affidavit (ECI-CA). Believe it or not, the ECI submits that the total number of eligibility documents is approximately three times the 7.9 crore potential voters of Bihar.
What, pray, is the basis of this extraordinary claim? Fortunately, the ECI provides a statistical table (ECI-CA: pages 35-39) on each of these documents. While its conclusions are dubious, it does provide some official data otherwise not easily available in the public domain. Here we fact-check the claims made by the EC and revise the initial estimate offered by one of the authors. (Rahul Shastri, The Hindu Data Point, August 1, 2025)
The accompanying table analyses ECI-CA claims for each of the documents. Column 2 reports figures claimed by the ECI. Since it has not presented the figures as a proportion of the population, we have done so in brackets against each figure in per cent of the total potential voters. Column 3 presents the actual position based on publicly available statistics, including the ones used by ECI-CA. Figures in Column 3 are presented as the percentage of the eligible electorate of 18-40 years, since this is the segment of the potential electorate whose names may not figure in the Electoral Roll (ER) of 2003 and hence who may be required to submit one of these 11 documents.
The first category of the six documents does not need much comment as these are either irrelevant or negligible, and in any case, they are not contested. The NRC does not apply to Bihar. Since the requirement for any I-card or document issued by the government, LIC or PSU is that it must have been issued prior to 1987, the ECI admits that it is virtually non-existent. Similarly, since the population of Scheduled Tribes in Bihar is merely 1.3 per cent and there were only 191 applications under the Forest Rights Act, this is negligible. The ECI data is a slight overestimate for government employees’ I-cards, as they include pensioners who are not relevant to the 18-40 age group. As for passports, the ECI data may be a small overestimate, as it includes invalid passports of those who have passed away.
The figures quoted by the ECI for matriculation certificates are for irrelevant reference years (1980 to 2025) and need to be restricted to 2001-2023. Thus, our estimate of 2.06 crore is much less than the 2.91 crore claimed by the ECI. However, since the proportion of matriculation is higher among the relevant age group (18-40 years), our estimate of 43.3 per cent is higher than 36.8 per cent for the entire adult population used by the ECI or that of 14 per cent for the entire population used in some other estimates. This may go up a few points if CBSE and ICSE matriculates are added, but that may be balanced by discounting a significant number of permanent migrants in this category.
Let’s now turn to the remaining five documents where the ECI-CA makes astronomical claims with scant basis. The claims for permanent residence certificates are in the same breath, titanic, self-contradictory and deluding. In the first sentence, the ECI-CA states, “Permanent residence certificate is not issued in Bihar”. Therefore, the relevant number should be zero. Yet in the next sentence, it claims — without citing any source — that 13.9 crore residence certificates have been issued in Bihar from 2011-25, a number which exceeds Bihar’s population. The claim is untenable for another reason: Residence certificates issued in Bihar have a validity of one year, and hence every year, lakhs of people reapply for them.
On the caste certificate, similarly, the ECI makes indefensible claims on the basis of total certificates issued during 2011-25 and not the number of unique persons to whom they have been issued. A person can apply simultaneously for OBC certificate for the state as well as the central government. The latter is valid just for a year. Besides, as caste certificates are mainly used by those aspiring for higher studies or government jobs, there is no reason to believe that the proportion of 18-40 year olds with caste certificates will be half of those with matriculation. The India Household Development Survey II had estimated in 2011-12 that only 16 per cent households had someone with a caste certificate.
The ECI’s claim on family registers makes a mountain out of a molehill. There is a valid but small number of 15.8 lakh individuals with Vanshavali certificates issued by the Panchayati Raj department (5.8 lakh in 18-40 age group). To this, the ECI adds 2 crore names contained in the Vikas Register 2.0 of the Bihar Mahadalit Vikas Mission, an internal administrative register to which no individual except government officials have access.
The ECI uses a grossly misleading method of counting all the birth certificates issued between 2001-24, whereas in the same paragraph, it concedes that those born after 2007 are not eligible to be on the electoral rolls of 2025. The correct figure is less than 2 per cent. NFHS-3 reported that just 2.8 per cent of births between 2001-05 had corresponding birth certificates; this proportion is bound to be much smaller for our cohort group (1985-2007), the majority of which was born before 2001.
In the house allotment certificate, the ECI adds up 1.18 crore beneficiaries of the Pradhan Mantri Aawas Yojanas (Gramin and Urban) and the Indira Aawas Yojana. But these beneficiaries receive a sanction letter and not a house allotment certificate. They do not qualify for the “land/house allocation documents” required under SIR.
What then is the realistic overall number of persons who may have at least one of the 11 documents? We must remember that the figures given in the third column are not additive but overlapping; there is a negligible proportion of those who are government employees or have passports or have birth certificates but are not amongst the 43.3 per cent who are matriculates. Of those in the table who may not be matriculates are only the landless allotted land by the government (1.2 per cent) and those included in the family registers (1.2 per cent). Hence, the proportion of individuals in the 18-40 age group with eligibility documents is estimated to be around 45.7 per cent (43.3 per cent + 1.2 per cent + 1.2 per cent) plus some exceptions like a small proportion of those who possess caste certificates but are not matriculates. Let us round it off, on the upper end, to 50 per cent.
The conclusion is stark: If the ECI strictly adheres to its arbitrary decision of asking for one of the 11 documents specified in its (“indicative but not exhaustive”) list, at least 50 per cent of those whose names do not figure in the Electoral Rolls of 2003 may face deletion from the voters’ list. As per the latest estimate (based on unofficial data from two districts), the ECI has been able to locate less than one-third of the electors on the 2003 rolls. Thus, the remaining two-thirds, 4.8 crore, would be asked for documents. Unless the Supreme Court intervenes and the ECI is made to modify or relax this requirement, 2.4 crore of the 7.24 crore electors on the current Draft Electoral Rolls face a loss of voting rights. This would be the largest disenfranchisement ever recorded in the history of any electoral democracy.
Shastri and Yadav work with the national team of Bharat Jodo Abhiyaan. Yadav has filed a petition in the Supreme Court challenging the SIR
What is the realistic overall number of persons who may have at least one of the 11 documents? (Illustration by C R Sasikumar) 
Share
