New Delhi: Israeli scholar Ornit Shani showed in her 2018 book, How India Became Democratic: Citizenship and the Making of the Universal Franchise, how independent India’s first electoral rolls were painstakingly prepared as India moved from limited franchise under the British Raj to universal franchise after Independence.

The officials responsible for drafting the first electoral rolls in what was to become the world’s largest democracy took special care to ensure that the marginalized and the ‘invisible’ sections of the society, such as pavement dwellers and refugees without robust documentation, found a place in the electoral rolls. In the process, the making of the electoral rolls ended up influencing the framing of the Constitution itself, Shani argues, leading to an independent election commission that is not subject to the whims and prejudices of local officials.

Perhaps it is because of this inclusive legacy that India’s election officials have for many years erred on the side of caution in removing names from the electoral register. However, if the complaints throughout most of independent India’s history have been of including duplicate names, and even names of dead persons, today the complaint is one of widespread omissions.

Millions of voters—by some estimates, nearly 2% of the eligible electorate—have been missing from Telangana’s electoral rolls in the latest elections and the deletions have not followed due process, according to election officials. Complaints about omissions from voter lists have also surfaced in other states. These complaints are not entirely without basis, a Mint analysis of electorate and census data suggest.

Every January, the Election Commission of India (EC) revises electoral rolls, the list of eligible voters by constituency, to include those who have turned 18 and new migrants from other constituencies, while deleting the deceased or those who migrated out. However, data from the 2018 electoral rolls suggests that there could be 52 million missing voters on the electoral rolls.

To arrive at this figure, we use the 2011 census data combined with age-wise mortality rates to estimate the current voting-age population. We find that for 2018, there should be 931 million eligible voters but there were only 878 million names on electoral rolls, an under-count of 6%.

According to our analysis, the missing names in the electoral register are largely those of women. The 2018 figures are in stark contrast to previous decades when the number of eligible voters actually exceeded census population totals. In 1991, electoral rolls exceeded the 1991 census count by nearly 5% and this increased to 10.2% in 1999 (compared to the 2001 census).

The under-counting seems to have persisted in 2019, as India heads towards Lok Sabha elections. The latest data from the 2019 electoral rolls reveals that the elector count according to the electoral roll falls short of the census estimates in 12 of the 14 states for which data is available.

The shortfall is the greatest in Haryana (13%) and Andhra Pradesh (12%), but Telangana’s count of voters according to the electoral rolls slightly exceeds our census projections. It is likely that Telangana’s electoral rolls suffer from both wrong inclusions (or duplication) and wrong exclusions, resulting in an overall over-count compared to census estimates.

There are two likely reasons for the decreased counts on the electoral rolls over the past two decades: greater scrutiny in the voter registration process and inadequate registration in the youngest age group.

In 1997, the EC began computerizing electoral rolls and benchmarking rolls against data from the census. In addition, booth level officers (BLOs) were introduced in 2007, tasked with verifying and updating data at the grassroots level.

S.Y. Quraishi, a former chief election commissioner, in his 2014 book An Undocumented Wonder—The Making of the Great Indian Election, argues that these initiatives could have helped improve the veracity of electoral rolls and weed out duplicates. The lower elector roll count could also be a result of declining interest in voting among youth. If we consider the age-wise breakup of the electorate in 2017 (the latest age-disaggregated electorate data made available by the ECI), we observe that in the youngest age group (18-23), there is a huge shortfall in registration with nearly 50% of the age group not counted in the electorate. A gender break-up of this figure was unavailable.

According to Quraishi, the under-counting of young voters reflects both the inability of the EC to register the newest eligible voters and the lack of interest among the youth to engage with the electoral process. The age-breakup also reveals that duplication may still persist: in the age bands of 30-60, we observe that the number of electors still exceed the census count.

Revisions to the electoral roll process may also have exacerbated discrepancies. BLOs, who were instituted to manage voter lists, may not be functioning properly.

According to a 2017 study by the not-for-profit Janaagraha Centre for Citizenship and Democracy, BLOs in major cities are under-paid, ill-trained, and over-burdened.

The BLOs are responsible for the updation of voters’ lists, including the contentious process of linking of electoral rolls with the Aadhaar database. Their lack of training can, therefore, be a serious hindrance.

The government’s National Electoral Rolls Purification and Authentication Programme (NERPAP), which links electoral rolls to the Aadhaar database, has generated further controversy.

Right To Information (RTI) activists have argued on the basis of replies to RTI queries that the deletion of voters from the list, based on Aadhaar, has been done without due verification on the ground by BLOs. According to critics, this lack of verification is behind the deletion of eligible voters in Telangana.

New research also raises other questions about the revisions of electoral rolls. A recent Economic and Political Weekly (EPW) study by Abusaleh Shariff of the US-India Policy Institute, Washington DC, and Khalid Saifullah of the Delhi-based Centre for Research and Debates in Development Policy argued that Muslims tended to be under-represented more than others in several states in the electoral rolls based on their analysis of disaggregated electoral data. If further research corroborates the findings of the EPW study, this could mean that the electoral roll preparation process is beset by systemic biases.

An e-mail sent to the EC on Monday, seeking comments on the issue of discrepancies in electoral rolls, remained unanswered till the time of publishing this story.

The EC has a rich legacy of being independent and inclusive. It must live up to that legacy and remove all doubts about missing voters from the electoral rolls of the world’s largest democracy.

Close