In the ongoing debate on delimitation, a conflict between a principle of democracy and one of federalism has become apparent. ‘One person, one vote, one value’ is a principle of India’s electoral system. What this should mean in practice, at a minimum, is that all Lok Sabha constituencies should have roughly the same number of electors.
The Indian republic is constituted as a “union of States”. States are the constituent units, and they have their individual identity and power. The delimitation debate, therefore, has to take into account the relative weights of individual electors and individual States. It has to be about federal democracy.
A two-step process
Article 81(2) of the Constitution takes into account federalism and democracy by providing for apportionment of Lok Sabha seats in a two-step process. The first step is to divide them among the States. It is in the second step that they are divided into constituencies within the States. Article 81(2) (a) provides allotment to each State in “such manner that the ratio between that number and the population of the State is, so far as practicable, the same for all States.”
In Article 81(2) (b), it is provided that “each State shall be divided into territorial constituencies in such manner that the ratio between the population of each constituency and the number of seats allotted to it is, so far as practicable, the same throughout the State.” The qualifier “so far as practicable” in both clauses allows for deviations, and delimitation acts go more into specifics.
The 84th amendment to the Constitution in 2001 and the 87th amendment in 2003 combined (when Atal Bihari Vajpayee was Prime Minister) separated this two-step process in more definitive terms, by basing them on two separate population counts.
The apportionment among States was linked to the 1971 Census; and the apportionment within States was based on the 2001 Census. In the current text of the Constitution, Clause 3 of Article 81 defines the reference to “population” in Clause 2 in two ways, as follows: “for the purposes of sub-clause (a) of clause (2) and the proviso to that clause, as a reference to the 1971 Census” and “for the purposes of sub-clause (b) of clause (2) as a reference to the 2001 Census.”
The Delimitation Commission appointed in 2002 followed this criteria and its recommendations were implemented in 2008 when Manmohan Singh was Prime Minister. The next inter-State redistribution of constituencies was barred until “the first Census taken after the year 2026 has been published,” by the 84th amendment.
Distribution of representation
Now, compare this distribution of representation to the distribution of land wealth in a joint family which has several constituent familial units. India is akin to a joint family and States, its constituent familial units. Each unit was given its share of land according to the size of its membership — the more the members, the higher the share. Each unit further divided their share among their individual members.
Over time, some units had more members, and this caused a decline in the share for their individual members; some others had fewer members, resulting in an increase in per person holding. That is what has happened to the value of votes across States. In 1967, every member in the Lok Sabha represented roughly 4.2 lakh-5.3 lakh electors across all major States. But in 2024, one MP from Kerala represented around 13.9 lakh electors, while in Uttar Pradesh and Bihar, it was 19.3 lakh each (all post bifurcation), and in Rajasthan, 21.4 lakh.
This means vote values of electors across States are not equal any more. To understand this better, let’s pick a middle point, called the national median, which is the population per MP of India. We assign this median a vote value of 1. Then, we calculate the value of vote in each State by comparing it to this median.
For example, if a State’s MP represents twice as many people as the median, its vote value is 0.5 — because one MP is doing the job of two. If another State’s MP represents half as many people, its vote value is 2 — meaning its people get twice the representation compared to the average.
Chart 1 and 2 illustrate the vote value of 20 major States in 1967 and 2024. In 1967, the variation in vote value across these States was relatively modest.
Chart 1 | The chart shows the vote value of 20 major states in 1967
By 2024, the gap from the median had widened significantly, with some States seeing a sharp rise in vote value and others a noticeable decline. Kerala’s vote value was 30% higher than the national median, followed by Tamil Nadu (13%), Odisha (12%), and Punjab (9%).
On the other hand, Rajasthan’s vote value was 16% lower than the median, and Uttar Pradesh’s and Bihar’s was 7% each below the median. When the relative size of the population within each unit increases, the value of each vote declines — as it happens with the landholding that gets divided, and vice-versa.
Chart 2 | The chart shows the vote value of 20 major states in 2024
If Lok Sabha constituencies are delimited based on current population, and regardless of State boundaries — i.e., when all votes across India are given the same value — the apportionment between States will undergo major changes. For instance, Rajasthan, which today has 4.6% of Lok Sabha members, could get 5.5%, while Kerala, which has 3.7%, will see its share shrink to 2.8% (Chart 3).
This data captures this conflict of two principles in India’s federal democracy — the relative representation of States, and individual voters.
Published – April 10, 2025 07:00 am IST