Are all votes equal, or do some matter more? Does every Indian have an equal say in influencing Parliament? The answer a straightforward “no".

There are some constituencies that are much bigger than others (which means individual voters had less say than those in other, smaller constituencies).

In this edition of Election Metrics, we will look at identifying these big and small constituencies, and where they are.

We will also examine which states have disproportionate representation in Parliament, and the impact of delimitation (drawing or redrawing the boundaries of constituencies) between the 2004 and 2009 elections.

2004

Prior to 2007, the last time delimitation of Lok Sabha constituencies happened was back in the 1970s. Thanks to the large time gap, there was significant variation in the size of constituencies in the 2004 election.

In the parliamentary election that year, the average constituency had about 1.28 million voters. This, however, did not mean that all constituencies had close to that number of voters. If you lived in the Thane or Outer Delhi constituencies, for example, your vote was worth approximately one-third of what it was worth in the average constituency. The table shows the top 10 biggest constituencies in 2004.

Notice that all these constituencies included the outskirts of large cities.

What about the smallest constituencies in 2004?

Unsurprisingly, they are all in either Union territories or small states. By definition, a parliamentary constituency is not supposed to spread over two states. Hence, some of the smaller states and Union territories (UTs) end up getting disproportionate representation in Parliament. In 2004, for example, Lakshadweep had a little more than one-hundredth of the voting population of Outer Delhi (the largest seat). In other words, the average voter in Lakshadweep had close to 100 times the influence in Parliament as the average voter in Outer Delhi.

A couple of seats stand out in this list of smallest constituencies, however. Ladakh, in Jammu and Kashmir, was one of the smallest constituencies. However, it was the biggest parliamentary constituency in terms of area, and its vastness and cultural homogeneity (and difference from the rest of Jammu and Kashmir) meant that it was a separate constituency.

Also interesting is the presence of Chandni Chowk in this list of smallest constituencies, given that Outer Delhi is the largest constituency. This shows that the distribution of votes across seats in Delhi was the most lopsided in 2004.

Delimitation and 2009

Let us start off by looking at the biggest constituencies in 2009.

The largest constituency in 2009 is much smaller than that in 2004 (showing the delimitation had some impact), but one thing doesn’t change—the large constituencies are all in big cities.

The biggest constituency in the 2009 general election was Malkajgiri, which is a part of Hyderabad. The four constituencies in and around Bangalore come next, which shows how much growth took place in that city in the 2000s. An unlikely candidate in this “biggest constituencies" list, however, is Uttar Pradesh’s Unnao, which is a rural constituency. Does this suggest bad delimitation or possible gerrymandering?

Not much changed in the list of small constituencies, though. The delimitation worked well for Delhi, and thus Chandni Chowk went out from the list. However, apart from Ladakh, the rest are constituencies in small states and Union territories.

It is interesting to note that putting together the two constituencies of Arunachal Pradesh, the total size is less than the size of the average constituency in 2009 (1.35 million)—indicating that Arunachal Pradesh is clearly over-represented in the Lok Sabha.

It is not the only over-represented state, though. For example, among states with just one parliamentary seat, only Nagaland has the appropriate representation—the rest are all over-represented.

Leaving aside these states, though (since we don’t have a choice, given constituencies cannot transcend state boundaries), let us look at how many seats each state ought to have by virtue of its size, and what the actual representation is.

For states that have more than one seat, we look at the ratio of the average number of voters per seat in that state to the national average number of voters per seat. This ratio is the “factor" in the table. This factor shows the degree to which a particular state has been under- or over-represented in the Lok Sabha (the higher the factor, the greater the degree of over-representation). Using this factor, we see how many seats each state should theoretically have (a minimum of one).

Among the large states (those that have at least 10 seats), we find that the most over-represented state in the Lok Sabha is Tamil Nadu, with 39 seats. Going by the number of voters in the state, it should only have 31 seats. At the other end of the spectrum, going by their number of voters, Uttar Pradesh should have 88 seats (currently 80) and Maharashtra 55 (currently 48).

This lopsided distribution is because in the 1970s the number of seats in the Lok Sabha for each state was frozen, the ostensible reason being that states that do well on the family planning and population control front (this being a hot topic in the 1970s) should not be penalized for their efforts. What the freeze didn’t take into account, however, was that states can also grow due to immigration. If you look at the above table, you will notice that the states that are most under-represented in the Lok Sabha (in percentage terms) are Karnataka, Maharashtra and Delhi, states that house India’s three biggest cities, all of which have grown due to massive immigration.

Close