Elections to the Lok Sabha and Vidhan Sabhas take place under the first-past-the-post (FPP) system. Under the FPP, a candidate only needs to get more votes than any other candidate to win a seat. It over-translates the votes of the winning party into seats, which is why the BJP won over 50 per cent seats in Lok Sabha despite winning only 31 per cent of the popular vote in 2014. The FPP also encourages political polarisation because a party with a sufficiently energised base needs to convince a relatively small number of unaffiliated voters in order to win an election. A party opposed by a majority of voters forms the government.
This democratic deficit of the FPP has been long recognised. The main alternative that has been suggested over the years is some version of the proportional system (PS). Under PS, voters do not vote for individual candidates in particular constituencies. Instead, they vote directly for a party. A party that gets 31 per cent of the votes wins 31 per cent of seats, usually subject to a threshold requirement that a party has to secure a basic minimum share of the popular vote (say, 5 per cent) in order to get any seats at all.
Our Constitution does not mandate the FPP, but it does require elections to be conducted on the basis of territorial constituencies. Although it is possible to design a mixed PS system based on territorial constituencies, PS may actually exacerbate polarisation. Under PS, all that a radical party needs in order to win 10 seats in a 100-member house is to convince 10 per cent of the electorate through a hateful and polarising campaign, even if the remaining 90 per cent detest its ideology. What we need is a voting system that goes beyond FPP’s yes-no binary and is sensitive to the strength of voters’ preferences — whether she loves a party, or merely tolerates it, or loathes it.
There is one such system, which would reduce democratic deficit, discourage polarisation and respect territorial constituencies. It is variously called the preferential vote, alternative vote or ranked-choice vote system (RCV), used in Australia, Papua New Guinea and the American state of Maine. Under the RCV, instead of voting for only one candidate, a voter ranks her candidates in order of preference.
Say a voter ranks candidate E as her number 1 choice, candidate C as number 2 and candidate F as number 3. In the first round of counting, every voter’s first-ranked votes alone are counted. If any candidate reaches the 50 per cent mark, she is declared the winner. If not, the candidate with the lowest share of first-rank votes— say, candidate E — is eliminated. The second-rank votes of voters whose first choice was E are now counted and added to the tally of remaining candidates. If a candidate now closes the 50 per cent mark, that candidate wins. If not, this process of elimination and re-distribution is continued until either one candidate crosses the 50 per cent mark, or is the last one standing.
The democratic deficit in RCV is markedly lower than FPP, because the winning candidate secures or comes very close to the 50 per cent mark. It also eliminates the need for strategic voting. An FPP voter whose first preference is the BJP but strategically votes for Trinamool because she wants to ensure the defeat of the CPM candidate can, under the RCV, honestly rank the BJP candidate as her first choice, the Trinamool candidate as her second choice and so on.
The RCV should also reduce political polarisation. Under this system, parties cannot win elections by relying solely on their base. They need not only enough first-rank votes but also a sufficient number of second and third-rank votes. This will require them to build broad social coalitions.
An optional RCV is not only constitutionally viable but even endorsed by the Constitution. When MPs elect our President or Vice-President, they also rank the candidates on the ballot — the Constitution calls it a “single transferable vote” (STV), but in a single-winner voting, there is no difference between STV and RCV. In Indian conditions, a limited optional RCV might be logistically better. Instead of mandating the ranking of all candidates, a voter can have the option of ranking up to three top choices: She may cast only a single-unranked-vote, or rank her top two or her top three candidates. Such a 1-2-3-vote system would deepen democracy and reduce political polarisation.
This article first appeared in the print edition on May 8, 2019, under the title ‘A Mandate More True’. The writer is an associate professor in law at the Universities of Oxford and Melbourne. He is also the General Editor of the Indian Law Review