Could a few swing states determine who governs India?

A significant number of Lok Sabha constituencies across India could see very close results

is chairman, InKlude Labs. Read Narayan’s Mint columns at www.livemint.com/ avisiblehand

India is in the midst of conducting its 18th Lok Sabha elections. In the first eight of these Parliamentary polls held after independence, the winning party had more than the required majority of seats to form a government, ranging from a high of 143 more for Rajiv Gandhi of the Congress in 1984 to a low of just 22 over the majority-mark for his predecessor Indira Gandhi in 1967. The reason to look at it this way is that the number of Lok Sabha seats has varied over time, from a low of 489 for the first elections held in 1951 to the current 543, so using a benchmark of the majority’s size allows direct comparison. From the ninth through the 15th general elections in 2009, the winning party was short of a majority, with the deficit ranging from 127 in the 2004 Congress victory for Manmohan Singh to 24 in the same party’s 1991 win that saw P.V. Narasimha Rao become prime minister. Negative majority margins necessitate coalition governments. The pattern changed in the elections of 2014 and 2019, when the Bharatiya Janata Party (BJP) won 10 and 31 seats over majoritymark, respectively.

Will the BJP or Congress maintain the trend of a single party returning more than the required majority to Parliament in 2024? It can only be either of these parties, as they are the only two contesting more than the majority number of 272 seats; BJP candidates have registered in 446 constituencies and Congress leaders in 327. Statistically speaking, the probability of the Congress achieving the halfway mark alone, with an 82% success rate required on contested seats, is very low. In terms of candidates, the third largest party at the moment is the Samajwadi Party (SP), with 62 constituencies. The Bahujan Samaj Party (BSP) has not yet announced its full list of candidates, as India’s multi-phase elections allow later declarations.

There are various types of voting systems around the world, too numerous and too arcane to review here. Suffice it to say that they attempt to balance proportionality, voter choice and local representation. Like the UK, India uses a first-past-the-post system (FPTP). This term is a reference to horse racing, where the first horse to cross the post is the sole winner, whether it leads by just a nose or a whole mile. The candidate with the most votes in a constituency wins, even if the person’s vote share is less than half the total votes cast, as often happens when there are more than two contestants in the fray.

In the 2019 elections, BJP candidates in as many as 224 Parliamentary constituencies won with a voter share greater than 50%, implying that they would have won those seats under any electoral system. An excellent analysis on Print.in puts paid to the unexamined notion that those 224 seats came exclusively from BJP strongholds in five states,

Uttar Pradesh, Bihar, Rajasthan, Madhya Pradesh and Gujarat, which together account for 200

Lok Sabha seats. Instead, it is less concentrated than expected in those states (except Gujarat) and reasonably well spread across other states like Maharashtra, Karnataka, and Haryana. In 2019, counter-intuitively, non-BJP parties secured over half the votes cast in 117 constituencies in states as diverse as Tamil Nadu, Kerala and Andhra Pradesh, but also including Uttar Pradesh and Maharashtra, among others.

One way to predict the outcome of these elections is to notionally award those 224 ‘safe seats’ to the BJP and the 117 others to non-BJP parties (with some adjustments to be made for differences in who is with the BJP this time versus the last), with the remaining 202 constituencies counted as ‘swing seats’ for analysis. But this could yield erroneous results for two reasons. One, the past is not necessarily prologue; and two, the nature (in a structural sense that matters to the FPTP system) of the BJP’s opposition at the state level has changed. For instance, in Uttar Pradesh, the ‘mahagatbandhan’ (grand coalition) between the SP and BSP of the last election has broken down since then, and the SP has allied itself with the Congress in the ‘INDIA’ bloc, even as the BSP fades. In Bihar, Nitish Kumar’s Janata Dal (United) has been allocated 16 seats in the NDA bloc, and it is hard to tell what his constant flip-flops will imply for results. In Maharashtra, there are now two Shiv Senas and two Nationalist Congress

Parties (NCPs). There is heated debate about whether citizens would prefer the original versions of these parties under Uddhav Thackery and Sharad Pawar, respectively, or the newer splinters that have been awarded the original party symbols.

Even though each seat counts as one and it matters not from which state it comes, in the 17 states with more than 10 seats each (that together make up 502), six states count as swing states. These are Maharashtra, Karnataka, West Bengal, Punjab, Chhattisgarh and Haryana. You could add Bihar to this list for 2024. These battle-ground states form a sort of cummerbund around India’s heartland. These aren’t ‘swing states’ in the American sense, as the whole state isn’t awarded to a sole winner, but because the FPTP winner in each constituency will likely be determined by a small margin and will be deeply influenced by the number of candidates, as well as local factors. There are a total of 152 seats in these six states (192 if you add Bihar). By my estimate, about half of these will be ‘too close to call’ and will thus determine whether the recent trend of non-coalition governments is maintained or reversed.

P.S. “With the slightest push—in just the right place—it can be tipped,” said Malcolm Gladwell, author of ‘The Tipping Point.’