Understanding Madhya Pradesh 2018… If We Can!

Two recent comments signalled that the election season in Madhya Pradesh was fully on.

First, Govind Singh, Congress leader of the opposition in the state legislature, fanned the flames of false alarm by accusing the state government of destroying the state’s main milk brand, Sanchi Milk, in order to promote Gujarat-based Amul milk.

Second, Chief Minister Shivraj Singh Chouhan of the Bharatiya Janata Party (BJP), said that while love was welcome in the state, jihad was not.

Whatever the rhetoric, it is clear that both principal parties are gearing up for assembly polls due late in the year.

It will be an opportunity for them to establish a new electoral benchmark, since 2018, to put it mildly, was the craziest provincial election in decades.

In a house of 230, the Congress got 114 seats, two short of a majority, with 40.9 per cent vote share.

The BJP got slightly less seats (109), but slightly more votes (41 per cent). The vote share difference between the two parties was just 0.1 per cent, and the seat difference was just five.

2018 was a comedown for the BJP anyway one looks at it, following its majestic sweeps in the preceding three assembly elections.

But the outcome was not cut-and-dried, since the Congress failed to gain a simple majority. And, as we shall see, just a few votes going this way or that, could have changed the outcome wildly. It was an extremely unsettled mandate.

Perhaps, then, it is unsurprising that the Congress government of Kamal Nath fell in March 2020, when 22 sitting Congress MLAs, led by Jyotiraditya Scindia, defected to the BJP, to pave the way for the return of Shivraj Singh Chouhan as Chief Minister.

So, why were the 2018 assembly elections so confusingly tight?

First, victory margins were agonizingly narrow in a very large number of seats. It was below a thousand votes in 10 seats, of which, the Congress won seven, and the BJP, three. An average seat gets more NOTA votes than that!

In eight seats, the margin was between 1,000-2,000 votes; the BJP won five, and the Congress, three. That’s an aggregate of 18 seats with a lottery outcome, in a situation where the Congress needed just two seats for a simple majority, and the BJP, seven.

And, as a table below shows, the carnage didn’t end there.

A full 71 seats, about a third of the total, were decided by a win margin of less than 5 per cent. Another 65 seats were decided by a win margin of between 5-10 per cent. That’s how close the 2018 elections were.

Second, the BJP lost some vote share at a time when the contest in Madhya Pradesh was becoming increasingly bipolar.

As the chart above shows, the ‘Others’ vote has shrunk quite considerably over the past two decades in Madhya Pradesh. This consistent trend, noted in both assembly and general elections, also added to the overall confusion.

Third, and this is rather interesting, the BJP won 41 seats while losing vote shares heavily. This apparent contradiction was because of the truly phenomenal leads it had built up over the previous three assembly elections. And this, perhaps more than anything else, points to the inherent resilience of the party in the state, which will hold it in good stead in the forthcoming elections.

Four, the verdict was patchy because, in between everything else that was happening, the BJP also made a number of handsome gains!

As the map above shows, the BJP made 30 gains in 2018. It also improved its vote share in 73 seats. As Swarajya will demonstrate later this year, in its special series on the forthcoming assembly elections in Madhya Pradesh, many of the outcomes were demographics-dependent. That is why Congress gains (dark blue) cluster in the west, around Gwalior, and in the south, while BJP gains (dark orange) cluster in the centre and the east.

This remarkable phenomenon is best presented in a dimensionless scatter plot of BJP win/loss margins versus vote swing in all 230 seats. BJP gains are marked in orange.

Readers may note two key points in the chart above: firstly, the large number of seats in which the BJP won or lost by a whisker; and, secondly, the magnitude of the vote swing in most of the seats gained by the BJP.

While the 2018 assembly elections are nearly unique, it is interesting to note that it was mimicked to a small extent last month in Karnataka.

There, too, while the BJP lost in the central and north-western regions because of internal flux, it made handsome gains in other parts of the state.

Both cases are indicative of an underlying fact, that the voting patterns are in the process of coalescing into a new paradigm.

Thus, in conclusion (if one is really possible in the midst of such flux), forecasting the outcome of these elections will be riddled with uncertainties, as they become more bipolar.

But this much is clear: if the BJP can get its act together, and engineer a 2 per cent swing in even a third of the seats, the party could very well sweep the state again.