As we approach elections for the biggest democracy in the world, a flurry of opinion polls has flooded the market. With allies such as the Rashtriya Lok Dal (RLD) and the Trinamool Congress leaving the INDIA bloc and the Telugu Desam Party's (TDP) Chandrababu Naidu and the Janata Dal (United)'s (JD-U) Nitish Kumar making a gharwapsito the National Democratic Alliance (NDA), the Bharatiya Janata Party (BJP)-led bloc seems to be a clear favourite in these polls. The opposition, meanwhile, is in deep disarray.
Prime Minister Narendra Modi, confident of the NDA's victory, has set a target of 400-plus seats for it in the upcoming polls. Various surveys have predicted that the NDA can win anywhere between 335 to 398 seats, while the INDIA bloc may get 93-166.
Opinion polls do get the direction right in most cases. However, their seat tally is known to be off the mark in several elections.
So, is elections 2024 a done deal? There are three factors that suggest otherwise.
1. Late Deciders
In India, 25% to 30% of voters decide whom to vote for on the very day of voting or just a couple of days before polls, as per the National Elections Studies reports published by the Centre for the Study of Developing Societies (CSDS) over the years. This means there are a lot of voters who are indecisive till the very end, and they can swing the elections in any direction.
Only 36% voters know which party to back even before the campaign starts; 12% do so after candidates are announced, while 18% make up their minds during campaigns.
As per the Axis My India 2019 exit poll, this number is even higher at 43%. Survey data underlines the significance of multi-phase, month-long poll campaigns. Some 43 per cent of respondents said they made their final decision either on the day of polling (14%) or a few days ago/after community meetings (29%).
Odisha (68%), Maharashtra (56%), Jharkhand (63%), Chhattisgarh (55%), Telangana (57%), Bihar (55%) and West Bengal (50%) are the top seven states with late deciders (figures in brackets are the percentages of such voters). All of them, except Chhattisgarh, have triangular or multi-cornered contests. Four states are from the East, and one each is from Central, South and Western zones. In all these states, at least half of the voters decide which party to back very late in the day.
2. A Small Number of Loyal Voters
Only 31% of respondents in the Axis My India 2019 exit poll said they were loyal to only one party and backed it election after election. This indicates only 31% of people in the country are hardcore, ideologically aligned voters.
In the 2009 Lok Sabha elections, the BJP, which got an 18.8% vote share, lost to the Congress, which netted 28.6% votes. This was the worst performance for the BJP since 1991. In contrast, in the 2019 Lok Sabha elections, the Congress got 19.7% votes against the BJP's 37.7%. That, in turn, was the Congress's worst performance ever. The combined bottomed-out or lowest vote share for both parties is 38.5%. Adjusting for the leadership factor, this almost matches the percentage of loyal voters as shown by the Axis poll. That is, it can be assumed that whoever is voting for the Congress today is mostly a loyal voter, just like whoever backed the losing BJP in 2009 in its lost cause could safely be assumed to be its loyal or ideological supporter.
A big (almost 70%) section of voters 'swing', that is, their preferences shift election to election. The remaining 31% are 'loyal' ones, and this nearly matches the share of people - 36% - who decide whom to vote for even before the campaigns.
The states where such loyal voters are high in numbers are Uttarakhand (60%), Gujarat (52%), Rajasthan (48%), Andhra Pradesh (44%), Madhya Pradesh (44%), Uttar Pradesh (38%), Kerala (37%) and Haryana (36%). Figures in brackets denote the percentages of such voters. Five of these states witness largely bipolar contests between the BJP and the Congress, while Kerala and Andhra Pradesh respectively witness a Congress vs Left contest and regional battles.
3. Local Candidate Factor
For 25% of voters - the second-biggest group - the local candidate is the most important factor determining their choice, according to the Axis My India exit poll 2019. For 37%, the prime ministerial candidate was important, 22% focused on the party symbol, 3% on manifesto promises, and 13% on other factors.
Most parties have not declared names for all seats yet, and hence, the 'candidate' factor has not been taken into account by most early opinion polls. To reiterate, almost 12% of voters decide whom to vote for after the candidates are announced. And twenty-five per cent - the share of people who focus on local candidates - is a huge number as well, higher than the lead of 18% the BJP clinched over the Congress in 2019.
Early Opinion Polls Have Gaps
Thus, while many early opinion polls tend to be directionally correct, their seat calculations can go wrong due to these factors.
- In the 2009 elections, the UPA's best estimate was taken to be 257 seats as per an agency, and it got 262 seats. For the Congress, the projection was 144, but it ended up getting 206.
In the 2014 elections, the NDA was expected to win in 275 constituencies, but it ultimately got 336. The BJP got 282 seats against a projection of 226
In the 2019 elections, a projection for NDA pegged its seats at 285, but it got 353 in the end. The BJP got 65 more than the projection of 248 for it
These numbers and patterns should cheer both the supporters and opponents of the BJP. While the former could hope that the NDA will cross the 400-mark given how opinion pollsters have underestimated it in the last two elections, the opposition could find some comfort in that old saying: "game abhi baaki hai".
(Amitabh Tiwari is a political strategist and commentator. In his earlier avatar, he was a corporate and investment banker.)
Disclaimer: These are the personal opinions of the author.