Bihar Election Forecast: BJP To Emerge As Single-Largest Party Both In Terms Of Seats And Vote Share

Every election has a parallax view, unique to the observer’s computations. That’s what generates a prediction spectrum of divergent outcomes. In Bihar, it looks like many are standing at the same spot this time. This is since, the general consensus is that Chief Minister Nitish Kumar, and the National Democratic Alliance (NDA), will convincingly win the forthcoming assembly elections in the Buddha’s state.

In this article, we shall therefore assess the plausibility of this view, and devise an election forecast for the next legislative assembly in Bihar. It has three sections: the numbers, the factors, and the forecast.

The Numbers

Bihari political alliances are always in a perpetual state of Brownian motion. As the cliche goes, change is the only constant. No one party has the heft to swing the state on its own, and in 2020, these are the three broad vote groupings:

• The mahagathbandhan (MGB): an alliance of Lalu Prasad Yadav’s Rashtriya Janata Dal (RJD, led by his sons, since Lalu is in jail), Congress and the communists.
• The NDA comprises of Nitish Kumar’s Janata Dal (United), the JD(U), the Bharatiya Janata Party (BJP) under Sushil Modi, and two caste parties – former chief minister Jitan Ram Manjhi’s Hindustani Awam Morcha (Secular) HAM-S, and the Nishads’ Vikassheel Insaan Party (VIP).
• The others constitute a significant chunk of the electoral pie, and include diverse caste groups like the Paswans’ Dalit Lok Janshakti Party (LJP), the Koeri Rashtriya Lok Samta Party (RLSP) of Upendra Kushwaha, Asaduddin Owaisi’s Muslim All India Majlis-e-Ittehadul Muslimeen (AIMIM), local rebels, and popular independents.

This is the current house position by alliance, as per the 2015 assembly returns:

This can be a little confusing, since the JD(U) was with the MGB in 2015, but is being treated in the above table as part of the NDA.

The BJP got its wings badly clipped on its own in 2015, winning only 53 seats with a 24 per cent vote share. They re-allied with the JD(U) only when Yadav coalition demands of the other sort grew too wearisome for Nitish Kumar to suffer, and he junked the MGB for the NDA once more.

Also, the skewing between votes and seats for ‘Others’ may seem inordinately high, until we recognise that this includes the LJP, which left the NDA this month, and is now on its own.

To complicate matters further, Chirag Paswan, who is the new LJP party chief following the recent demise of his father, the legendary Ram Vilas Paswan, says that he will not contest against the BJP in Bihar, but only against the others.

All of these contradictions, and more, will be at work in the forthcoming election, in varying degrees of influence.

The Factors

The JD(U): Chief Minister Nitish Kumar has flip-flopped so many times in the past decade that he makes a merry-go-round look linear. If he continues, even after that, to be the most popular politician in Bihar, then it says more for the respect he commands from his electorate, than any journalist may define in words. His acceptability is widespread, and he is the main draw, cutting across alliances. But, his core vote base is restricted to about 20 per cent, which means that he can’t win the state on his own. In 2020, it is projected that the JD(U) will poll under 20 per cent, and will have a lower strike rate than the BJP.

Congress: Any way you look at it, India’s oldest political party looks set to get squeezed once more. Like it has now become in Kerala, the Congress will be reduced to a Muslim party in Bihar as well, and remain consigned to political purgatory. This is in spite of them being part of a numerically formidable MGB alliance. The probability of the Congress crossing into double digits, in either vote share or seats, is woefully low.

The MGB knows this, which is why it is desperate to hold on to Bihar’s primary caste combo, and capitalise one last time upon what remains of an old Mandal-era formula: Lalu Prasad Yadav’s disgusting ‘M-Y’ axis, which successfully sought to secure a block vote of Muslims plus Yadavs (approximately 30 per cent), via artificially concocted identities based on cultural separatism. Electorally, and ideologically, its underpinnings are similar in varying degrees, to a tenacious secular narrative stemming from the tonier parts of Kolkata, which tries to portray the bhadralok’s bhadrakali as somehow unrelated to an ancient pantheon that includes a certain king of Ayodhya; or hoary Periyarist atheism in devout, Tamil-speaking lands, which endeavours to position Shiva’s son as somehow un-Vedic, or pre-Vedic.

The RJD: The M-Y axis is still broadly intact, to the extent that it consistently polls well over 30 per cent in the seats it contests, irrespective of alliances or opponents. But in a first-past-the-post system, it is the aggregate tally which counts, so the best the RJD can hope for in this election, is to profit from those pockets where the LJP hurts the JD(U). They stand little chance in a straight fight against a JD(U)-backed BJP, where the LJP stays away, and this truth will show itself in the RJD’s lower strike rate against the BJP, vis-a-vis other opponents.

The LJP: There are three views on Chirag Paswan’s bold gamble to go it alone, with hardly 8 per cent of the vote share. One, the LJP has delusions of grandeur; if so, then the party will wither away.

Two, it is playing a double game in cahoots with the BJP, with both looking to stiff Nitish Kumar in a reverse-Maharashtra, after the elections. This is far-fetched hokum which strains credulity, but even if so, then Nitish Kumar will make Chirag Paswan’s life as miserable as he can in the years to come.

Or, three, the LJP genuinely wants to occupy some of the space which will be vacated by the RJD and the Congress, with a Muslim-Dalit permutation of its own (particularly in those seats where the JD(U) will be dependent on BJP votes); if so, then the LJP will eventually come into conflict with the BJP in an existential manner, because of the latter’s attitude towards loathsome identity politics.

Either way, the party’s days are now numbered, and it is difficult to see it lasting more than another election cycle or two, unless it goes back to the NDA. One reason is due to state demographics: Dalits and Muslims are rarely found together in dominant numbers; it is usually either-or, in Bihar.

More fundamentally, the age of the Paswans is slowly passing. Instead, it is an era of a Guru Prakash which beckons, wholly concurrent with the inexorable growth of a truth, that there can be no dharmic society without Dalit centrality within. This abrupt assertion of independence by the LJP on election eve, then, is perhaps more a reflection of its rising irrelevance, than anything else.

The Communists in Bihar are primarily of the Maoist variety. Both the Beijing and Moscow factions bow to Naxalbari, since the Communist Party of India (CPI) and the Communist Party of India-Marxist (CPI-M) are junior partners to the anarchist CPI(ML)(L). They routinely poll around 3-4 per cent, and win a handful of seats in traditional pockets of support. This time, they are in formal alliance with the RJD and the Congress. However, the risks of romancing extremism were underscored recently, when two wanted Left candidates were arrested by the police while filing their nominations..

Nonetheless, media machines will pit the Left against the rest, in shades of anti-national versus national, and secularist versus fascist. Nitish Kumar and the Yadavs will stay out of this, while the Congress will provide supporting fire to the Left. Metaphorically, it will be Kanhaiya Kumar versus the Bihar Regiment, and the break-India ‘Tukde-Tukde’ gang from Jawaharlal Nehru University versus ‘Galwan ke balwan’.

Whether this facet has an impact on outcomes or not, it will offer a good deal of drama, and the fully-democratic absurdity, that voters actually get to choose between the nation, and those who wish to break it. In Bihar, the Left may be contesting on 30 seats under an MGB banner this time, but they will be lucky to win in five, at best.

The Sushant Singh Rajput tragedy. Sadly, the news is now entertainment, so, for all the raging prime-time debates we have had to endure, the politics surrounding the tragic death of a son of Bihar will not be a primary election issue.

The Wuhan virus. As a race, Indians are surprisingly resilient when it comes to overcoming natural calamities; families know how to soldier on, even after taking a beating. This epidemic is no different. The livelihood of a very large number of migrant workers was affected, when they were instigated by forces in Maharashtra and Delhi into moving back home, to Bihar, during the lockdown. But the situation has passed, those affected know that the NDA government in Bihar had nothing to do with these machinations, and people are returning to work. Consequently, while Leftist online rags may make much of this matter, it too, will not be an election issue.

On the contrary, the Bihar NDA government’s commendable felicity in containing the virus will be touted as an indicator of administrative efficiency, and will aid in countering any anti-incumbency that might have set in.

The BJP: Come counting day, the BJP will be the largest party in Bihar, both in terms of vote share and seats – without the Yadav vote. Then will come the next step – of breaking the Muslim-Yadav axis (similar to how the BJP benefited in Uttar Pradesh from 2014 onwards, once the Jat-Muslim axis there was broken). This is the sole, remaining stumbling block between the BJP, and a government of its own in Bihar. In the interim, Narendra Modi is set to commence his campaign in a few days; it will only add more votes to the NDA’s kitty.

The Forecast

Election data from the past 15 years was used to generate a pair of prediction curves for the MGB and the NDA. There is simply too much flux and disparity within the ‘Others’ segment to generate a stand-along type curve, even if it attracts more than 20 per cent of the popular vote; consequently, the ‘Others’ are numerically treated as the balance outcome of what is primarily an MGB-NDA clash, and not as a three-way split.

This actually makes electoral sense too, because if we look at elections in the past decade, the “Others’ have consistently polled around 20 per cent, with little to show by way of seats, upsets, or the derailment of either major alliance.

Similarly, the Muslim factor is visibly more pronounced only in Muslim-majority areas. In the rest of the state, it is seen that the Muslim vote becomes largely irrelevant in the face of a consolidated NDA grouping.

In simple terms, the ‘Others’ act as ‘vote-cutter’ parties in far fewer seats that we would expect. Even in the case of the LJP, which is now in the ‘Others’ column, and on its own, it will not do as much damage to the JD(U) as it wishes, since its own vote clout is outstripped by the support the JD(U) gets from the BJP.

Regarding patterns, it is clear that Bihar votes differently in general and assembly elections. However, the 2014 Lok Sabha polls offer an interesting insight into the JD(U)’s actual electoral prowess, which we cannot overlook, since it is the one election when the party went it alone (with disastrous consequences). So too, with the 2010 assembly polls, when the Congress contested on its own, or 2015, when the BJP was in alliance only with the LJP, or 2019, when the Left contested alone.

For all the confusion and coalition-shifting, though, these alliance vote shares demonstrate a surprising linearity of correlation when plotted against seats won, and offer forecasts with a +/- 10 seat error margin. The advantage of such a trend is that it permits us to simultaneously factor in the LJP vote which stays with the BJP in half of the 243 seats, and that which may or may not gravitate away from the JD(U) in the rest.

One unknown factor as on date, which cannot be suitably captured by these curves, is if the LJP enters into any sort of deal with the MGB (tacit or open) in non-BJP seats. One option in that case is to discount the LJP vote share by half, to simulate the absence of LJP candidates in BJP seats. Another option is to assume that Chirag Paswan will stick to his word, and not upset the NDA apple cart any further, but this is Bihar, and you never know.

This issue notwithstanding, the NDA curve was tested separately against the BJP for control. Close conformance between synthetics and actuals was seen for the crucial 2015 data point, thereby raising confidence levels.

In addition, the prediction curves were also tested against data from three opinion polls: the ABP C-Voter poll of 26 September (done before the LJP departure from the NDA became fully apparent), the Times Now C-Voter poll of 12 October, and one by Lokniti-CSDS for India Today of 20 October.

As the table below shows, all three polls appear to be overestimating the vote-to-seat conversion of ‘Others’; in addition, CSDS in particular appears to be overestimating the ‘Others’ vote share and impact, while underestimating the NDA vote share:

The inference is that the ‘Others’ will, in reality, conform to an extremely low vote to seat conversion ratio, like what they normally exhibited earlier; meaning, that they will get a disproportionately lower number of seats, than their vote share would otherwise project. Or, to close the circle, it means once again that their vote share is overestimated.

This anomaly gets even more accentuated, when we plug the surveys’ vote share numbers into our matrix:

Now, you can’t have a tally more than 243 seats, or a scenario in which ‘Others’ get 118 seats. What these absurdities obviously mean, is that the ‘Others’ vote-to-seat conversion is grossly overestimated, and that the NDA’s vote-to-seat conversion is underestimated; exactly what the raw poll inferences suggest. The conclusion is that the actual impact of the ‘Others’ will be less than the surveys predict, in two ways: one, that they will get a lower vote share, and two, that they will get less seats.

This also means four more points: One, that the MGB needs fewer votes to get the same number of seats as the NDA does. This is the power of the M-Y axis. Two, our prediction curves could be a little flatter. Three, Nitish will pull the aggregate NDA tally down because of his JD(U)’s lower strike rate. And four, that Nitish Kumar needs the BJP now more than ever. He may not like, it but that’s the way things are in Bihar.

Therefore, the net outcome of the surveys and our prediction curves is that the NDA will get more seats for each vote, and that the ‘Others’ will get proportionately less, with the MGB filling the gap in-between with their potent caste calculus. This gives us a final, optimal table:

If anything, a few MGB seats may shift to the ‘Others’, but the NDA will still secure a comfortable majority in Bihar, with the BJP emerging as the single largest party.

All election data sourced from the Election Commission of India website.