One of the most surprising facts of the 2019 elections was the ability of the BJP to retain its hold over the state of Uttar Pradesh, where an alliance of three regional parties was expected to make it lose its majority. Clearly, the alliance did not work to its potential. The BJP won 62 seats; the BSP and SP 10 and 5 respectively. The Congress, nearly mute spectator of the Battle of UP, lost Amethi while retaining Rae Bareli.
Giant bite of the pie
The first achievement of the BJP was to increase its vote share. In 2014, it won the state with 42.3% of the votes. It increased its vote share to 49.5% in 2019 and, in fact, crossed 50% if one considers that it contested only 78 seats.
The shocking news for the SP and the BSP is that they actually did a little worse than their cumulative vote share of 2014, when they had contested separately.
Strike rates reveal that within the alliance, it was the SP that underperformed. The SP won only 13.7% of the seats it contested, against 26.3% for the BSP.
In the absence of publicly-available survey data results, one can only speculate that the SP bears a somewhat greater responsibility for the defeat than the BSP. It could very well be that fragments of BSP voters did not transfer their votes to SP candidates to the extent that was expected.
Or it could be that the Yadav support for the SP was not as cohesive as the SP thought it would be. According to CSDS data, support for parties from their core group has somewhat declined in recent years.
It could also be a combination of both. An examination of vote shares in seats contested shows that the difference between BSP and SP is not great — 40.8% and 38.4% respectively.
West was best of all
The vote share map of the BJP (top right) reveals that it did the best in Western Uttar Pradesh, as it had done in 2014 and 2017. But in 2019, the BJP increased its vote share the most in the Lower Doab — where the SP has a number of traditional strongholds — and in Central Uttar Pradesh, the subregion that has the largest number of Rajput strongholds. The BJP keeps underperforming (if one can use that word in this context) in Eastern Uttar Pradesh, where caste equations are less favourable to it. The BJP also performed well in a number of Muslim-dominated seats in Rohilkhand.
How caste calculus worked
It may seem strange to speak of caste when a result appears to have been so overdetermined by a single factor. But to focus on Narendra Modi alone leads to ignoring the role of other contributing factors to the victory.
One such factor — which is hard to write about as it is poorly documented — is the role of money and the role of organisation. It is worth remembering that the image of Modi alone is not sufficient for the BJP to win elections. In order to do that, it needs to be backed by a strong organisation, by troops on the ground.
The other contributing factor is the caste calculus. I have argued in a previous piece (‘How UP was conquered’, The Indian Express, print edition of May 25, 2019) that the mahagathbandhan failed for having focused exclusively on a narrow caste calculation. That does not mean, however, that caste plays no role. Quite the contrary.
An examination of the caste composition of candidates and winners shows that clear choices were made by parties, choices that correspond to well thought-out electoral strategies, and public discourse.
In 2019, both the BJP and the Congress distributed nearly half of their tickets to upper-caste candidates (BJP 45%; Congress 43.3%). The BSP had a more balanced distribution of caste groups among its candidates, while the SP bet heavily on OBCs (54%). And it gave half of its OBC tickets (10/19) to Yadavs.
Within caste groups, there are clear variations. The BJP distributed 15 tickets to Brahmins and 13 to Rajputs. It fielded only four Jats, and just one Yadav. Nineteen other tickets were distributed to 10 different OBC castes. Kurmis got 6, jatis like Nishad, Rajbhar, Lodh, Saini, Modh Ghanchi, etc. got only one or two each. The same pattern was seen among the SCs — Jatavs got only four tickets out of 17, while a variety of other groups received a small number of tickets each.
UP: The big picture
The point here is this: First, while the BJP claims to look beyond caste and be an inclusive party, it clearly favours its historical core support base. Second, data confirm that the BJP pursued in 2019 the same strategy that helped it sweep UP in 2017: cater to all the groups that are not affiliated traditionally with either the SP or the BSP. And that means a marginalisation of Yadavs, Jatav Dalits, and Muslims.
In retrospect, the SP did perhaps commit the mistake of relying too much on vote transfers. It sought to maximize the representation of Yadavs, forgetting that this particular alignment has created a strong perception of preferentialism that is clearly rejected by voters.
These are only data snapshots on an election that is far more complex than caste or alliance mechanics. I do not claim that these two factors alone explain the outcome. The truth is that we simply don’t yet have the instruments to assess the impact of the new forms of communication that the BJP has invented, or the effect of policy on electoral choices. But while we speculate on the weight of factors that are hard to measure, we should not forget at the same time to look at more conventional factors of electoral politics that still inform the working of democracy.