DUSU, By-Elections & ‘Peak Modi’, But 2019 is Still a Dead Heat

What do the recently concluded DUSU elections say about the upcoming 2019 Lok Sabha elections?

Too caught up to read the story? Listen to it instead:

The University of Delhi (DU) immeasurably helped to create Prime Minister Narendra Modi’s political brand. His sprint towards the stunning 2014 victory had begun on 6 February 2013 at DU’s Shri Ram College of Commerce. The auditorium was packed with students eager to listen to his sermon on the Gujarat model of development. A bevy of television channels were going live across the country, to an audience of hundreds of millions of curious viewers.

The Gujarat Chief Minister was very nervous about this outing. Why? Because Modi had a complex about his lack of facility in English; he feared being mocked by DU’s hip and elite students. Yet he spoke passionately in Hindi for 75 minutes.

“I am an optimist. India must learn from global examples to become a global leader. The 21st Century belongs to us. We just need to rebrand the country.”

His rousing speech was peppered with words that were to become his trademark through that extraordinary campaign: development (20 times); challenges (16); youth (15); growth (12); mitron or friends (180)!

Two thousand students sprung to their feet in a standing ovation that seemed to last forever, resonating across the country. Modi himself was astonished at the applause. Then it came to him in a flash. He could win the following year, riding on the mantra of hope, youth, aspiration, development, more development, and even more development; it was irrelevant to his constituency whether he spoke in Hindi or English.

His political appeal was elemental, not cosmetic. A prime minister was born on that day. In DU.

There are four other reasons why DU, and what happens on that campus, is crucial for Modi and the Bharatiya Janata Party (BJP):

• Its two lakh students are drawn from all over the country, speaking every language, following every custom; it’s the closest surrogate for the “young India” that fanatically supported Modi in 2014 and can catapult him again to power in 2019
• Since quotas were introduced for Other Backward Caste (OBC) students around 2006, DU’s enrolment has skewed in favour of Jats and Gurjars from the Hindi heartland states, which Modi had maxed in 2014, winning nearly 90 percent of all seats. However unrealistic, he desperately needs to retain that base
• The BJP’s and Congress’ numbers in DU are a solid proxy for NDA’s and UPA’s national vote shares among students, for the simple reason that regional parties are not directly in the fray. For example, a Rashtriya Janata Dal (RJD) supporter from Bihar is likely to vote for the Congress, while a Telangana Rashtra Samithi (TRS) voter from Telangana could ballot for the BJP, in DU
• Finally, even when the BJP was polling in single digits at the national level (7.35 percent of the votes in the 1971 Lok Sabha polls), it was winning DU polls (Arun Jaitley was president of the Delhi University Students’ Union, or DUSU, in 1974!). Therefore, DU is the oldest and most robust political garh (bastion) of the BJP. If there is a BJP or Modi wave in the country, it must register with gale force in DU

Delhi University Captures the National Political Mood

Given the above reasoning, it’s irrefutable that vote shares thrown up in DUSU elections should give vital clues about the national political mood. I also concede that psephologically, these polls do not make for a scientific sample; so, it would be foolish to project the NDA’s or UPA’s gains or losses from these numbers. But can you glean some tangible insights into the upcoming 2019 battle? You bet!

Here are the key data pointers from the DUSU polls:

• The BJP (or ABVP) has not gained a cent since Modi’s tidal wave in 2013; its vote share has been rock-steady (stagnant?) at 35 percent, from 2013 through 2018 (although 2017 saw a dip to 27 percent)
• The Congress (or NSUI) has gained a healthy five-and-a-half percentage points, from under 27 percent in 2013 to over 32 percent in 2018
• The Aam Aadmi Party (AAP) plus Left alliance polled nearly 14 percent of the votes; had there been a mahagathbandhan or major Opposition alliance (of the kind that is being planned for the 2019 election), the anti-BJP vote would have gone up to 46 percent (since the BJP is anathema for AAP+Left voters)

Crucial, Stronger, More Empirical Validation!

Now to the next critical question: Does the DUSU data synch with other inputs? Does it violate or validate what other credible sources are throwing up? Because if it does validate, then our inferences would have a stronger ring of truth, right?

Let me recall the significant crunching we had done on by-elections held after the Gujarat polls of December 2017. This by-election data is “circumstantially random”, since no systematic bias gets created when legislators die or resign.

Here, then, are the contours of our “real world sample” of ten parliamentary and 21 Assembly by-elections, spread over 15 states, in which over 1.25 crore people cast their votes for nearly 19 political parties:

For those who delight in numbers, do click through to see the granular worksheet.

There is an almost perfect overlap between the DUSU poll numbers and the by-elections’ real voting data:

• The BJP (or NDA) got 35 percent of the votes in DUSU; they’ve got 36 percent of the actual votes cast (normalised) in the by-elections since the Gujarat polls
• The Congress (or UPA) got 32 percent in DUSU, and 32 percent of the real votes in the by-elections
• And while anti-BJP opposition parties (AAP+Left) got 14 percent in DUSU, they (BSP, SP, plus others) notched up 13.3 percent of the actual votes polled in the by-elections

We can now say, with a huge amount of confidence, that vote shares across the country are hardening at 36 percent for NDA, 32 percent for UPA, 14 percent for potentially anti-BJP regional parties, and 18 percent for the rest.

Wait, What About “Peak Modi”?

But wait. Whenever I have done the above analysis, I have been energetically accosted by pro-Modi experts for ignoring a crucial variable (or imponderable): and that’s Modi himself. They argue, quite justifiably, that all this data was gathered when Modi himself was not on the ballot. They assert, again justifiably, that he has a “delta” in which people come out and vote for only Modi, not for the BJP or RSS or any other factor.

I readily concede this argument. There is a “peak Modi” phenomenon, that is, if Modi is in the fray, they would vote for him, irrespective of whether they had voted for non-BJP outfits in other elections.

Now, is there a way we can estimate “peak Modi”? Yes, there is:

• Before Modi, the highest that the BJP had hit was 25.59 percent in the 1998 General Election; so, let’s call this the “core BJP vote”; to make a generous pro-Modi assumption (yes trolls, we do this too!), let’s further assert that there was no growth in the “core BJP vote” until Modi came along
• In 2014, Modi took BJP to 31.34 percent of the national vote; therefore, Modi, solely by himself, added 5.75 percent to the BJP’s vote
• Now let’s make another generous pro-Modi assumption (yes trolls, we can do this a second time too!), that is, “peak Modi” has stayed at 5.75 percentage points, despite the fact that his approval ratings have fallen from 75 percent plus in 2014 to about 50-55 percent in 2018; statistically, we should be scaling “peak Modi” down to about 4 percent, but in deference to the trolls, let’s keep it at 5.75 percent

In the ultimate analysis then, we could be looking at the following score:

• “Peak Modi” could log in 41.75 percent (that is, NDA’s currently polled vote of 36% plus “the Modi-only” vote of 5.75 percent)
• “Peak Opposition” (that is, the pre-poll mahagthbandhan of UPA+BSP+SP+TDP+AUDF etc) could log in 45-46 percent (as validated in several polls)

So, who will win? Go figure.