Poll insights in unlikely places

When political journalists travel to different parts of the country during elections, we ask people some obvious and occasionally boring questions. “Who’s winning from this constituency?” And, “Why will you vote for so and so party?” In many areas, people are charmingly candid and forthcoming in disclosing who they think will win and, also, why. Passing by Rajiv Gandhi’s friend, Satish Sharma’s parliamentary constituency (Sultanpur in Uttar Pradesh) some years ago, I stopped at a tea shop. The boy serving us seemed up to our oft-repeated questions, and answered with, “Sharma will be walloped in the polls, as no one likes him.” I promptly called up a confidant of Sharma’s and told him bad news was awaiting his leader. I don’t know whether he believed me or told Sharma what lay ahead, but the results seemed to vindicate that talkative tea boy: Sharma was roundly defeated.

Sometimes it is possible to tell the outcomes of polls just by talking to ordinary people. Around the country one encounters, often, so much clarity about why people vote. I met a young bureaucrat during one of my travels who had put together an interesting theory on how to predict polls. His thesis was simple: if there is a drought in a state, then the ruling party will lose. To buttress his point, he had collected data on elections and droughts, and he had found a very high correlation between them. More authoritative studies on elections have also discovered strong connections between the two.

The desire for clarity drives pollsters and academicians to attempt to ascertain the motivations that underlie voting behaviour. Like Bollywood, Indian democracy holds great fascination for western academia. A number of social scientists belonging to eminent universities are trying to crack this interesting riddle in a polity divided on the basis of caste, region and communities. A Google search will reveal many scholarly articles on voting patterns in India and the rest of the world. In 2004, economist Arvind Virmani wrote a paper entitled, “Economic Growth, Governance and Voting Behaviour”. The writer prefaced his enquiry by admitting that voting behaviour is an “extremely complex phenomenon and depends on a host of economic and social factors (e.g. caste, feudalism, coercion/fear), as well as political alliances”. He noted that economics plays an important role, but is not the sole determinant. His conclusions were significant, and his framework may help anticipate the way people will vote in upcoming parliament and state elections. Virmani’s assertion is that high growth rate helps in re-election. He quotes some political economy studies conducted in the United States, which claim that the last six months before elections do not really influence voting. In other words, voters look more at the overall experience of a term than what may have happened in its last year. 

With that in mind, how will India vote? Are we better- or worse-off than we were in 2003–04? Look at the figures: in 2003, the sensex was 3001; since then, the sensex has risen dizzyingly to 20,000. India has now become an almost trillion-dollar economy. As Finance Minister P Chidambaram has said, the average growth rate for the entire United Progressive Alliance (UPA) rule is around 7.5 per cent. Any government handout will show that India is richer, that Indians are eating and living better. Although the growth rate has slipped to about five per cent—for the first time since 2003—the economy is now far bigger than it was 10 years ago. The government has also transferred considerably more wealth to rural areas and the poor than ever before.

Despite these obvious achievements, the UPA government seems uncertain about re-election. Runaway inflation and allegations of corruption against top leaders have destroyed workers’ morale and credibility of the top leadership. The moneybags and the middle class are livid at how this government has dumped them for the poor. This next year will test Virmani’s hypothesis: does high growth rate really win elections, or does a vote depend on more?