Low polling nationwide has been a notable trend in the first few phases of the ongoing general election 2024 in India. So, would anyone in particular gain from this?
Also read | Supreme Court denies interim order on booth-wise voter data
Voter turnout decreased six times in the Lok Sabha elections held between 1957 and 2019. In 10 elections with increased turnout, the incumbent was victorious six times and lost four; in six elections with decreased turnout, the incumbent won four times and lost twice. This data, when analysed, shows that there is no statistically significant bias for or against the incumbent based on changes in turnout. An examination of comprehensive voter turnout data, however, must unavoidably be multidimensional.
It is conventional wisdom that in Indian elections, higher voter turnout is bad for incumbents. Nonetheless, a wave election was noted in 2014 and 2019. This time around, would the incumbent have any disadvantage if there was no wave like then?
The belief in America
Conventional political wisdom does exist in many countries. For instance, a deeply embedded belief in American political culture is that increased voter turnout systematically benefits Democrats. Many claimed that in the 2016 elections, Donald Trump’s victory may have been reversed had voter turnout been a little higher, resulting in a possible surge in Democrat votes in Pennsylvania, Michigan, and Wisconsin. However, Daron R. Shaw and John R. Petrocik, the writers of the book, The Turnout Myth: Voting Rates and Partisan Outcomes in American National Elections (2020), thought that Hillary Clinton’s support hardly varied depending on the electorate’s size.
How about the Indian myth? In multiple publications in 2018, Milan Vaishnav and Johnathan Guy talked about their analysis of election data from 1980 to 2012 in 18 major States — voter participation “is not necessarily pro- or anti-incumbent; rather, the relationship between these two variables is likely shaped by the specific context at hand”.
Perception about party’s prospects
In 2024, a number of possible explanations for decreased turnout are floating, which include extremely hot weather, the COVID-19 pandemic effect, price rise, job losses, and voter indifference. However, the belief that a particular party would win handily may sometimes be a major deterrent to voting. Bill Clinton was well ahead of Bob Dole in the opinion polls for the 1996 U.S. presidential elections. This led to the lowest polling percentage in 72 years — just 49%. In reality, 9% of likely voters told Harris, a polling organisation, just before the election that “if the opinion polls show that Bill Clinton is certain to win with a big majority”, they would be “very likely not to vote, because there is no real point in doing so”.
But what impact does this have on different political parties? According to Harris’ poll, 10% of Mr. Clinton supporters and 9% of Dole supporters, apparently evenly, said they were reluctant to cast ballots due to Mr. Clinton’s perceived big victory. But the study also revealed that only 14% of Dole supporters and 54% of Clinton supporters thought Mr. Clinton would “win with a big majority”. Thus, it is possible that a greater number of Mr. Clinton’s supporters than Dole supporters abstained from voting. Therefore, in general, the party whose followers hold a stronger belief in a perceived foregone conclusion will be the one more affected by it.
The traditional support base for each party differs across various socioeconomic strata, especially in a diverse society like ours. In the American context, political scientists Michael D. Martinez and Jeff Gill, in a 2005 paper in The Journal of Politics, presented a methodology for evaluating the potential impacts of turnout on an election outcome using simulations based on survey data. By adjusting simulated turnout rates for five U.S. elections between 1960 and 2000, they noticed that Democratic advantages from higher turnout (and Republican advantages from lower turnout) have consistently ebbed since 1960, correlating to the erosion of class cleavages in U.S. elections.
The U.S. elections from 2010 to 2020 were then analysed by Spencer Goidel, Thiago Moreira, and Brenna Amstrong in a 2023 paper published in American Politics Research. Expanding upon Martinez and Gill’s methodology, for instance, to “increase” turnout by 5%, one could “add” the most likely voters to the electorate from those who did not vote until the turnout rate increased by 5%. One may calculate the extent to which this five-point increase in turnout benefited or harmed either party based on the predicted likelihood of these new voters casting Democratic or Republican ballots. Democrat votes would have increased by 1.5% in 2010 with a 15% rise in turnout, according to estimates, but only by 0.4% in 2020.
In India
However, I could not find any comparable research in an Indian setting. Nonetheless, the distribution of habitual voters’ voting patterns ought to differ significantly from those of difficult-to-predict non-habitual voters. Furthermore, the effects of a few percentage votes from non-habitual voters in an Indian election will almost certainly be far more nuanced than in a two-party system such as the U.S. Additionally, the effects would vary across States. So, until the electronic voting machines are counted, these non-habitual voters rushing to the polling booths put political commentators in the realm of unknown unknowns. Without these non-habitual voters, however, everything is much easier to comprehend, and the situation is a known unknown.
Atanu Biswas is Professor of Statistics, Indian Statistical Institute, Kolkata
