The "High Turnout" in ongoing Lok Sabha polls - What could explain it?
The ongoing Lok Sabha elections have clearly shown a higher turnout in most states as compared to previous Lok Sabha elections. Is this an one-off phenomenon? A look at the polling percentages in recent assembly elections suggests that there has been a general increase in voter participation over the past few years. What could be the reason?
A GIS map tracking the poll percentage increases across various constituencies in the country (for which data is available) from 2009 to 2014 could possibly show voter trends on the basis of geography/ rural-urban divides. This map was prepared from data released so far by various Chief Election Commissioners of various states for the six phases held till this piece was published.
Figure 1 shows the voting turnout (in percentage) for each of the Lok Sabha elections from 1952 to 2009.
Here is a comparison of voting percentages from 1977 onwards till the third phase of the 2014 Lok Sabha polls (for which we have ECI confirmation of turnout data so far).
Select State to view graphs of Vote Turnout in % from 1977 to 2014
Table 1 shows the polling percentage for states in the current elections (2014%) for seats polled so far (Held_14), previous LS elections (2009%), last held assembly elections (LastAS%) and previous held assembly elections (PrAS%). Table 2, which focuses on differences in poll percentage (2014-2009 LS corresponds to difference in poll% between 2014 and 2009, LSAsblyDIf - difference between 2014 and last held assembly elections, and ASDif - difference between the previous and last held assembly elections).
Table 2 clearly shows that for most states, peak voter turnout had already occured in the last held assembly elections (but for Maharashtra). (Data for the large state of Uttar Pradesh is not included in this study because of lack of authenticated information from the Election Commission. Newspaper reports point to about 60% polling as the tentative figure provided by the ECI. This compares well to the turnout in 2009 (47.75%) but is only marginally higher than the polling percentage in the assembly elections in 2012 (59.52%) which is itself substantially higher than the 45.96% mark registered in the 2007 assembly elections).
Strikingly, most assembly polls held in 2013 (the previous year) had shown the highest levels of polling (perhaps even ever as Figure 1 points out). Table 2 also shows that compared to last held assembly elections, the voting %s in the present Lok Sabha elections has been lower except for Jharkhand and Odisha among the bigger states. There is certainly something to turnout increases in the past five years, which has to be taken into account while studying the higher voter turnouts in 2014 as compared to 2009.
We could reasonably deduce that the surge in voting as compared to the previous Lok Sabha elections had much to do not only with the specificity of this election, but the fact that voter enthusiasm is on the rise for the past few years as assembly level data shows. Yet the highest levels of polling were in states which had the most recent assembly elections adds credibility to the idea that this was because of the proximity of the timing of the polls to be close to the Lok Sabha polls.
What explains this voter enthusiasm? Is it a wave for/against incumbency? Is it simply a register of the effort made by the ECI to enhance voter participation? That is, has the surge been simply because of correction in voter rolls by the ECI? We shall try to answer the last question first.
ECI Roll Updates & Corrections:
A good way to find that out is to look at changes in electorate as compared to increases in the population in eligible voters (above 18 years of age) published by the Census of India. Popln01, Popln11 represent eligible voters population (above 18 yrs in millions) in 2001 and 2011 census respectively. Poplninc% represents the increase in this population in the 10 years. Voters04, Voters14 represent the electorate (in millions) in ECI's electoral rolls for 2004 and 2014. Voterinc% represents the increase in the electorate between 2004 and 2014. "Cleanup?" represents whether the increase in eligible voters surpasses that of the electorate.
As Table 3 shows, the relative increase of electorate levels (Voterinc%) as compared to eligible voters' population between 2001 and 2011 (PoplnIc%) in a comparative 10 year bucket is lower as a whole for India and also for a number of larger states. One can reasonably assume that the ECI has effectively cleaned up its electoral rolls for this period and thereby the relative increase in the electorate levels has been lower than that corresponding to actual population increase (captured in the variable Cleanup). This would explain to an extent as to why there has been an increase in polling percentage recently.
The Anti-Incumbency Explanation
It is now well accepted that there is a large anti-incumbency feeling among voters against the United Progressive Alliance. Does that explain the high polling numbers? Or could one say that the high polling numbers as compared to 2009 explain a definite anti-incumbency wave? Or a pro-opposition - in particular for the BJP- consolidation?
While high polling numbers in previous Lok Sabha elections have definitely carried a political message (60+% voting in 1984 signifying a wave in favour of the Congress; 60+% voting in 1998 signifying a decisive support to an opposition coalition led by the NDA), the high polling numbers in 2014 have followed similar numbers in the assembly elections. In the assembly elections, no significant pattern has emerged in terms of explaining anti-incumbency or pro-incumbency against state governments, corresponding to high polling percentages. But what is clear is that in assembly elections held close to 2014, especially in the previous year, the Congress party has done relatively worse in all elections which has had high polling. This could signify an anti-incumbency effect against the national party in such states (Madhya Pradesh, Rajasthan, Delhi, Chhattisgarh & Punjab).
One could therefore surmise that there is indeed an anti-incumbency trend that is evident in the high polling numbers across various states, particularly in parts of northern and western India. It is difficult though to attribute the high polling numbers to favour a particular opposition party though, across states. It can be surmised that which-ever opposition party (at the national level) is in a direct fight against the Congress, it stands to benefit in the anti-incumbency trend in states with continuing high polling numbers as seen in respective assembly elections.
This limited study will not detail the trends shown in Figure 2, which could be elaborated upon when full data for all constituencies are made available post the elections.
 Thanks are due to Anil Padmanabhan of The Mint for sharing official data received from the ECI for turnouts in Uttar Pradesh.Constituencies marked with green have shown increases in turnout, while those marked in light red have shown decreased turnouts.
EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.