Opinion
Karnataka’s electoral contest is closer than seat shares suggest
Summary
Which party achieves power is mostly down to the wire as voting patterns reveal of a state with its very own poll dynamics
As Karnataka goes to vote, it may be worth engaging with its complex electoral data. Few states in India witness such intriguing electoral dynamics. We pose and explain two puzzling observations. First, parties face diverging outcomes even in closely held Lok Sabha and Vidhan Sabha elections. That is to say, voters in Karnataka are sensitive to whether they are voting for general or assembly elections. For instance, in the 2004 Lok Sabha polls, the Bharatiya Janata Party (BJP) got about 64% of the state’s seats, but in the Vidhan Sabha elections held the same year, merely 35% seats. In the 2008 Vidhan Sabha, it got 48% seats, but in the Lok Sabha the year after, some 68% of Karnataka’s 28 seats. The next cycle was even more intensely divided. In the 2013 Vidhan Sabha, the BJP secured only 18% of seats, which went up to 61% in the 2014 Lok Sabha (while the Congress party’s seat share went down from 54% in the 2013 Vidhan Sabha to 32% in the 2014 Lok Sabha). And most recently, again, the BJP’s 46% seats in the 2018 Vidhan Sabha went to 89% of the state’s seats in the 2019 Lok Sabha (as the Congress’s fell from 35% to 4%). At one level, this indicates that the BJP is the preferred party in Karnataka for the Lok Sabha, while not as much during Vidhan Sabha elections. This is not quite true, as we discuss later.
Second, Karnataka routinely witnesses an otherwise unusual pattern where the party with most votes overall often loses by a count of seats. While this is not impossible in a first-past-the-post electoral system, it is not usual either, indicating a leakage in capturing the popular mandate. In the nine assembly elections that have taken place in (modern-day) Karnataka, this has happened four times. Even in the other five elections, the vote share difference is not disproportionately large either (compared to seat share). While this leakage indicates wasted votes, at the same time, it also suggests fierce electoral competition.
Why do typical Karnataka voters aspire for divergent leadership in Delhi and Bengaluru and why do popular-vote winners lose elections? The answers to these two puzzles can be found by closely looking at how the state votes for the Congress party in particular.
Consider the two charts published alongside. In the first one, we show the seat and vote share of the last four Lok Sabha and Vidhan Sabha elections in the state. While the divergence in seats is significant, it disappears when we see vote shares. This indicates that the poll performances of the Congress and BJP in terms of drawing votes has not significantly deviated between closely held central and state elections.
As far as a vote preference is concerned, manifested most accurately in votes cast, it is clear that these two parties have performed similarly in the two sets of elections; the only exception is the 2013 Vidhan Sabha. But this graph also shows something far more interesting, which is that the Congress vote share is not significantly lower than the BJP’s even in Lok Sabha polls. This implies the general perception that voters in this state do not want the Congress to represent them in the Lok Sabha as much as they do in the Vidhan Sabha needs to be revised. The Congress is not really deficient in votes across general and state election cycles.
We tested this across all nine Vidhan Sabha elections. The second chart shows the difference in seats and votes between the Congress and its nearest rival, whether it was the BJP, Janata Party or Janata Dal (Secular). While a negative value shows that the Congress’s figure was lower than the rival’s, the more interesting thing is that the difference in vote share between the Congress and its nearest party is significantly less than the difference in seats. Note that of the six times this party lost in seats (where the blue bar is below the X-axis), in four of those (1978, 2004, 2008, 2018), it had more votes than the seat winner (blue bar below but brown bar above the X-axis). Also note that in all six elections which the Congress lost seat-wise, the vote share difference was never more than 7%.
Somehow, it seems voters cannot not vote for the Congress in Karnataka. It may be some form of ‘Congress Complex’ this state has. The party does lose in the state, but is not clearly rejected.
The past five Lok Sabha elections suggest a similar story. Despite losing badly in all these elections (except in 1999), the Congress had secured more than 32% votes, merely 2-4 percentage points lower than the BJP in 2004, 2009 and 2014 (in 2018, the difference was huge though, 20%).
All of this implies that vote banks are not clustered together. This geographical spread of each party’s stronghold groups is what leads to coalition politics in Karnataka so often. Indeed, Karnataka is one of the most complex states to understand the electoral politics of.
Part of this complexity can be traced to its history. The state of Karnataka was formed in 1956 by amalgamating Mysore state with parts of states of Hyderabad, Bombay, Madras and Coorg, each with a distinct identity and culture. During early decades, the legislative assembly was dominated by the Congress party, but over time the party began to lose its dominance, although never got decimated. The Congress Complex of Karnataka is perhaps also the party’s Achilles heel, something the BJP and JD(S) could identify in regional patterns to work on.
Karnataka not only stands apart from other South Indian states in its broad political dynamics, its electoral politics displays fine patterns that are remarkably peculiar, as this analysis has shown.
Yugank Goyal & Arun Kumar Kaushik are, respectively, teachers at Flame University and OP Jindal Global University, and are co-authors of Who Moved My Vote? Digging through Indian Electoral Data.
