Is Karnataka Assembly Election verdict a foregone conclusion?

Elections for Karnataka Vidhan Sabha are scheduled to be on 5th May 2013. Unlike other recent assembly elections, there is not much speculation about the expected results. Not only the main stream media but even social media political gurus are of the opinion that nothing can stop the BJP whitewash and Congress would form the next government with comfortable majority. The major reason for this consensus is the election result of the Urban Local Bodies (ULBs) in Karnataka held in March 2013.

In these ULB polls, Congress emerged as the number one party by winning 1960 wards out of 4952 declared results. BJP and JD (S) were tied for second position with 905 wards each. After the results were declared, political pundits were quick to predict the impending doom of BJP not only in the upcoming assembly elections but also in the parliamentary elections (expected to be held in May 2014). Their rationale is that results of ULB polls is a good leading indicator of an assembly or a parliamentary election. Intuitively, this assumption sounds logical. After all, the same voters who voted in the local body elections would be voting in the assembly and parliamentary elections. According to one estimate, the ULBs covered approximately 30% voters of the state. So theoretically, this argument is very strong that verdict of assembly and parliamentary elections in the state is a foregone conclusion.

However, when one looks at data, this idea is not supported. Take Uttar Pradesh, Delhi and Mumbai for example. In UP, BJP has always been the number one party in Mayoral elections since 1995. The party however could never excel in any assembly or parliamentary election held in the state since 1999. In the assembly elections held in February last year, BJP was at third position and polled its lowest vote share in the state in the last twenty years (both assembly and parliamentary included). However in the Municipal Elections held just after five months of assembly elections, BJP won 10 out of 12 Mayoral elections in the state. In 2006, Municipal elections were held before assembly elections of 2007 and the party had won 9 out of 12 Mayoral seats. This performance was in no way an indicator of its performance in 2007 assembly elections, when the party was at third position in the state. Even if one looks at urban assembly constituencies, the performance of BJP was nowhere close to what it achieved in Mayoral or Municipal elections. This electoral behaviour is not specific to Uttar Pradesh. In Mumbai, BJP and Shivsena are controlling the Municipal body since last seventeen years but failed to win a single parliamentary seat in 2004 and 2009 Lok Sabha elections. Even in assembly polls, held in the same year as parliamentary elections, NCP and Congress had an upper hand over the saffron alliance. MNS, which ensured that Congress-NCP win all the Lok Sabha seats in Mumbai, could not stop BJP-Shivsena from winning the BMC for fourth consecutive time.

In Delhi too, the story is somewhat similar. BJP won the MCD polls comfortably in 1997 and 2007, but could not win the assembly elections held in 1998 and 2008. In 2009 it lost all the seven parliamentary seats in the state. As opposed to Karnataka, where ULB elections reflect mood of 30% of total electorate of the state, the local body elections in Delhi and Mumbai reflect mood of 100% of electorate. And still these elections proved to be a bad leading indicator of assembly and parliamentary elections.

This does not mean that local bodies can never be indicative of trends in assembly and parliamentary elections. In West Bengal, the loss of Left Front in Municipal and Panchayat elections in 2008 gave a good indication of the future rout in 2009 parliamentary and 2011 assembly elections. In Orissa, before dumping BJP in 2009, BJD fought the local elections in 2008 without allying with BJP. Its good performance in the local polls gave BJD enough confidence to sever its ties with BJP and win the subsequent assembly and parliamentary elections in the state on its own.

So what do we make of Karnataka ULB results? Is Karnataka like Mumbai, Delhi & UP or is it like West Bengal and Orissa? The most logical thing to do here would be to look at Karnataka’s past electoral history. In the 2007 ULB elections too, Congress had emerged as the number one party in the state. It had won 1,606 seats (compared to 1,960 won now), JD (S) was a close second with 1,502 seats (compared to 905 now) and BJP was at third position with 1,180 seats (it is now at second position with 905 seats).

When local body elections were held in 2007, BJP and JD(S) were in coalition government in Karnataka. In 2001 urban local body elections, JD (S) had won only 415 seats and this more than doubling of seats in 2007, gave enough confidence to JD(S) to sever ties with BJP and go for snap polls few months after ULB elections. This was very similar to what BJD in Orissa after an excellent performance in Orissa Local Body polls. The difference was that local body polls in Orissa mirrored the assembly and parliamentary results whereas in Karnataka, the ULB polls had no correlation with assembly and parliamentary election results. BJP which was at third position in ULB polls, not only emerged as a single largest party but was only three seats short of majority (which it achieved in subsequent by elections). In Parliamentary elections held in 2009, the party won 19 out of 28 seats in the state. BJP won more seats from Karnataka in 2009 parliamentary elections than from any other state (including Gujarat and MP) whereas in ULB polls its performance was far worse than even UP, where it is neither the opposition nor ruling party since 2002. As Table 1 below shows, since 2001 BJP has been the largest party in every assembly and parliamentary election despite being at third position in ULB polls twice. In the latest ULB polls, position wise BJP has improved. It is now at second position (along with JDS) unlike last ULB polls when it was at third position.

Karnataka Election Results since 2001

 

ULB 2001

Assembly 2004

Parl 2004

ULB 2007

Assembly 2008

Parl 2009

ULB 2013

Cong

2322

65

67

1606

80

62

1960

BJP

562

79

109

1180

110

140

905

JD(S)

415

58

47

1502

28

22

905

JD(U)

457

5

1

NA

0

0

NA

Others

1178

17

0

719

6

0

1182

Total

4934

224

224

5007

224

224

4952

Table 1: Karnataka Election Results since 2001. For ULB polls, wards have been considered. For parliamentary, assembly wise leads and for assembly elections, seats won. Dark Green colored cells show seats won by single largest party.

If BJP could be the number one party in the assembly and parliamentary elections, despite being a distant third in ULB polls, the 2013 ULB results are not that bad for BJP. If KJP and BJP seats of 2013 ULBs are added, it exceeds the BJP tally of 2007. Also on several seats, neither BJP nor KJP won but the vote split benefited other parties. In the assembly election to be held next month, BJP seats would depend on the extent of damage that KJP has inflicted on the party. The most important factor would be candidate selection by all major players in the state.

However the larger point here is that there is no precedent in near past to suggest that ULB polls give any indication of which way wind is blowing in Karnataka.

Last Two Assembly and Parliamentary Elections

 

Vote Share

Seat Share

 

Assembly
2004

   Parl
2004

Assembly
2008

Parl
2009

Assembly
2004

   Parl
2004

Assembly
2008

Parl
2009

BJP

28.3

34.8

33.9

41.6

79

109

110

140

INC

35.3

36.8

34.8

37.6

65

67

80

62

JDS

20.8

20.4

19

13.6

58

47

28

22

Others

15.6

8

12.3

7.2

22

1

6

0

Table 2: Last Two Assembly and Parliamentary Elections. In 2004, assembly and parliamentary elections were held on same day. For parliamentary, assembly segment leads are shown.

In 2004, Lok Sabha and Vidhan Sabha were held on the same day in Karnataka. BJP was leading on 109 assembly segments in the Lok Sabha elections but could win only 79 seats in Vidhan Sabha. Polling booths were same for both elections. The time difference for pressing the EVM button in the two elections was probably less than a minute. BJP’s vote share was 6.5% less in Vidhan Sabha compared to Lok Sabha election even when elections for both were held on same day. This happened because voters are smart enough to understand the difference between a state and national election even if elections are held on same day. One can safely assume that the same voters would vote differently in a state election held 50 days after ULB polls.

We will have to wait for opinion polls with good representative samples to get a better sense of political situation in the state. Opinion polls too would not give a near accurate picture and have been off the mark in past (mostly in states with three or four cornered contests like Karnataka and UP). In this book review of Indianomix, I have explained problem with polling in India but they are still better than ULB polls in indicating the trends. Or better wait till 8th May.

Advertisements

4 thoughts on “Is Karnataka Assembly Election verdict a foregone conclusion?

  1. very well analysed! does away with a lot of punditry we watch on television.
    My question here is – why do we think this phenomena happens? Does it mean indian voters can make the difference between voting at local/national level? is this the new indian voter that knows which issues matter and how to use his rights?

    • Thanks. This is not a pan India phenomena but state/region specific. For some states, people vote same party/alliance for parliamentary, assembly and local bodies. In others this is not the case. The most important differentiating factor according to me is not the issues but the candidates. This of course needs more research.

  2. Well put argument! A quick glance suggests that there is very low correlation between ULB and Assembly/Parl results though the underlying assumption that the voters in this part of the country have the ability to differentiate between types of election increases my curiosity further. I wish to explore the characteristics of these smart voters escp demo-graphical sense and the other possible factors which could have played role in making/keeping them smart while casting votes!

    • Thanks. There is not enough data for Indian elections to understand this phenomenon. As I said in the above comment, we surely need more research and data to understand this phenomenon and also to identify these “smart” voters.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s