Despite Pro-Incumbency, Karnataka Heading for Hung House! Any Takers??



Just as everybody was increasingly asking why there was a dearth of pre-polls for Karnataka, a string of them were published, almost simultaneously. In this post, we comparatively analyze 4 latest major tracker polls - Cfore, Karvy, Cvoter & VMR.  

Both the media and opinion polls had been trying for months to desperately paint the picture that the Siddaramiah-Congress government in Karnataka was facing massive anti-incumbency, while the Congress vehemently denied it.

Now a clutch of pre-polls vindicated the Congress assertion. A poll of poll as later seen in this analysis, project the Congress vote share above their 2013 performance, indicating that rather than anti-incumbency, the Congress is actually enjoying pro-incumbency. 

Secondly, when asked whether life improved under a Siddaramiah government, more than half the respondents nodded in agreement. If voters confirmed their  life improved during the last 5 years, where is the question of anti incumbency? 

Thirdly the personal stock of Siddaramiah within the state is flying mighty high. 46% of respondents not only rated him their best Chief Minister in recent decades but also by a similar percentage as their best preferred Chief Minister. There is practically little challenge to him either within the party or outside. He stands as the tallest politician in the state, without any challenge.

Everyone agrees that Siddaramiah succeeded in making the contest a Presidential race. If so, the numbers he poll should normally translate to those of the party he leads. If you look at the breakdown for the VMR numbers for the Most Preferred CM candidate, you find collectively Congress candidates commanding nearly 48% of votes, BJP 33% and JDS nearly 20%. This is very close to the Cfore overall party wise vote shares. 

The ELECTORAL EQUATIONS for Karnataka is fairly simple. The Congress crosses simple majority mark when their vote share hovers around 40%. Each 1% increase from there, increases their majority further. If Congress touches 46% as Cfore had predicted, it is clear what we would see is a Tsunami Wave. For the BJP, they need to first restrict the Congress below 40% and touch 37% to cross the simple majority mark. In the hypothetical case the Congress vote share crosses 40%, then the BJP need to proportionately increase their vote share past 37%.

Accordingly if Cfore indeed truly reflects the ground situation, then it is fairly clear that Karnataka is a NO Contest. Under the circumstances, it is very tempting for the BJP to use unethical Push Polls techniques  in an attempt to induce bandwagon, tactical voting & contagion effects. Political opinion polls are not always an accurate reflection of public opinion. They could be also an "Opinion Making" Polls as the JDS spokesperson on the IndiaToday panel summarily dismissed the Karvy Poll. The aim of these pre-polls would to narrow the gap so that both their cadres and voters remain enthuse to make the contest more competitive than it is really is on the ground.

The non-Cfore bunch of pre-polls apparently did not disappoint. They apparently scaled down Congress below 40%, while projecting BJP inching towards the 37% mark and promising shamelessly & openly that after Modi's campaign the BJP vote share would increase further. Though vote wise and seat wise, all these pre-polls look almost similar, a closer analysis indicates wide variations in their internals. For the purpose of analysis, we have limited it to solely vote shares as seen below.

Poll of Poll: Votes Shares by Region


Poll of Poll: Overall Vote Share


The Poll of Poll indicates the following

- Overall State, Hyderabad Karnataka & Bangalore: Congress most likely to lead
- Central Karnataka: BJP most likely to lead
-  Mumbai Karnataka & Coastal Karnataka: Neck to Neck
-  South Karnataka: Fuzzy but Congress likely to get better of JDS

There is a temptation to brand Cfore an outlier, particularly since Congress part funded them as their internal survey. However, not only their projected 46% vote share for Congress finds high correlation with Siddaramiah's personal ratings as confirmed by other pre-polls, the fact as seen below, this number is also in sync with their demonstrated vote share during the last 5 years as seen below:


The BJP faces a lot of constraints. They face a strong resistance at vote share band 33-34%. This is mainly due to their extremely narrow social base, which is made worse recently by erosion in their urban, Dalit & Lingayat segments. Even a 1% loss in each of these segments can bring BJP down to 30%. However, the non-Cfore pre-polls are all projecting BJP above this resistance bands with no explanation why this buoyancy is happening. As seen earlier, the bogey of anti incumbency against the Congress government was busted by their own data! Accentuating their challenges, is the fact that BJP has no presence in approximately 70 out of 224 seats, nearly one third the total assembly seats.

The JDS though depleted organizationally by defections, facing a funding crunch and loss of secular votes is putting up a plucky fight in South Karnataka. Vokkaligas are consolidating around JDS and this should erode some 1-2% vote share of both the BJP & Congress respectively. The JDS has no presence in nearly 140 seats, losing their deposits in 2013 in these seats. In 2014, the JDS was reduced to just 11% vote share. Consequently projections of pre-polls of JDS vote share hovering around 20% overall at the state level, i.e. nearly 100% increase viz-a-viz 2014 looks grossly unrealistic. 

A tacit understanding between BJP and JDS is expected but only on select seats, not across the board. The JDS is fighting basically an existential battle. So it is highly unlikely they would transfer a significant chunk of their vote to BJP overall if they are not assured that they would not gain from it. If they end up 20-30 seats, then Shah would immediately get into his act to wholesale buy up their MLAs. they need to cross 40 seats  to be safe and hence they would aim to increase their vote share as much as possible. Being a zero sum game, either BJP's or JDS' vote shares can get boosted, not both simultaneously as reflected in the non-Cfore pre polls. 

Most of these pre-polls on the face look as push polls. A push poll is an interactive marketing technique, most commonly employed during political campaigning, where the aim is to manipulate or alter prospective voters' views/beliefs under the guise of conducting or disseminating findings of an opinion poll. These are primarily to distribute propaganda rather than to understand the views and opinions of the public.
 
As we get closer to voting day expect more such polls, that project BJP decisively leading due to the "Modi Effect". Just as  the BJP carpet bombs their advertisements in each and every paper, they are surely cash rich to buy off most polling agencies. 

Would such tactics work? Sometimes it does. But not always. We have Vajpayee in 2004 where push polls faced their Waterloo. In the US, Trump overcome literally a battery of push polls to win the US Presidency. Ditto Brexit and attempts to marginalise Jeremy Corbyn. So let's wait for May 15th to find out...

Archives: 

Karvy & Cvoter Polls: Karnataka Stuck in a Time Warp! Nothing Changed Since 2013!! Read Here
Karnataka Polls 2018: Who among pollsters will end up the Winners & Losers? Read Here
KarnatakaElections: Would the spectre of a JDS-BJP Coalition Spook Off Voters??? Read Here


Karvy & Cvoter Polls: Karnataka Stuck in a Time Warp! Nothing Changed Since 2013!!




For many of us living in Karnataka and follow both national and state poltics very closely, the findings of the IndiaToday-Karvy and News9-Cvoter pre-polls ended up flummoxed. Both polls suggested that the vote shares of major parties were caught in a time warp viz remaining frozen since 2013!. Not surprisingly both suggested that it may take Modi's campaign during the the last two weeks to unfreeze the state!

Based on such data, media speculates that the Battle for Karnataka as still open as these extracts from a LIveMint article reflects 



Wow! Just think about it ! BJP vote share stuck at 33%, Congress at 36%! How many takers are there for this hypothesis??

It is as if issues such as State Government accepting separate religious status for Lingayats, the Dalit revolt, escalating crimes against women in the country, Kannada pride & anti-Hindi sentiments, to name a just a couple of issues, had absolutely no impact on voting intentions. If the shoe was on the other feet, i.e. if the BJP was a likely beneficiary of any one of these issues, our slavish media and pollsters would have no doubt bombarded us by now their fallout, analysis after analysis 24x7!

Both the surveys are absolutely mum on the fact that in 2014 BJP registered their highest vote share of 43% in the state. Yet in 2019, their polls show BJP at 33% level - experiencing a huge negative swing of 10%. Why no explanation for such a whopping swing away forthcoming from these pollsters? 

They are also mum on the demonstrated vote share growth of the Congress during the last 5 years due to expansion of their social base due to their AHINDA coalition building strategy. At their projected 36% vote share, this too is a big swing away from what the Congress demonstrated in the last urban-rural civic elections in 2016.

But at least here, there is some feeble explanation is attempted - anti-incumbency! The moot question is where is the anti-incumbency if the projected vote share of the Congress by these respective polls are at their 2013 levels? Their own vote share data contradict their claim of anti-incumbency


Even more interesting is the data they use to back up their anti incumbency claims. Take Cvoter. Look at how the question is formulated as if they apparently using push-poll techniques- "Who are you ANGRY with?". 52% of the respondents Cvoter apparently found were not angry with anyone  at least with the list they gave as options. Maybe if Cvoter had expanded their choice to include spouse, girlfriend, boyfriend, teacher, Arnab Goswami etc they would have got a better instantaneous response rate to this question!

This should also better explain why Cvoter found the BJP unable to capitalize on anti-incumbency - simply because none existed according to their own data. This number of 52% incidentally finds high correlation with CSDS & Cfore surveys pro-incumbency wave.


A more significant contradiction of their vote share projections of the Congress is when they give their caste wise projections. The CSDS Lok Sabha 2014 data provides a good baseline of vote share distribution on the basis of caste wherein the Congress overall registered a vote share of 42% and BJP 43%. But look at the caste wise projections for the Congress by Karvy in 2019 - the party has improved its share across the board viz-a-viz 2014. If this be the case, one would have expected the Congress to increase their vote share from 2014 and move towards the range projected by Cfore & CSDS 46-47%. Instead we find Karvy scaling back the Congress to 36% - their vote share registered in 2013!


Even more interesting is this table of Cvoter published in TheQuint. Rather than provide data on voting intention share distribution by parties, Cvoter opted to provide the findings of a question "Who Will Form The Next Govt in Karnataka?". Coincidentally this is also a much commonly used push poll technique that are designed to induce such effects like "Contagion and Bandwagon".

Contagion and the bandwagon effect are not simple phenomena. There are a range of variables at play during an election. The bandwagon effect comes in play where voters who think a particular political party will win the election may end up voting for that party or candidate. Now if Cfore and CSDS 17% margin leads for the Congress are lapped up by the electorate, even more voters may be induced to vote the Congress due any bandwagon effect in play. So it becomes imperative for the poll strategy of political rivals to counter such a perception by releasing counter polls that narrows down such a wide margin by commissioning what is known in the polling industry as hired gun polls.

There are other two types of other effects push polls are known to be commonly designed to induce viz. tactical voting and contagion. Tactical voting is motivated by the intention of voters to affect which party wins the election. The move of voters to the perceived majority view is called “contagion”.

The difference between tactical voters and those backing a perceived winner shows how voting behavior under a tactical consideration will be more complex than voting behavior under a bandwagon effect. For example, in a three-party election under the first-past-the-post method, Voter A and Voter B think their most-preferred candidate is going to get the fewest votes. Voter A is a tactical voter; Voter B is affected by the bandwagon effect.

Both Voter A and Voter B will tend to vote for their second-preferred candidate if their ballots are perceived to be crucial in their constituency, such as in the case where their second-preferred candidate is expected to win the election by a small margin. If voters believe their ballots are trivial in their constituency, then Voter B will tend to vote for the candidate expected to win. Although Voter A’s most-preferred candidate has no chance of winning, Voter A will still vote for their most-preferred candidate because a vote for their most-preferred candidate or for their second-preferred candidate cannot change the result.

The Karnataka Elections Lends Itself The Perfect Fertile Grounds For Push Polls Utility. 

On one hand,  we have a greatly resurgent Congress with a clear charismatic leader with a pan Karnataka appeal. Almost everyone agrees the Congress have displayed a vice like grip on the narrative and have won the Battle of Perception that they are not only poised to win but win with an increased majority. Anecdotal evidence are also available in the form of crowds they are pulling which are relatively much larger and ecstatic than their other two rivals. Though the BJP have thrown a lot of dirt at them, nothing succeeded as of yet in sticking on to the Congress which is often reflective of a party on the upswing.  

On the other hand, we have a BJP whose main challenges are greatly faction ridden party, discredited leadership, narrow social coalition base which is made narrower through erosion of urban voters, Lingayats & Dalits. To their bad luck instead of an opportunity to capitalize on anti incumbency they are challenged by a pro incumbency wave in favor of the Congress. The JDS find itself organizationally weakened by key defections and paucity of funds. They are also hit by desertion of their secular vote base, primarily the Muslims. On the plus side, what's going for them is consolidation of Vokkaliga votes. 

It is clear to voters that neither the BJP nor JDS on their own can dethrone the Congress. However, both can have a small window of opportunity to share power if the mandate throws up a hung house. JDS could be kingmaker if not king having the luxury to choose between Congress and the BJP. For a BJP, the window opens up full play for Shah to wholesale buy MLAs as he did in the North East & Goa to form a BJP government.

Under the circumstances, it is very tempting for the BJP to use Push Polls to induce bandwagon, tactical voting & contagion effects. Political opinion polls are not always an accurate reflection of public opinion. They could be also an "Opinion Making" Poll as the JDS spokesperson on the IndiaToday panel dismissed the Karvy Poll. 

Don't be surprised if a spate of polls are released on the eve of voting, projecting a "razor sharp contest" or BJP in clear lead. Don't get carried away by media brands like IndiaToday too.. Go by the transparency & robustness of methodology and internal consistency of the data of polls published.



 

Like us on Facebook

Follow us on Twitter