The Indian Voter in 2014 Devesh Kapur Neelanjan Sircar Milan Vaishnav March 25, 2014
The Lok Surveys Why? To understand the social and polibcal reconfigurabons taking place across India today What? MulB year, panel survey of 65,000 households How? Piggy back off quarterly CMIE Consumer Pyramids survey When? Biannually
Survey calendar Q4 2013: Pre elecbon survey Q3 2014: Post elecbon survey Q1 2014: AspiraBons & AnxieBes
Key findings The electoral horse race Five myths about the Indian voter? Regional parbes are surging Good economics good polibcs Voters are fed up with dynasbes Lack of informabon breeds criminality Vote your caste, not cast your vote
The electoral horse race
The bofom line EsBmated 2014 vote share* UPA 23% NDA 31% * Lok Survey conducted between Oct Dec 2013
The bofom line CSDS (7/2013) Lok (4Q/2013) CSDS (1/2014) HT Cvoter (1/2014) Times Cvoter (1/2014) ABP Nielsen (2/2014) UPA 29% 23% 28% 23% 22% 23% NDA 29% 31% 36% 34% 36% 31%
Big Gains for NDA 7% 0% 0% 9% 5% 9% 6% 7% 8% 11% 11% 12% 13% 15% 23% AP BIH CTG GUJ HAR KAR KER MAH MP ORI PUN RAJ TN UP WB *Note: Lok Survey; Maharashtra and Punjab projecbons include coalibon partners
Big Losses for UPA AP BIH CTG GUJ HAR KAR KER MAH MP ORI PUN RAJ TN UP WB 22% 22% 13% 7% 9% 8% 10% 9% 6% 4% 3% 0% 1% 4% 9% *Lok Survey; Maharashtra and Kerala include coalibon partners
Urban rural differences UPA NDA 35% 33% 32% Vote share (%) 30% 25% 20% 15% 10% 24% 28% 22% 23% 27% 24% 5% 0% Rural illiterate Rural literate Urban illiterate Urban literate Source: Lok FoundaBon
Social bases of support Vote share (%) 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% UPA NDA 42% 35% 35% 31% 25% 19% 21% 21% 23% 12% Upper Caste OBC SC ST Muslims Source: Lok FoundaBon
Actual seat share will depend on: Campaign effects Candidates Alliances
1. Are regional par4es rising?
RelaTve Vote Share In India s NaTonal ElecTons Source: ElecBon Commission of India
All India vote share estmates CSDS (7/2013) Lok (4Q/2013) CSDS (1/2014) HT Cvoter (1/2014) Times Cvoter (1/2014) ABP Nielsen (2/2014) UPA 29% 23% 28% 23% 22% 23% NDA 29% 31% 36% 34% 36% 31% UPA + NDA 58% 54% 54% 57% 58% 54%
Reversal of fortune? UPA NDA Vote share (%) 35 30 25 20 15 10 31.5 21.5 23 31 5 0 2009 (actual) 2014 (projected)* Source: Lok FoundaBon
2. Good economics good poli4cs
The economy dominates 25% 21% 18% 14% 8% 7% 5% 3% Economic Growth CorrupBon InflaBon Personal Income Law and Order Access to Govt Benefits Leadership IdenBty Source: Lok FoundaBon
Growth: large state level variaton 50% % of respondents influenced by growth 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% 43% 43% 41% 38% 32% 31% 28% 22% 21% 20% 19% 17% 13% 7% CTG PUN GUJ WB RAJ TN KER KAR AP MP HAR MAH UP BIH Source: Lok FoundaBon
Top electon issue, by state CorrupBon Economic growth InflaBon/price rise Kerala 43% Chhaosgarh 43% Maharashtra 29% Bihar 40% Punjab 43% Haryana 25% Upar Pradesh 33% Gujarat 41% Andhra Pradesh 26% West Bengal 38% Madhya Pradesh 25% Rajasthan 32% Tamil Nadu 31% Karnataka 22% Source: Lok FoundaBon
Are Muslim voters different? 30% Muslims All India 25% 20% 21% 21% 21% 18% 22% 25% 15% 14% 14% 10% 5% 9% 7% 3% 3% 7% 8% 4% 5% 0% Access to Govt Personal Income CorrupBon IdenBty InflaBon Law and Order Leadership Economic Growth Source: Lok FoundaBon
Bharat vs. India? rural urban 30% 27% 25% 23% 23% 20% 15% 16% 12% 19% 19% 17% 10% 5% 9% 5% 2% 4% 8% 6% 6% 4% 0% Access to Govt Personal Income CorrupBon IdenBty InflaBon Law and Order Leadership Economic Growth Source: Lok FoundaBon
3. Are voters fed up with dynas4es?
DynasTcism among MPs 2004 2009 20% DynasBc Bes 29% DynasBc Bes 80% No dynasbc Bes 71% No dynasbc Bes Source: Kanchan Chandra and Anjali Thomas Bohlken
DynasTc preference? % prefer to vote for dynasbc candidate 60 50 40 30 20 10 0 54 46 No Yes Source: Lok FoundaBon
Why voters prefer dynastes Reason Percent Better at doing politics because it is their occupation 45% Likely to succeed because of greater exposure to politics 40% Makes it easier to deliver services 15% Source: Lok FoundaBon
4. Lack of informa4on breeds criminality
Lawmakers and law breakers? % of Lok Sabha Members of Parliament 35% 30% 25% 20% 15% 10% 5% 0% Serious charge Minor charge 30% 24% 19% 15% 2004 2009 Source: Vaishnav (2014)
Uncovering support for criminal candidates % willing to vote for candidate with serious charges 30% 25% 20% 15% 10% 5% 0% 26% Direct QuesBon Source: Lok FoundaBon
HOW MANY of the following candidates trouble you? Control Group Experimental Group 1. Wealthy Candidate 2. Poor Candidate 3. Independent Candidate 1. Wealthy Candidate 2. Poor Candidate 3. Independent Candidate
HOW MANY of the following candidates trouble you? Control Group Experimental Group 1. Wealthy Candidate 2. Poor Candidate 3. Independent Candidate 1. Wealthy Candidate 2. Poor Candidate 3. Independent Candidate 4. Candidate with serious criminal charges
Uncovering support for criminal candidates % willing to vote for candidate with serious charges 60% 50% 40% 30% 20% 10% 0% 48% List Experiment 26% Direct QuesBon Source: Lok FoundaBon
Support vs. prevalence Support for candidates with serious charges 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% TN AP RAJ CTG KAR PUN Source: Lok FoundaBon; ADR KER MP MAH HAR GUJ UP 0% 2% 4% 6% 8% 10% 12% 14% 16% WB ORI 2009 Candidates with serious criminal charges BIH
5. Vote your caste, not cast your vote
IdenTty ranks 8/8 25% 21% 18% 14% 8% 7% 5% 3% Economic Growth CorrupBon InflaBon Personal Income Law and Order Access to Govt Benefits Leadership IdenBty Source: Lok FoundaBon
Caste and vote choice % of respondents demonstrabng bias 50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% 46% Co ethnic affinity 36% Ethnic bias Source: Lok FoundaBon
HOW MANY statements trouble you? Control Group Experimental Group A corrupt candidate wins An uneducated candidate wins An educated candidate wins A corrupt candidate wins An uneducated candidate wins An educated candidate wins An [Upper Caste/OBC/SC] candidate wins
Uncovering caste discriminaton % of respondents demonstrabng bias 60% 50% 40% 30% 20% 10% 0% 36% 57% Ethnic bias (direct) Ethnic bias (list) Source: Lok FoundaBon
% of respondents vobng for NDA IdenTty predicts vote choice, not electoral outcomes 45% 42% 40% 35% 35% 30% 26% 25% 25% 25% 23% 23% 20% 17% 15% 10% 9% 5% 0% 2009 2014 2009 2014 2009 2014 2009 2014 2009 12% 2014 Source: Lok FoundaBon Upper Caste OBC SC ST Muslim
Conclusion Indian society is changing and, with it, the preferences of Indian voters While the majority of India remains rural, rural India is increasingly shaped by urban sensibilibes Classic rural urban divide is geong blurred Social bias conbnues to be endemic in India However, these biases affect the levels of polibcal support based on social idenbbes, not changes
Conclusion (cont d) The changes in vobng behavior are being driven mainly by economic factors and governance This elecbon is clearly a vote for change. PotenBally it is a pivotal elecbon that could fundamentally affect India's trajectory But we will not really not know that unbl mid May!
Thank you Web: hpp://indiaintransibon.com hpp://carnegieendowment.org Twiper: @MilanV @NeelanjanSircar Print: Sunday Times of India Indian Voter: Inside and Out