India Assembly elections 2018- Karnataka, MP, Rajasthan, Chhattisgarh, Telangana

[jaichind]

Looks like a few seats are left and they gave up for now ...

                Won   Leading   Total      Vote share
BJP            100          4       104         36.2%
INC+           79                    79          38.4%  (INC 78 Pro-INC Ind. 1)
JD(S)-BSP    38                    38          18.7%  (JD(S) 37 BSP 1)
KPJP             1                       1           0.2%

If you take NDTV's analysis of 2013 results IF BJP KJP and BRSCP were united and take into account that out of the 11 OTHER in 2013 it is 4 INC rebels and 5 BJP rebels you get a net swing of

BJP           +12 seats
INC+         -17 seats
JD(S)-BSP   +4 seats

From a vote share point of view (if we take BJP 2013 as BJP+KJP+BSRCP) it is

BJP           +3.7%
INC+         +1.8%
JD(S)-BSP  -2.4%

Which make this election a mostly status quo election with a small swing toward BJP relative to 2013

[jaichind]

Exit poll summery

Note that only 222 out of 224 seats are voting since in 2 seats a candidate has passed away ergo no election will held.

                                                 BJP   INC   JD(S)   Others
Times Now-Today's Chanakya      120    73     26         3
News Nation                               107    73    38          4
NewsX-CNX                                106    75    37          4
Republic TV-Jan Ki Baat               105    78    37          2
Dighvijay News                           105    78    33          6
ABP News-C Voter                       103    93    25          1
Times Now-VMR                           94     97    28          3
IndiaTV-VMR                                87     97    35          3
India Today-Axis My India             85    111   26          0
Suvarna                                      85    111   26          0

Average w/o Chanakya                 97     90    32          3
Average w Chanakya                   100    88    31          3
 

It seems Republic TV-Jan Ki Baat exit poll mostly nailed it.

[jaichind]

Karnataka 2008 assembly election

              Contest          Won           Vote Share

INC           222               80                34.76%

INC rebel                         3                 1.24%

JD(S)+      220               28               19.05%

JD(S) rebel                      1                 0.33%

BJP           224            110                33.86%

BJP rebel      2                0                  0.99%

BSP          217               0                   2.74%

SP            121               0                   0.93%

SKP           21               0                    0.40%

CPM+        19               0                    0.35%  (CPI was part of CPM+)

JD(U)        72               0                    0.33%

JD(S)-BSP clearly over-performed in terms of seats relative to vote share.  A comparison versus 2008 has both BJP and INC+ over-performing 2008 vote shares and JD(S)-BSP under-performing 2008 JD(S) vote shares but JD(S) gets more seats in 2018 and both BJP and INC+ gets less seats.

[Dr Oz Lost Party!]

Looks like BJP will get at most 104 seats ... not enough to block JD(S)-INC.

                 Won   Leading   Total      Vote share
BJP              97          7       104         36.2%
INC+           76          3        79          38.4%
JD(S)-BSP    38          0        38          18.7%
KPJP             1                       1           0.2%

*Sigh of relief*

[jaichind]

All done.  Note vote share has NOTA in it that will have to filtered out when computing effective vote share.

                Won     Vote share
BJP            104         36.2%
INC+           79         38.4%  (INC 78 Pro-INC Ind. 1)
JD(S)-BSP    38         18.6%  (JD(S) 37 BSP 1)
KPJP             1            0.2%

[jaichind]

Now the Karnataka governor (pro-BJP)  has a decision to make.  Should he invite the largest party (BJP) or should he invite a post-election alliance of INC and JD(S).  if he goes with the former then BJP can use all sorts of financial inducements to get defections from both INC and JD(S) where the MLA will then resign and run for re-election as the BJP candidate. 

This is what took place in 2008 when the BJP won 110 out 224 seats and was invited the form the government.  Of course back in 2008 there was no real post-election deal between INC and JD(S) to form the government as they did not have the numbers either. 

As a precaution JD(S) is already putting their MLA in a resort and most likely limit cell phone usage.  INC I suspect will do the same soon. 

[jaichind]

Pro-BJP Karnataka governor invites BJP as the largest party to form the government.  He gives BJP 15 days to prove its majority which is rather large going by history.  This seems to give BJP a lot of time to poach for defections from INC and JD(S).   INC and JD(S) has escalated this to the Supreme Court and a a three-judge bench of the Supreme Court will hear the INC-JD(S) case. 

Not sure why BJP is doing this.  The goal of BJP forming the government in Karnataka is to create a media momentum of the BJP victory narrative.  By forming a government through defections takes away from the narrative an merely forces the INC and JD(S) base to combine in 2019 against BJP.  JD(S) has some strength in Maharashtra and doing this merely means JD(S) in Maharashtra will also join up with INC-NCP in the upcoming LS elections.

What the BJP should do is to let JD(S) form the government.  As the larger of the two parties various elements of INC will eventually find this problematic and that will hurt INC-JD(S) in 2019 at the grassroots level and the JD(S)-INC government might even fall apart before 2019. 

[jaichind]

In Karnataka Now that the BJP CM Yeddyurappa has been installed he as 2 weeks to prove his majority.  The seat distribution is

BJP               104
Pro-INC Ind.     1
KPJP                2
INC                78
BSP                  1
JD(S)              37

BJP has to get to 112 since one BJP MLA will have to be the Speaker.  It seems 2 INC MLA are already AWOL and most likely have defected to the BJP.  The Pro-INC Ind. seems to be going back and forth.

The rest of the INC MLAs are in a luxury resort.  All their smart phone have a special INC app installed that will monitor all incoming and outgoing phone calls.  I suspect JD(S) has something similar set up.

[jaichind]

Based on what the prices various JD(S) MLA have been quoted from the BJP, the price the BJP seems to be willing to pay for a MLAis around $17 million.  Since BJP needs 8 MLAS (10 MLAs to be certain) it seems that the BJP budget for this government formation phase is around $170 million.  Small peanuts compared to the potential monetary benefits for the key BJP kingpins.

[jaichind]

BJP has destroyed democracy, says Shiv Sena

https://www.theweek.in/news/india/2018/05/17/bjp-has-destroyed-democracy-says-shiv-sena.html

Again. So called BJP ally SHS came out against BJP in the Karnataka crisis.  Of course all this could be one massive ruse to set up a situation where to get SHS to come back into the NDA fold the BJP will have to accept being the junior partner in Maharashtra the next election cycle. Something the BJP will not accept so SHS has to come out with a very negative tone publicly to make sure that the BJP  knows that SHS means business.   

[jaichind]

Prelim Karnataka 2018 assembly election (only 222 out of 224 seats voted)

              Contest          Won           Vote Share

INC+         222                79             38.75%   (SWJI was part of INC+)

INC rebel                          0              1.03%

JD(S)+      217                38             18.79%  (BSP was part of JD(S)

JS(S) rebel                       0               0.18%

BJP          222               104             36.57%

BJP rebel                         0               0.71%

KPJP          30                 1                0.21%

Which is not too far away from my adjusted exit poll based seat projection algorithm

To get                                        BJP    INC  JD(S)   Others
                                                107    86     26         3

In fact given the vote shares my seat projection would be what one would expect.  It seems there must have been some BJP->JD(S) tactical voting to explain this.  Looking at the seats that JD(S) won over INC by narrow margins one does notice that in most of them the BJP vote share is lower than the 2013 BJP+ bloc (BJP KJP BSRCP BJP rebels) in an election when the BJP bloc rose which sorts of validates my theory.

If one takes 2013 election results and group vote share by blocs: INC+ (INC plus rebels), BJP+(BJP, KJP BSRCP BJP rebels) and JD(S)-BSP (JD(S) BSP JD(S) rebels) the 2013 to 2018 vote share shift are:

INC bloc:     +1.36%
BJP bloc:     +3.04%
JD(S) bloc:   -2.72%

So relative to 2013 the BJP bloc gains are fairly minor relative to INC bloc which given Karnataka's record of voting out incumbents since the 1980s is a pretty good performance by INC. It seems what did INC in was two factors a) Dalit vote swing toward JD(S)-BSP in INC-BJP marginals helped BJP and b) BJP tactical voting for JD(S) in several INC-JD(S) marginals.

[jaichind]

Indian Supreme court ruled that the new BJP government must face a floor test this Sat instead of 2 weeks from now.  Also the floor test must be a open vote and not a secret ballot.  All things equal most likely BJP will not be able to rustle up enough INC and JD(S) defectors in time although BJP claims that at least 12 INC and JD(S) MLAs are in touch with them to defect.  Of course the Indian Anti-Defection Law means that any such defection would mean disqualification and these MLAs will have to run for by-election on the BJP ticket.

[jaichind]

Karnataka BJP CM Yeddyurappa resigns ahead of floor test.  BJP clearly does not have the numbers.

[Lachi]

Based on what the prices various JD(S) MLA have been quoted from the BJP, the price the BJP seems to be willing to pay for a MLAis around $17 million.  Since BJP needs 8 MLAS (10 MLAs to be certain) it seems that the BJP budget for this government formation phase is around $170 million.  Small peanuts compared to the potential monetary benefits for the key BJP kingpins.

Ahh, wonderful, trying to win over a majority the traditional Indian way, bribery.

Seriously though, is what the BJP doing even legal?

[jaichind]

Based on what the prices various JD(S) MLA have been quoted from the BJP, the price the BJP seems to be willing to pay for a MLAis around $17 million.  Since BJP needs 8 MLAS (10 MLAs to be certain) it seems that the BJP budget for this government formation phase is around $170 million.  Small peanuts compared to the potential monetary benefits for the key BJP kingpins.

Ahh, wonderful, trying to win over a majority the traditional Indian way, bribery.

Seriously though, is what the BJP doing even legal?

Well, the bribing part would not be legal although BJP will deny that.  The on the surface controversy is a Karnataka Governor who was a BJP MP asking the largest party, BJP, to form a government despite the fact that an post-election alliance of INC and JD(S) have a majority.  While this decision is  problematic it is not illegal.  The governor has discretion to invite who he or she feels would give the state stable government.  The governor also gets state the time frame of a trust vote (15 days) and to appoint a Speaker (on a temp basis) (in this a BJP one) that can decide the type of trust vote (voice or secret ballot.)  Nothing that is done above the table is illegal but clearly biased to help in the bribing of MLAs to support a party that does not have a majority.  In the Supreme Court came in pretty much said the same thing and forced a trust vote in 48 hours without a secret ballot.

[jaichind]

Now that the Karnataka BJP government has fallen, INC and JD(S) can get to the hard bargaining for power.  INC for caste balance reasons are asking for 2 INC DCM, one Dalit and one Lingayat or Muslim. JD(S) only wants to create one while the INC  Lingayat and Muslim factions are battling it out to see if there is a second DCM from which commuity it would come from.  Both INC Lingayat or Muslim factions  are claiming that if their bloc does not get a DCM position then their community will abandon INC in the 2019 LS elections.

In many ways if INC-JD(S) were to form a pre-election alliance in Karnataka in 2019 LS elections it might be had for BJP short run but might be very positive for BJP's growth in Southern Karnataka.   The Vokkaliga vote which are with JD(S) are a landowning Upper Caste group which would normally vote BJP.  But JD(S) has held on to this voting bloc because of the perception that JD(S) is the Vokkaliga party.  If JD(S) were to form an alliance with INC which represents Dalits, Mulims and lower OBCs this could push the Vokkaliga vote to the BJP and lay the basis of BJP being competitive in Southern Karnataka on the long run. 

[jaichind]

JD(S)'s HD Kumaraswamy to be sworn in as Karnataka CM on Wed.  INC JD(S) haggling continuing with INC demanding two INC DCM, one Dalit and one Lingayat (I guess the idea of a Muslim DCM is gone which makes since INC can gamble that Muslims have no choice but to vote for the largest anti-BJP party which in Karnataka is INC.)  INC will get Speaker role which gives the INC the upper hand in any future INC-JD(S) split and INC needs to get defections from JD(S) to form a government on its own.

It also seems that INC and JD(S) has agreed to contest 2019 LS election in Karnataka as allies.  JD(S) also likely to join INC-NCP alliance in Maharashtra where it has some presence.  

[jaichind]

A bunch of by-elections coming up that will test opposition unity against BJP.   Voting will be May 28th and counting will be May 31st.

The most key one is UP Kairana LS seat where the BJP MP passed away and the BJP will run his daughter.   All opposition parties (SP BSP INC AAP) will back RLD who nominated a ex-BSP MP Muslim.

In 2014 LS it was

BJP    50.6%
SP     29.4%
BSP   14.3%   
RLD    3.8% (backed by INC)

My current model based on 2014 results plus some implied swings for this eat for 2019 which also assumed that this seat will go to SP under a SP-BSP-RLD alliance with INC tactically withdrawing has it at

BJP    50.8%
SP      46.4%

This seat has a relatively high concentration of Jats which went to BJP in 2014 due to Jat-Muslim communal rioting.  RLD's strategy of nominating a Muslim candidate to demonstrate to its future allies (SP BSP INC) that it now has regained its hold over Jat votes that the RLD label can get Jat votes to vote RLD despite a Muslim candidate.  The 2017 assembly elections seems to indicate that the RLD Jat is coming back a bit.  If you take the 5 assembly segments and add up the votes it comes out to

BJP        38.2%
SP-INC   21.6%
BSP       18.5%
SP           4.2%
INC         5.5%
SP rebel   2.8%
RLD         7.6%

Which if you allocate SP-INC vote 75/25 to SP and INC you get implied relative strength of
 
BJP        38.2%
BSP       18.5%
SP         23.2%
INC        10.9%
RLD         7.6%

Where it seems some of the BJP Jat vote of 2014 has gone back to RLD.

One should also assume some sympathy factor for the BJP daughter of the deceased BJP MP so this result is up in the air.  If RLD can win then it proves both the ability of the opposition to combine its votes as well as its ability to beat by the BJP's main narrative that the Opposition, especially the INC, are the Muslim party.





Then there is the Maharashtra Palghar LS seat where the BJP MP has passed away.  BJP will field a recent INC ex-MLA defector while SHS has jumped into the race and nominated the son of the deceased BJP MP.  In 2014 it was

BJP    53.7%   backed by SHS
BVA   29.6%   backed by INC-NCP
CPM    7.7%

BVA has since defected to join BJP bloc which should give BJP the edge.  But the SHS jumping in with the son of the deceased BJP MP has made this by-election a difficult battle for BJP.  INC will also run with NCP backing.

My 2019 model for this seat has BVA running for NDA, NCP running with INC support and SHS running separately and has it at a close 3 way battle

BVA    31.7%
NCP    27.7%
SHS    26.8%
CPM      7.1%

If you look at the 2014 assembly segments for Palghar when BJP SHS INC and NCP blocs ran separately you get

BJP    21.2%
SHS   23.1%
INC     8.4%
NCP+ 35.2%  (BVA was allied with NCP and one can argue 65% of this support is for BVA)
CPM    5.9%

But in a jolt for BJP, now BJP ally BVA will also run.  So if you use the assembly election segment results from 2014 and assume 65% of NCP+ vote are for BVA you get very close 4 way race.

BJP     21.2%
SHS    23.1%
BVA    22.9%
INC     20.7%  (backed by NCP)
CPM     5.9%

One can argue that SHS should have the edge given the additional edge of the sympathy factor of running the deceased BJP MP's son.  The family of the deceased BJP MP has completely gone over to SHS and are now denouncing the BJP.  This race will also be fun to watch to see how the result will come out.

[jaichind]

Picture of JD(S)'s HD Kumaraswamy swearing in ceremony in Karnataka that should have BJP somewhat worried.   Ancient rivals like INC and TDP, AITC and CPM, SP and BSP on the same stage at the same time.



Not captured is AITC's Mamata Banerjee and TDP's Naidu which are here


Missing is TRS's KCR who is now isolated.  TRS's position is that in Telangana INC is its current main enemy of today but BJP is the enemy of tomorrow. So TRS's goal is a non-INC non-BJP third front.  With JD(S) lining up with INC this position is a non-starter.  It will now be INC-Third Front alliance to take on Modi in 2019.  Other than BJD which is in a similar situation, TRS will fight a lonely battle against both.

[jaichind]

Latest CSDS National Survey seems to generate results similar to my current model

http://www.lokniti.org/pdf/Lokniti-ABP-News-Mood-of-the-Nation-Survey-Round-3-May-2018.pdf

It has (different from Jan 2018)

BJP           32 (-2)
BJP allies    5 (-1)
INC          25 (--)
INC allies    6 (+1)
BSP            4 (+1)
BSP allies   6 (+2)  (SP JD(S) INLD)
Left            3 (--)
Others      19 (-1)

My current model has
BJP          31.8%
BJP allies   3.9%
INC          21.5%
INC allies   7.5% (I really have 9.3% but that includes 1.8% for TDP which I predict will be INC ally)
BSP           4.2%
BSP allies   4.7% (using SP JD(S) INLD)
Left Front   3.9%

My current model BJP+ and INC+ numbers mostly match the survey.

Breakdown by region is similar to my as well

East
BJP+  43  (--)
INC+  25 (+4)

West and Central
BJP+  48 (--)
INC+  43 (+3)

North
BJP+  39 (-6)
INC+  21 (-1)

South
BJP+  18 (-7)
INC+  38 (-1)

Where BJP+ is very strong in the West relative to 2014 just like my model.  Whey my model diverge from the poll is it has BJP+ losing ground in North India whereas by model has the same but my model has grand alliances in UP (SP BSP INC RLD ..) driving voters that do not like this alliance to BJP.  

What is concerning to BJP is the following question: Modi government should be given another chance in government

Yes 39 No 47.  To be fair regional wise it does not look as bad
South:                 24/53
North                   44/45
West and Central: 44/49
East:                    46/40

BJP+ could gain a lot seats in the East and squeeze out some victories in the North West and Central areas to get a small majority.

PM preference is also looking better for Rahul Gandhi
Modi      34 (-3)
Gandhi   24(+4)

Satisfied with Modi government:
Yes   47 (-4)
No    47 (+7)

Andhra Pradesh  26(-9)/68
Bihar                 69(+7/29
Gujarat              55(-2)/40
Karnataka          50(-1)/45
Kerala                28(+4)/64
Madhya Pradesh 48(-6)/46
Maharashtra       51(-5)/47
Odisha               63(-3)/28
Rajasthan          57(-2)/37
Tamil Nadu        19(-14)/75
Uttar Pradesh    49(-6)/44
West Bengal      44(-2)/45
Telangana          29(NA)/63

It seems Bihar is going well for Modi although alliance problems could cause problems for NDA and BJP can look forward to making gains in Odisha.  Several key Northern states are on the edge (UP MP and Rajasthan)

[jaichind]

Another LS by-election in May 28th is Maharastra's Bhandara - Gondiya.  Here the BJP MP who was in INC but defected to BJP and won in 2014 on the BJP ticket.  Recently this BJP MP defected back to INC and resigned his seat.  In 2014 it was

BJP     50.6% (backed by SHS)
NCP     38.2% (backed by INC)
BSP       4.3%

Since this seat was allocated to NCP in 2014 the INC decides not to field the now incumbent INC MP.  SHS and BSP both will not contest so the race will be BJP vs NCP.   SHS is still deciding if it will support BJP NCP or stay neutral.

My 2019 model for this seat has it at

NCP    40.6% (backed by INC)
BJP     33.1%
SHS    15.0%

Using 2014 assembly segment results you get for this seat

BJP    34.3%
SHS   11.3%
INC    18.4%
NCP   18.3%
BSP    10.7%

So BJP support and INC-NCP support are similar.  The by-election will be decided by how the SHS and BSP vote goes.  The BJP will run a SHS defector who was in the BJP in 2014 before defecting to SHS because he was not given a ticket to contest in the 2014 assembly elections.  BSP vote will most likely go NCP given recent problems the BJP has had with the Dalits.

[jaichind]

CSDS National Survey has other breakdowns of LS election by key states.  MP and Rajasthan which are coming up later in 2018 are problematic for BJP

MP  
INC    50
BJP    40

Rajasthan
INC    44
BJP    39

With MP being a surprise.  Rajasthan is about where I think it will be (before the Modi campaign kicks into gear and pushes up the BJP vote share) but that the BJP is that far behind in MP is a surprise.

Other states are

UP (big trouble for BJP and that BJP is this low is unexpected and if INC joins up with SP-BSP then BJP will face total wipe-out in UP)
BJP       35
INC       12
SP-BSP  53

Bihar (unexpected lead of BJP despite on the ground reports of a RJD-INC surge)
BJP+     60
INC+     34

Maharashtra  (problem here is SHS will almost certainly run separately so INC-NCP at 40 is bad for BJP)
BJP+    48
INC+    40

Gujarat (sounds about right)
BJP     54
INC     42

WB (looks about right with BJP being the main alternative to AITC pushing aside INC and Left Front)
AITC    44
BJP      24

Overall it has BJP+ at 274.  But that assumes SHS does not split off in Maharashtra which is very likely and will cost BJP at least a dozen seats.



All India voting intentions relative to prev polls


As mentioned before "Give Modi another chance" came out to 39/47 the poll has it broken down by community shows a BJP Upper Caste-OBC counter-consolidation against the opposition Muslim-Dalit consolidation could still be effective

 

[jaichind]

In Nagaland there will be a LS by-election to replace Neiphiu Rio who defected from NPF to form NDPP and captured power in Nagaland in an alliance with the BJP in a close race with NPF.

INC will now back its ancient rival NPF to battle against NPF.  On paper the election is NDPP vs NPF but it is really INC vs INC.  The NPF and NDPP candidates were both INC MLAs elected in 2013 and defected to NPF in 2015.  Both were re-elected in 2018 on the NDPP and NPF ticket but just 3 years ago both were INC MLA in opposition to the NPF government.

Of course now INC is a shell of itself even 3 years ago with NDPP-BJP and NPF pretty much taking over political space in Nagaland.  INC will have to capture power at the federal level to be relevant in Nagaland again. 

[jaichind]

By-polls in progress.  There seems to reports of fairly large scale EVM malfunction.  Opposition parties claims that the BJP fix is in by having EVM berak down in areas of Dalit or Muslim areas to reduce turnout there. 

[jaichind]

It would be useful to count the number of MLA that BJP, BJP allies and other pro-BJP (relatively) have as I think we are reaching peak BJP.  The next cycle of assembly election results will most likely lead to losses by the BJP in terms of MLAs.

If you go with

https://en.wikipedia.org/wiki/Member_of_the_Legislative_Assembly_(India)

Out of 4120 seats BJP has 1518, BJP allies 352.  If you add in other pro-BJP parties like AIADMK, YSRCP, AINRC, NUZP and BVA it is another 178.   So pro-BJP bloc is has 2043 out of 4091 seats (29 are vacant) which is exactly around half of the seats.

I suspect this is the peak of what the BJP can accomplish.  They are almost certain to lose seats in MP,Rajasthan, and Chhattisgarh assembly elections later this year.   In 2019 I seem likely they will lose seats in AP, perhaps make minor gains in  Telangana, most likely lose a bunch of  seats in Arunachal Pradesh (having 57 out of 60 means you can only go down), make good gains in Odisha, most likely lose seats in Maharashtra, Haryana, J&K, and Jharkhand.

It might at least a couple of election cycles before the BJP can get to this peak again.