Japan 2016 - July 10

[jaichind]

With 17.8% of the vote it is

LDP    49.2%
DP      50.8%

A dump of the urban vote came in.  DP takes the lead.

[jaichind]

With 27.3% of the vote it is

LDP    49.5%
DP      50.5%

[jaichind]

With 28.9% of the vote it is

LDP    49.6%
DP      50.4%

[jaichind]

In Kyodo 3rd  with 31% of the vote counted it is

DP        72.9%
ORA      17.1%
Ind         5.1%
PJK         3.0%


PJK at 3% is pretty sad.

[jaichind]

The various TV channels projects LDP wins in Hokkaido 5th.  I do not see how given how close the vote is but I guess they are using projections based on how each zone is voting. 

[jaichind]

With 29.2% of the vote it is

LDP    49.9%
DP      50.1%

[jaichind]

In Kyodo 3rd  with 39% of the vote counted it is

DP        72.0%
ORA      18.3%
Ind         4.6%
PJK         3.0%

ORA has to rethink its name.  Running in an election in a non-Osaka prefecture with the word Osaka in the name of the party seems to be poisonous

[jaichind]

With 31.4% of the vote it is

LDP    50.4%
DP      49.6%

[jaichind]

NHK count seems to be faster now so I will go with them

With 44% of the vote it is

LDP    50.5%
DP      49.5%

[jaichind]

With 51.6% of the vote it is

LDP    49.6%
DP      50.4%

DP retakes the lead.  But all media outlets insist on their call for LDP.  I guess they know what they are doing.

[jaichind]

With 66.3% of the vote it is

LDP    53.3%
DP      46.7%

Some LDP urban strongholds came in.

[jaichind]

In Kyodo 3rd  with 99% of the vote counted it is

DP        65.4%
ORA      20.8%
PJK         6.5%
Ind         4.6%

PJK made a small comeback.  Exit polls were very accurate.

[jaichind]

With 70.9% of the vote it is

LDP    54.0%
DP      46.0%

I can see why all those media outlets made the call that early when it was neck-to-neck.  I still feel the final results will be closer than this.

[jaichind]

In Kyodo 3rd total result

DP      65,051   65.4%
ORA    20,710   20.8%
PJK       6,449     6.5%
Ind.      4,599     4.6%

Back in 2014 it was

LDP    59,437   35.8%
DPJ    54,900    33.1%
JCP    26,655    16.1%
JIP     24,840    15.0%

It seems that DP did manage to merge into it the DPJ-JCP vote from 2014.  ORA did manage to capture the majority of the JIP vote from 2014 which is to be expected since Kyodo is part of Kinki.  What should be disappointing is that whatever LDP-KP voters that managed to come out seems to split their vote and did not vote for ORA.  This calls into question ORA's ability to capture votes outside the Hashimoto JRP core from Kinki.

[jaichind]

With 75.9% of the vote it is

LDP    53.8%
DP      46.2%

[jaichind]

With 79.3% of the vote it is

LDP    53.4%
DP      46.6%

[jaichind]

Looking at the results of Hokkaido 5th in zones that are done reporting it seems that DP was able to consolidate the DP-JCP vote from 2014 which would have made the election 50-50.  The LDP victory seems to have come from NPD bloc support for LDP.

[jaichind]

With 92.3% of the vote it is

LDP    53.3%
DP      46.7%

[jaichind]

All votes counted in Hokkaido 5th



LDP    135,842   52.4%
DP      123,517   47.6%

Amazing how close NHK exit poll was on this one as well.

Back in 2014 it was

LDP    131,394   50.9%
DPJ      94,975   36.8%
JCP      31,523    12.2%

LDP successfully consolidated the LDP-KP vote got the anti-JCP vote from NPD and the old JIP base.  This seems to show that Abe' strategy of harping on the anti-JCP theme is effective against the DP-JCP alliance strategy.  

[jaichind]

I guess the main lessons from these two by-elections are

1) LDP-KP vote mostly intact and not too far off its 2014 levels.
2) DP-JCP alliances does work on the ground and will be able to mostly merge their voting blocs
3) LDP still has the advantage given that is anti-JCP message can peal off the old JIP and ORA vote bases
4) ORA's brand not so hot outside of Osaka and most estimates on how it will do in the Summer 2016 are mostly likely inflated.

[jaichind]

There is already news that ORA will be having an internal debate on the name of he party.  The setback in Kyodo plus problems with TCJ merger talks is leading to some within the party to advocate for a name change.  Problem is that the Hashimoto core in Osaka is most likely going to oppose this.  So basically this is just a micro version of the JIP split back in July 2015 between the Osaka bloc vs Tokyo bloc.  I guess ORA has to worry about keeping its Osaka base against the local LDP which would stop at nothing (including an alliance with JCP) to take it down and at the same time try to expand outside Osaka.

[jaichind]

Kyodo 3rd district by-election exit polls



LDP only 28% of the voters which is small and of course a result of no LDP candidate running.  As expected LDP voters split their vote between DP ORA and PJK.   DP which is 23% of the electorate and JCP which is 7% of the voters of course went to DP.   ORA which was 8% of the voters as usual went for ORA.   Many of 28% of the voters of the unaligned are really DP voters are as expected went DP.

[jaichind]

I would say that this was a good night for the Opposition. This result was the same as 2014, but with Opposition forces combined, it shows how much can be done.

All things equal I would agree.  There were several special factors going on in Hokkaido 5th which might make the results harder to justify extrapolating to the rest of Japan but I believe they will most likely cancel each other out.  Namely:

1) The old LDP incumbant 町村 信孝(Machimura Nobutaka) was actually a powerful faction leader (the same faction that Abe is from in fact.)  Given his power within the LDP it is very likely that his vote in 2014 was artificially high since he was effective in bring home the pork.

2) Machimura passed away in 2015 and his son-in-law ran on the LDP ticket.  The sympathy factor most likely boosted the LDP vote. 

3) As I mentioned before NPD went from backing DPJ in 2014 to backing LDP to 2016 which must have swung some votes toward LDP.

Most likely 1) canceled out 2) and 3) so the net narrative that DP-JCP alliance will retain the DP and JCP vote shares is still fairly valid.  But out of these factors 3) is the more problematic one for Abe/LDP.  3) was necessary, in retrospect, for LDP to win and indeed do well in Hokkaido in any future election.  But the negative image of NPD will wear down on LDP's image in the rest of Japan especially when most of the mideeds of NDP leader 鈴木 宗男(Suzuki Muneo) took place while he was in the LDP pre-2005.  Having Suzuki part of the LDP alliance brings back bad memories of the "Old LDP."

[jaichind]

Time for an update on DP-PLP-SDP-JCP grand alliance deals.  Over the last couple of weeks deals were reached in

岩手(Iwate), 秋田(Akita),  群馬(Gunma), and 岡山(Okayama).    Of this Akita is most consequential as it puts it in play but it still leans LDP. Iwate was leaning PLP anyway but this deal makes it a lock.  Rest will still go LDP deal or no deal.



青森(Aomori) - DP candidate -> solid LDP
岩手(Iwate) - PLP candidate -> soild PLP
宮城(Miyagi) - DP candidate -> tossup/lean LDP
秋田(Akita) - DP candidate -> tossup/lean LDP
山形(Yamagata) - center-left independent (former DPJ MP) -> tossup
栃木(Tochigi) - center-left independent -> solid LDP
群馬(Gunma) - DP candidate -> solid LDP
新潟(Niigata) - PLP candidate -> tossup/lean LDP
富山(Toyama)- center-left independent -> LDP landslide
石川(Ishikawa) - center-left independent -> LDP landslide
福井(Fukui) - center-left independent -> LDP landslide
山梨(Yamanashi) - DP candidate -> tossup
長野(Nagano) - DP candidate -> tossup/lean DP
滋賀(Shiga) - DP candidate -> tossup
鳥取(Tottori/
島根(Shimane) - center-left independent -> LDP landslide
岡山(Okayama) - DP candidate -> Solid LDP
山口(Yamaguchi) - center-left independent -> LDP landslide
徳島(Tokushima)/
高知(Kōchi) - center-left independent -> solid LDP
愛媛(Ehime) - center-left independent -> LDP landlide
長崎(Nagasaki) - DP candidate -> LDP landslide
熊本(Kumamoto) - center-left independent -> LDP landslide
宮崎(Miyazaki) - center-left independent -> LDP landslide
沖縄(Okinawa) - OSMP candidate -> lean OSMP

The most critical ones left to get a deal on are 福島(Fukushima) (rumor is that a deal is close), 三重(   Mie) (this one is most difficult as the local DP branch is fairly hostile to JCP), and 大分(Ōita).  A deal in the first two makes them tossup/lean DP and a deal in the third on makes it tossup.

[jaichind]

If you reran the 2014 election except you pooled together all of the DP/JCP vote - and added in the JIP vote from outside of Osaka, what would the seat breakout be?

Because that's what you seem to suggest a snap election would resemble.

Yes, the magazine  週刊現代 or Shūkan Gendai (Modern Weekly) did this fairly recently.  What it did was for 2014 JIP candidates that joined ORA they do not count those votes to the opposition alliance vote share and for those 2014 JIP candidates that joined DP they count those votes for the opposition alliance.  This is a very close proxy to all 2014 JIP votes in Osaka for ORA and all JIP votes outside Osaka for DP.  See below for a post I had on this a few days ago.

It comes out to LDP-KP at 257 out of 475 seats for a bare majority.  My main critique of this approach is that it just does a brain-dead adding up of votes from 2014 instead of a dynamic approach to try to  model how the 2014 JIP vote might flow, what is a likely swing away from LDP-KP since 2014, and what JCP alliance might have on DP votes defecting to LDP as well as anti-DP JCP voters staying home.

But as unsophisticated this model was, it worked for Hokkaido 5th district.  This model would still project a close LDP victory since in 2014 the LDP candidate barely won above 50% of the vote which is indeed what took place in the by-election.

Magazine 週刊現代 or Shūkan Gendai (Modern Weekly) Business came out with their projection on what a double election would look like.  It seems fairly negative on LDP's chances based on successful DP-JCP alliance, a swing away from LDP-KP, and ORA mostly being a dud outside of Osaka with anti-LDP votes consolidating around DP.  It has



Which would translate into for Lower House where LDP-KP would be reduced to a very narrow majority

           District       PR         Total
LDP       167         58          225    
KP            6          26           32
DP         105         64         169
JCP           1          19           20
ORA          5          12           17
SDP          1            1            2
PLP          2            0             2
Others     8            0             8

Overall, these sort of PR seat results would imply that LDP-KP vote share would be around 41% or even lower as the ORA could have many stranded votes in PR section outside Kinki where it has some support but not enough to cross the threshold to get seats which in turn would push up everyone else's seat share relative to their vote share including the LDP-KP.  These same PR seat results would also imply that DP would be around 36%-37% in terms of vote share which would be a complete shock.

For Upper House it has (and it is a shame it does not break out the PR seat and district seats)

LDP            44
KP              14
DP              45
JCP              5
ORA             6
AEJ             0
SDP             2
PLP              2
OSMP          1
Ind.            2

Frankly there lots of things that seems fishy about this projection other than it being to optimistic about how well a DP-JCP alliance can work on the ground.   At lot of these numbers seems inconsistent.  Namely

1) If JCP PR seat count for Lower House is 19 then the JCP PR vote share is most likely around 10%-11%.  But in Upper House they only project 5 seats for JCP.  Knowing that JCP is for sure going to win one of the Tokyo District seats (with 6 up for grab and JCP vote base being very disciplined) then that leaves JCP winning 4 PR seats in the Upper House which would more imply JCP getting around 8%-9% of the PR vote.  Even if we accept the JCP voter might tactically vote on PR section as well it is much more likely that JCP will over-perform in the Upper House PR section (which is an all Japan zone so the threshold for getting seats is low) whereas the PR section in the Lower House is by region making the threshold for getting seats much higher..

2) It has SDP getting 2 Upper House seats.  This seems implausible.  SDP does not have any prospects of winning a district seat in the Upper House so the 2 has to come from PR.  But that means that SDP PR vote share would be around 4%-5% which does not seem very likely when last few election cycles it has been in the 2% range.  With the narrative of opposition consolidation around DP in this poll a surge of SDP to 4%-5% seems even less likely.  One possibility is that in 大分(Ōita) the DP-JCP opposition unity candidate which is still being negotiated and most likely a DP candidate might end up being a SDP candidate or DP running as SDP candidate given that Ōita is the one prefecture where SDP has significant strength.  

3) It has ORA at 6 seats in the Upper House.  As long as ORA still retained some of its base on Kinki it should in theory capture 2 大阪(Osaka) and 1 兵庫(Hyōgo) district seat.  If it does not and DP-JCP can work out alliances deals in both prefectures then ORA could be reduced to just 1 Osaka district seat.  In that situation ORA will have 5 PR seats which in turn would imply a PR vote share of around 10-11%. Alternatively ORA could win 2 district seats and have 4 PR seats which would imply a PR vote share of  8%-9%.  I guess all this is possible.  My main issue with this projection is that I find it hard to believe that DP-JCP would get their act together (and withdraw candidates they already nominated)  to prevent ORA from winning 3 district seats.  But if that comes to pass then ORA would be at 3 PR seats which imply a ORA PR vote share of around 7% which I find too low.  I would think that given the DP-JCP alliance the Center-Right anti-LDP DP voter would most likely vote ORA so ORA should not be that low.

4) It has PLP winning 2 seats in Upper House.  Hard to figure out which 2.  It is clear PLP will win in 岩手(Iwate).  To fit the narrative the PLP should also win the tossup seat in 新潟(Niigata) where the PLP candidate is the opposition unity candidate.  But that would leave PLP with no PR seats which is possible but seems to contradict the narrative.  In 2013 LDP landslide PLP came very close to winning 1 PR seat.  For PLP in 2016 to do almost as bad as 2013 given the turnaround in the election climate as per this poll also seems contradictory.  

This projection has 2 independents winning.  Since this is projecting DP led opposition doing much better than expected then the 2 independents winning must be the second DP candidate in 愛知(Aichi) who is running as an independent which will now win the second DP seat in Aichi and the ex-DPJ ex-incumbent of  山形(Yamagata) who left DPJ but now is running as an independent with all opposition support.  The LDP backed independent in 神奈川(Kanagawa) under this projection is not expected to win the 3rd out of 4 Kanagawa seat.