Japan 2016 - July 10

[jaichind]

朝日 (Asahi) projections



                 District        PR           Total
LDP              38           19            57
KP                 7             7             14
ORA               3             4              7
DP                19           11           30
SDP               0             1              1
JCP                 2            6              8
AO                 1            0               1
Pro-DP Ind     3            0               3

All things equal I do not buy this poll.  I do not feel the PR vote and district vote results are consistent.  It seems to project ORA underperforming on the PR vote section and falling to around 8% losing votes to LDP.  For LDP-KP the PR vote seems to imply a vote share around 51%-52% on the PR.  Even with ORA falling to 8% on the PR section I find it hard to believe that LDP-KP will win 51%-52% of the PR vote.  And if LDP-KP is at 51%-52% then LDP should crush DP in the district vote at a much higher level then this projection indicates.  

This projection looks very much in line with the Kyodo and Mainichi in which case I suspect those projections would contains the same flaws, in my view.

[jaichind]

A comparison between my projection with the most pro-LDP projection which is 朝日 (Asahi) reveals that for single member seats Asahi has it LDP 24 to Opposition 8.  I have LDP 21 Opposition 11 which several opposition victories neck-to-neck victories which could easily go the other way.  Asahi and I completely disagree on some fundamental dynamics which actually cancel each other out in large part ergo the single member district projections are not that far apart.  They are

1) Asahi has LDP-KP at a great result at around 51.5% in PR and ORA at a disappointing around 7.5% in the PR section as there is a swing from ORA to LDP.  While I have LDP-KP at 44% for PR of which some went to ORA which I have 9% at PR.

2) Asahi has LDP-KP getting an unusually swing away from, especially in the district votes, in the North due to TPP.  I factored that in as well but my model uses elasticity and trend at the prefecture level which already contains that trend as the North has been trending away from LDP anyway.

3) Asahi has LDP losing some of its PR vote in the district vote to the opposition plus ORA tactically voting against LDP for opposition candidates where ORA is not running.  I have ORA mostly voting for LDP where ORA are not running as well as some DP voters defecting to LDP in the district vote to protest the DP alliance with JCP.

It seems that difference between Asahi and myself for 2) and 3) canceled out a lot of the effect of 1).  We will see who is right July 10th.

[jaichind]

One surprise in the media projections has both 日経新聞(Nikkei)  and 日経新聞(Nikkei)  projecting that for 北海道(Hokkaido) the DP winning 2 of 3 seats with the LDP winning 1 as opposed to the conventional wisdom has it other way around.  This seems to be a function of the anti-LDP trend in the North due to TPP as well as the effect of a LDP rebel running in the fray splitting the LDP vote.  Even my LDP pessimistic projection has LDP narrowing winning 2 seats versus DP winning 1.

[jaichind]

共同通信社 (Kyodo) projection chart



                District        PR           Total
LDP              38           21            59
KP                 6             7             13
ORA               2             3              5
DP                18           10           28
PLP                2            0               2
SDP               0             1              1
JCP                 3            6              9
AO                 1            0               1
Pro-DP Ind     3            0               3

I allocated 2 of the 6 independent winners as PLP as they are de facto PLP.

Wow.  What a projection.   This projection has LDP-KP PR vote share at around 55% and ORA at 6%.  This is the same as 朝日 (Asahi)  but even more extreme.  Again this projection makes no sense as if LDP-KP can win 55% on the PR vote then it should just sweep the district elections but these projections indicates that DP and the opposition doing reasonably well in the district seats.  This makes no sense.

[jaichind]

日経新聞(Nikkei) projection chart.  This one is not clear what the medium projection is but  I will try to infer it by eyeballing it.  





LDP                   59
KP                     13
ORA                    5
DP                    27
PLP (ind)             2
SDP                    1
JCP                   11
AO                     1
pro-DP ind          2

LDP-KP bloc (including pro-LDP independent) at 73 which is pretty close to other media projections.

Nikkei also projected out of 32 single member districts LDP is ahead in 16, Opposition ahead in 2, and 14 neck-to-neck.  



Opposition ahead in 岩手(Iwate) and 沖縄(Okinawa) which makes sense.  Tossups surprises are 青森   (Aomori), 秋田(Akita), 岐阜(Gifu), 岡山(Okayama), and 愛媛(Ehime).  Rest are well know tossups.

[jaichind]

毎日新聞(Mainichi) chart on PR result.  I will have to eyeball it to get the medium projection



LDP       17
KP          7
ORA        5
DP        11
SDP        1
JCP         7

This would put LDP-KP PR vote share more like 48%-49% which I feel is more reasonable but still way too high but it is lower than other media projections for LDP-KP PR vote share.  It also has ORA around 9%-10% which is higher than other media projections.

[jaichind]

It is said that for the Tokyo Governor election DP-PLP-SDP-JCP will work toward a joint candidate.  Weather this will work I guess will depend on the candidate.

[jaichind]

Overall the positive surprises for the opposition from all these media polls seems to be

1) DP has a real shot at winning 2 seats in 北海道(Hokkaido)
2) DP competitive in 青森(Aomori) and may even have the upper hand
3) DP had upper hand in 宮城(Miyagi) and Center-left opposition independent has the upper hand in 山形(Yamagata)
4) 秋田(Akita) might be in play for DP
5) LDP-KP not as strong in 神奈川(Kanagawa) which makes LDP-KP plans to capture 3 seats with the help of a pro-LDP ex-YP independent unlikely to succeed  
6) JCP has a solid shot at winning the 4 seat in 大阪(Osaka) over ORA#2 candidate.  
7) DP#2 has a solid shot if not the upper hand to win the 6th seat in 東京(Tokyo)
8 ) Center-left opposition independent within striking distance of LDP in 愛媛(Ehime) which is a total shock.

On the other side of the ledger the positive news for LDP is that

1) LDP still very competitive in 長野(Nagano) and may have the upper hand.
2) LDP most likely has the upper hand in 滋賀(Shiga) where it was thought it was DP ahead.

[jaichind]

There is something very strange about these various media surveys.  I complained about this before but lets look at the two most extreme survey results in terms of likely LDP-KP vote share and compare to demonstrate my point.

We have Nikkan Gendai



                District      PR       Total
LDP            36           15         51
KP               6             7          13
ORA            3             3            6
DP             18           11          29
PLP             0              1           1
SDP            0              1           1
JCP             5              8         13
pro-DP Ind  5              0          5

and

共同通信社 (Kyodo)



                District        PR           Total
LDP              38           21            59
KP                 6             7             13
ORA               2             3              5
DP                18           10           28
PLP                0            0               0
SDP               0             1              1
JCP                 3            6              9
AO                 1            0               1
Pro-DP Ind     6            0               6

The Nikkan Gendai survey has LDP-KP at 22 PR seats which puts the LDP-KP PR vote share around 43.5%  The Kyodo survey has LDP-KP at 28 PR seats which puts the LDP-KP PR voet share around 55.5%.  Yet the Nikkan Gendai survey has LDP-KP winning 42 out of 73 district seats while Kyodo has LDP-KP winning only 44 district seats.

This would be like two separate USA polls where the gap between the GOP generic Congressional vote share has a gap of 12% between the poll yet the two poll has the GOP only winning around 15 more seats in the one poll versus the other.  One of the two, or most likely both, must be wrong.

Someone else posted a blog post complaining about something different.  They pointed out that if you read the details of the surveys done by two separate news agencies, in this case 読売新聞(Yomiuri)  and 日経新聞(Nikkei)  respectively





You noticed that both surveys claimed to have called 50,943 registered voters of which 27,640 responded.

Also both the 共同通信社 (Kyodo) and 毎日新聞(Mainichi)  also claimed to have polled "around 27,000" registered voters.

It seems that at least these 4 news agencies really just used the same pollster.  This not that absurd since for exit polls in the USA many TV networks just used the same consortium to do exit polls ergo the mess up in FL 2000.  

What is weird is that these news agencies, using the same poll, came out with different results, especially on the PR vote section.

I suspect the explanation for this is the centralized survey really just covered the district votes and the PR section are derived by polls done by the media agencies themselves.  This would explain why the media projection would have PR and district result which are not consistent since they are from different sources.  Of course I would expect some basic QA before they publish this stuff.  Someone should have said "these results on the PR and district section does not seem to jive, lets look more into this before just throwing it over the fence."  I guess no one did this.

[jaichind]

毎日新聞(Mainichi) poll on support for Constitutional change has it being opposed 36/45



LDP voters are for 58/28
KP voters are 44/38
ORA voters 47/42
DP voters 18/75
JCP voters 15/80

It seems surprising that ORA voters are split down the middle on this.  It shows that the ORA vote still has a good chunk of the old center-right anti-LDP part of the pre-2012 DPJ vote base which is less hawkish but has a libertarian streak.

[jaichind]

DP leader 岡田 克也 (Okada Katsuya) indicated that if the DP candidate for his home prefecture 三重(Mie) failed to win against the LDP he would step down as leader of DP.  Not a bad bet.  If DP lost Mie then DP will most likely lose most of the tossups which means Okada would be tossed out anyway by the DP.

[jaichind]

Final NHK poll

Abe approval  47/36



Party ID

LDP      36.4
KP          5.5
ORA       2.1
DP         8.9
PLP        0.1
SDP       0.5
JCP        4.8



As expected all major parties gained in polarization right before an election.  The rule of thumb for NHK polls right before an election is the LDP-KP PR vote share should be NHK polls plus 1%-3% which would put LDP-KP PR vote share at 43%-45% which puts in fairly in line with other polls once we calibrate them with previous results to filter out pro-LDP biases.  

It seems with this NHK poll we should be looking at LDP-KP PR vote of around 44% which is the basis of my prediction model.  The main risk is that DP levels of support seems low given LDP-KP levels of support although this is not true for polls that ask explicitly for PR vote in which case DP is well into the double digits.  I guess it is this risk that made all the most media projections have LDP-KP PR vote  at the 48%-56% range.

One good news for opposition alliance is that approval for the 4 party opposition alliance (DP-PLP-SDP-JCP) which had been well underwater are now nearly even 45/47



Abenomics approval is 44/49 which is not bad under the circumstances

 

[jaichind]

Who would be the oddsnon favorite to secede him? Hosono?

Perhaps.  I suspect 長妻 昭 (Nagatsuma Akira) and 蓮舫(Renhō) will be in the running as well.  A lot depends on how DP loses the election, badly or marginally. 

[jaichind]

If Abe fails to amend the constitution in a referendum, would he resign?

I doubt it will really get to that.  Unless LDP wins by a unexpected landslide in July 10th, to get a 2/3 majority will require LDP, KP, ORA plus various non-LDP right parties (AEJ, PJK) and various ex-PFG/ex-YP independents.  To get that entire bloc to agree to a controversial change for the constitution seems pretty hard.  Of course if LDP completely hits a home rome and does much much better than expected then Abe will have the mandate to try.  In that case it is not clear he will lose the referendum and may likely win it given the resulting demoralization of the center-left opposition.   

[jaichind]

毎日新聞(Mainichi) Tokyo poll.

Abe cabinet approval 42/37.  Right before 2013 Upper House elections it was 51/28
Constitutional change support 32/51.   Right before 2013 Upper House elections it was 40/42

 

[jaichind]

RealpoliticsJapan has a chart to summarize each media projection of the 32 1- seat districts



Which is

毎日新聞(Mainichi)  LDP 26 Tossup 3 Opposition 3
読売新聞(Yomiuri) LDP 16 Tossup 13 Opposition 3
日経新聞(Nikkei)  LDP 16 Tossup 14 Opposition 2
朝日 (Asahi) LDP 20 Tossup 3 Opposition 8 (something is missing this is 31 and not 32)
産経新聞 (Sankei) LDP 22 Tossup 2 Opposition 8
共同通信社 (Kyodo) LDP 22 Opposition 10

Asahi and Kyodo are more anti-LDP whereas Yomiuri, Nikkei and Sankei are more pro-LDP.

Opposition stated goal is to win at least 10 out of the 32 seats.  It seems they might be on the verge of doing so or will be close.  My current projection has them winning 11 out of 32 which is slightly more optimistic than the most pro-opposition projection.   

[jaichind]

There is some growing concern that DP will get hurt on the PR vote due to the party name change.  On the PR vote, the voter WRITES the name of the party or the name of a candidate from the party list.  Commonly known abbreviations are ok.  So for example LDP is officially called 自由民主党.  But 自民 (Liberal Democrat) or even 自(Liberal) would be acceptable for a vote for LDP on the PR slate since people refer to them as  自民 (Liberal Democrat)  or 自(Liberal) all the time.  Back when DP was DPJ the official name was 民主党, a vote for 民主(Democratic) or 民(Democrat) would be acceptable as a vote for DPJ.  Now that DPJ is now DP or 民進党, a vote for 民進(Democrat Progressive) or 民 (Democrat) would be ok but now 民主(Democratic) would not.  The reason is LDP in theory has 民主(Democratic) in its name and now  民主(Democratic)  is not in DP's name of 民進党.  So  traditional DPJ voters who are used to writing   民主(Democratic) on their ballot might not get counted under DP.  DP is making the argument that mass media still refers to DP often as 民主(Democratic)  so a vote for 民主(Democratic)  should be a vote for DP.  Each prefecture will make its own call how to handle this but this issue is causing concentration at DP headquarters.  I guess that is the cost of changing party names.

[jaichind]

Yahoo came out with an astounding projection.

At an high level it has
 

                   PR           My estimated          District              Total
                                 PR vote share
LDP            16                 31.0%                41                     57
KP                8                 15.5%                  3                     11
ORA             6                 11.5%                  4                     10
DP             11                 21.5%                13                      24
JCP              7                 13.5%                  9                     16
Pro-DP Ind.                                                 3                       3

Total Seats


District seats


PR seats


It seems to have a KP PR vote share holding up but massive KP -> JCP tactical voting in various multi member districts leading JCP to win 9!!! district seats.  It has DP PR vote share holding up but massive defection of the DP vote in single member districts to LDP leading the LDP to sweep the single member districts 27-5 over the opposition alliance with LDP actually winning 沖縄(Okinawa).   The ORA  vote which is very high also voted heavily for LDP in single member districts.  It seems SDP and PLP vote completely fell apart and went wholesale to DP or JCP.  PJK and NPR got a good bloc of votes but not enough to get a seat but their votes went to either LDP or ORA helping ORA to capture a seat in 東京   (Tokyo) and 神奈川(Kanagawa) where the DP vote was split between two DP candidates.

LDP-KP vote share is 46.5% which is low when compared to other media projections but this projection sees a lot of potential LDP votes going to ORA on the PR section.  Net net this projection seems a major polarization with LDP and ORA on one side and JCP on the other with KP defecting to JCP in some places and DP defecting to LDP in other places.

[jaichind]

On the PR vote, the voter WRITES the name of the party or the name of a candidate from the party list.

Good God, that's almost as asinine as Australian Senate ballots.

Both systems are similar.  Japan is a bit harder since unlike Australia the names of the PR party list candidates do not appear on the ballot.   You have to remember his/her name and write it down.  This leads to fractional results.  Say there are two candidates with last names of Tanaka in the running.  A voter who writes down Tanaka without the first name will still have that vote counted but it gets prorated between the two Tanakas in ration of the valid votes they received.  So if Tanaka A and Tanaka B got vote shares of 4 to 1 then that vote counts Tanaka A 0.8 votes and Tanaka B 0.2 votes.  The ballot paper does list the names of the parties but a voter has the option of writing down the party he or she want to vote for in addition to the name of the candidate leading to the problem I point out above.

[jaichind]

It seems to have a KP PR vote share holding up but massive KP -> JCP tactical voting in various multi member districts leading JCP to win 9!!! district seats. 

What would be the profile of a KP -> JCP swing tactical voter ? O_o

It turns out such a situation is quite possible in theory.  The typical JCP and KP voter and working class and lower middle class in terms of class profile.  The KP vote base is actually fairly pacifist and could be turned off by Abe's subtle attempts to move toward the hawkish Right.  It is said that during the anti-Abe protest of Summer 2015 over the new security legislation there were KP supporters involved without KP party support.   The KP leadership wants power and social respectively which justifies the LDP alliance. 

Just a few days ago, a JCP MP had to step down due to comments he made that the Japanese military budget is a "budget to kill people."  ORA also just announced backing of LDP candidate in Okinawa to try to capture the hawk vote from LDP.  In Yahoo scenario this leads to Dove/Hawk polarization where the ultra-dove KP voter votes for KP on the PR slate but then votes for ultra dovish JCP in the district vote.   LDP voter who are angry over this also vote ORA in the PR vote even as they vote LDP in the district vote. Hawkish DP votes still vote DP in the PR vote but defect to LDP in the district vote.  PLP SDP and VPA voters vote JCP in the PR section or tactically for DP.

[jaichind]

Japan election website go2senkyo.com was able to collate and show detailed versions of the updated media proejctions of 4 of the 6 media houses that came out with projections on a seat by seat basis.  They are  朝日(Asahi) 毎日(Mainichi) 日経(Nikkei) and 読売(Yomiuri).





There seems to be a bug in this chart for 朝日(Asahi) where in 岩手(Iwate) they labeled it DP when it should be pro-PLP independent.  

We also have from before 共同(Kyodo)

 
With this I can construct projection charts for all of them to compare

For 朝日(Asahi)  (I am going to fix the bug in the chart) it would be

1 seat districts LDP-opposition 23-9

                       PR           My estimated          District              Total
                                     PR vote share
LDP                19                 37.5%                38                     57
Pro-LDP Ind                                                     1                       1
KP                    7                 14.0%                  6                     13
ORA                 4                    8.0%                  4                       8
DP                 11                  21.5%                17                     28
SDP                 1                    2.0%                 0                       1
PLP                  0                    1.0%                 0                       0
JCP                  6                  12.0%                 3                       9
Pro-DP Ind.                                                     4                       4    (1 DP, 2 PLP, 1 AO)


For 毎日(Mainichi)   it would be

1 seat districts LDP-opposition 27-5

                       PR           My estimated          District              Total
                                     PR vote share
LDP                19                 37.5%                43                     62
Pro-LDP Ind                                                     0                       0
KP                    7                 14.0%                  6                     13
ORA                 4                    8.0%                  3                       7
DP                 11                  21.5%                15                     26
SDP                 1                    2.0%                 0                       1
PLP                  0                    1.0%                 0                       0
JCP                  6                  12.0%                 3                       9
Pro-DP Ind.                                                     3                       3    (1 DP, 1 PLP, 1 AO)



For  日経(Nikkei)   it would be

1 seat districts LDP-opposition 23-9

                       PR           My estimated          District              Total
                                     PR vote share
LDP                18                 35.0%                38                     56
Pro-LDP Ind                                                     0                       0
KP                    8                 15.5%                  7                     15
ORA                 3                    6.0%                  3                       6
DP                 12                  23.5%                20                     32
SDP                 0                   1.5%                   0                       0
PLP                  0                    1.0%                 0                       0
JCP                  7                  13.5%                 1                       8
Pro-DP Ind.                                                     4                       4    (1 DP, 2 PLP, 1 AO)


For 読売(Yomiuri)   it would be

1 seat districts LDP-opposition 22-10

                       PR           My estimated          District              Total
                                     PR vote share
LDP                16                 31.5%                37                     53
Pro-LDP Ind                                                     0                       0
KP                    7                 14.0%                  7                     14
ORA                 5                  10.0%                  3                       8
DP                 12                  23.5%                19                     31
SDP                 1                    2.0%                 0                       1
PLP                  0                    1.0%                 0                       0
JCP                  7                  14.0%                 3                     10
Pro-DP Ind.                                                     4                       4    (1 DP, 2 PLP, 1 AO)


For  共同(Kyodo)  it would be

1 seat districts LDP-opposition 22-10

                       PR           My estimated          District              Total
                                     PR vote share
LDP                21                 41.5%                38                     59
Pro-LDP Ind                                                     1                       1
KP                    7                 14.0%                  6                     13
ORA                 3                    6.0%                  2                       5
DP                 10                  19.5%                18                     28
SDP                 1                    2.0%                 0                       1
PLP                  0                    1.0%                 0                       0
JCP                  6                  12.0%                 3                       9
Pro-DP Ind.                                                     5                       5    (2 DP, 2 PLP, 1 AO)


My projection which I will post soon is very similar to the 読売(Yomiuri) projection which in turn is the most negative on LDP.   The  毎日(Mainichi) projection is the most pro-LDP.  It is interesting that the two most pro-LDP media outfits  読売(Yomiuri) and 日経(Nikkei)  came out with the most pro-DP projections.  We also have the 共同(Kyodo) with the most extreme PR vote share projection for LDP-KP yet at the same time having pretty conservative district seat share projections for LDP-KP given that large vote share.

[jaichind]

If we take the average of the 4 media projections it comes out to LDP-KP 71, DP-PLP-SDP-Independents 33, ORA 7, and JCP 10.

If we do a comparison of Japan Upper House elections since 2001 it would be interesting.  First we should use the Japanese term 第三極  or Third Pole which refers to various non-LDP center right forces (they might be LDP rebels, hawks, populists, libertarians) that often take votes away from the main center-left anti-LDP opposition bloc.   In this election the Third Pole parties are ORA, NPR, and PJK although only ORA is relevant.   A chart of Upper House election results which lumps pro-LDP independents as part of LDP has and Center-Left bloc include parties like DPJ PLP SDP DP etc etc.

                LDP+              Third Pole            Center-Left          JCP
2001          80                       6                       30                    5
2004          60                                                57                    4
2007          46                       3                       69                    3
2010          60                     12                       46                    3
2013          76                     17                       20                    8
2016          71                       7                       34                    9

In 2001 Third Pole was Ozawa's LP which back then was an anti-LDP conservative party which split the anti-LDP vote and gave the LDP+ a massive landslide.  For 2004 LP merged into DPJ which evened the odds.  In 2007 Third Pole are LDP populist splinters NPN and PNP which worked with DPJ to trounce LDP+.  In 2010 Third Pole are LDP libertarian and hawk splinter YP SPJ and NPR which took some LDP votes but also took a bunch of anti-LDP votes.  2013 Third pole are YP and JRP which took enough anti-LDP vote to lead to a collapse of the Center-Left Bloc.  2016 on these projections would be the continued recovery of the Center-Left bloc as well as the rise of JCP as the Third Pole declines.

[jaichind]

三重(Mie) poll



It has

Abe Approval 39/37
District seat   DP 39 LDP 36
PR vote LDP 33 DP 29 KP 11 JCP 7

This the first poll I have seen that actually has KP PR vote above 10% anywhere.  Of course KP will get 13%-15% they never poll at this level.

My current model has DP winning this seat DP 51 LDP 48.5 so this poll fits with that narrative. 

[jaichind]

時事通信(Jiji) projection  



which if you read their article you can construct

1 seat districts LDP-opposition 23-9

                       PR           My estimated          District              Total
                                     PR vote share
LDP                17                 33.5%                38                     55
Pro-LDP Ind                                                     0                       0
KP                    7                 14.0%                  7                     14
ORA                 4                    8.0%                  3                       7
DP                 13                  25.5%                19                     32
SDP                 1                    2.0%                 0                       1
PLP                  0                    1.0%                 0                       0
JCP                  6                  12.0%                 2                       8
Pro-DP Ind.                                                     4                       4    (1 DP, 2 PLP, 1 AO)

This projection has a last minute surge for DP from LDP and JCP on the PR section even as the district results are not affected.  LDP-KP PR vote share at 47.5% which is low by media projection standards.

[jaichind]

Latest 朝日 (Asahi) poll

Abe approval 41/36 from 45/36 last poll
Want LDP-KP to continue their majority 44/35
Want LDP-KP-ORA-PJK to gain 2/3 majority for Constitutional change 36/41

PR vote

LDP   35  (-3)
KP       7  (--)
ORA     7 (+3)
DP     16 (+1)
JCP      6  (--)

Given historical house effects of Asahi poll this seems to imply a LDP-KP PR vote share of around 45%-47% range.