2019 Japan Unified Local Elections(April) and Upper House elections (July 21st)

[jaichind]

A couple of noted political analyst projections.  Frankly I know these two guys and last few election cycles their projections were even further off than my projections.  Anyway their projections are slightly more negative for LDP than CW 



          District    PR      Total
LDP         37      17        54
KP            7        7        14
JRP          3         4          7
DPP          3         4         7
CDP          9       11       20
RS            1         1        2
SDP          0         1        1
JCP           3         5        8
OPPN      11                 11  (Center-Left Opposition backed independents)

          District    PR      Total
LDP         38      17        55
KP            7        7        14
JRP          3         4          7
DPP          3         3         6
CDP        12       13       25
RS            1         1        2
SDP          0         1        1
JCP           2         4        6
OPPN        8                   8  (Center-Left Opposition backed independents)

Of course these projections are already obsolete because they did not take into account that RS's 山本太郎(Yamamoto Tarō)  will run on the PR list and not Tokyo.

My current projection which takes that into account is more positive for LDP.  I suspect after the mega polls this weekend my projection will become the most anti-LDP projection out there.

          District    PR      Total
LDP        41      18        59
KP            7        7        14
JRP          2         5          7
DPP          3         3         6
CDP        11       11       22
RS            0         1        1
SDP          0         1        1
JCP           3         4        7
OPPN        7                   7  (Center-Left Opposition backed independents)

[jaichind]

JX poll on 大阪(Osaka)

The level of support for the district candidates are

JRP
LDP
KP
JCP
JRP
CDP
DPP

The PR section has it at

JRP  30
LDP 17
JCP   8
KP    6
CDP  6

The PR level of support shows that the JCP lead over JRP for the 4th and last seat will not last and once the JRP machines goes into gear to allocate its vote correctly between the two candidates it should be JRP 2 LDP 1 KP 1 with JCP CDP and DPP out in the cold.  Only way JCP can beat JRP for the 4th seat if there are large scale tactical voting by CDP and DPP voters for JCP.  This seems unlikely as 大阪(Osaka) has been trending JRP since the JRP blowout victory in local elections back in April.

[jaichind]

Abe "promised" that the rise of consumption tax from 8% to 10% later this year will be the last one within the next 10 years which might be fiscally unsustainable but he has no choice lest he loses more support in the Upper House elections due to the consumption tax increase issue.

https://www.bloomberg.com/news/articles/2019-07-03/japan-s-abe-signals-no-sales-tax-hike-beyond-10-for-10-years

[jaichind]

Election campaigning officially starts today.

[jaichind]

Gender balance of candidates of various parties.  LDP KP have very low number of women candidates while SDP CDP JCP have high number of women candidates (along with Center-Left-JCP joint candidates)  DPP and JRP are in the middle

[jaichind]

長野(Nagano) poll.

Abe cabinet approval/disapproval 40.5/59.1

Support/opposition for Abe Constitutional revision 30.3/54.8

PR Vote

LDP   24.9
KP      5.1
JRP    1.6
DPP   3.5
CDP  11.0
SDP   1.4
JCP    3.7

Grouped you get

LDP-KP         30.0
Center-Left   15.9
Third Pole      1.6
JCP               3.7

Historically LDP-KP PR vote in 長野(Nagano) is around 7% less than all Japan levels.  

Back in 2016 PR poll was

LDP              33.6
KP                  4.1
PJK                0.2
NPR               0.2
ORA               1.3
DP                20.1
PLP                0.1
SDP               1.2
JCP                8.9

Grouped you get

LDP-KP         37.7
Center-Left   21.4
Third Pole      1.7
JCP               8.9

Since 2016 there seems to be a lot more undecided with JCP losing the most ground relative to 2016.

[jaichind]

Noted political analyst 有馬晴海(Arima harumi) indicated that for the 32 1- member districts, despite the recreation of the Center Left-JCP alliance they would not be able to repeat the 2016 21-11 split between LDP and  Center Left-JCP alliance due the the decrease in the salience of the TPP issue.  He figures it will be 23-9.  I agree and my projection is 24-8.

He has the 32 seats as 14 LDP well ahead 1 Opposition well ahead (I assume 沖縄(Okinawa)) and 17 battleground seats.  17 is surprisingly large.  I tried to come up with this list of 17 where the opposition has a chance or are well ahead and all I came up with

青森(Aomori)
岩手(Iwate)
宮城(Miyagi)
秋田(Akita)
山形(Yamagata)
福島(Fukushima)
山梨(Yamanashi)
新潟(Niigata)
長野(Nagano)
三重(Mie)
滋賀(Shiga)
愛媛(Ehime)
大分(Ōita)

Which is 13.  I find it hard to get 4 more where the opposition has even an nominal chance.  Gun to my head I came up with

佐賀(Saga) - LDP put it its list to focus on
群馬(Gunma) - it is an open seat
徳島(Tokushima)/高知(Kōchi) - 高知(Kōchi) has an anti-LDP lean  and LDP put it on its list to focus on
鹿児島(Kagoshima) - there is a LDP rebel in the fray

Other than 沖縄(Okinawa) the places the LDP is likely or could lose in my view are the list of 13 above.  The other 4 are long shots. 

佐賀(Saga)  and 徳島(Tokushima)/高知(Kōchi) are only there because LDP said they might be in trouble there.  To give LDP credit back in 2016 愛媛(Ehime) which looked like a sure thing for LDP had the LDP sounding the alarms that they might be in trouble.  In the end LDP narrowly won. This year the same thing might play out where 佐賀(Saga)  and 徳島(Tokushima)/高知(Kōchi) might turn out to be the surprise in terms of how close it became.

[jaichind]

First media projection to come  out base on the mega survey I keep on talking about is Asahi whose projections have  historically  pro-LDP lean (making up for its anti-LDP policy house effect)

Medium projection are

          District    PR      Total
LDP         41      18        59
KP            7        7        14
JRP          3         5          8
DPP          3         2         5
CDP        10       11       21
RS            0         1        1
SDP          0         1        1
JCP           4         5        9
OPPN        6                   6  (Center-Left Opposition backed independents)



Which would produce

LDP+           73
Center Left   34
Third Pole      8
JCP               9

And when added to the class of 2016 it gives us

LDP+          144
Center Left   71
Third Pole     15
JCP              15

With anti-Constitutional revision bloc (Center Left + JCP) at 35.10% of the seats.  

This is a bit more pro-LDP than pre-campaign polls and projections but still has Center-Left + JCP being able to block Constitutional revision

[jaichind]

岡山(Okayama) poll

PR vote

LDP  28.8
KP     8.9
JRP   0.7
DPP  1.2
CDP  6.8
RS    0.2
SDP  0.3
JCP   3.9


Historically LDP-KP outperforms on PR in 岡山(Okayama) around 5%-6% above its national level of support.  The responding rate is so low one cannot gain much from this poll other than the fact that a lot of the 2017 HP vote has gone over the CDP and not DPP.

[jaichind]

Just be clear that is going on with these media projections which is the story I mostly constructed by reading these polls over each election cycle.

First, 2-3 weeks before the election a common unknown pollster does a massive poll all over Japan with sample sizes of 100K or even 200K respondents.  The raw data are given the various media houses  which pay for this massive effort with the cost spread over all the media houses.  Then each media house uses their own weights plus various ground reports to generate their projections.  These projections are inversely correlated with the political lean of the media house in question.  So anti-LDP media house like 朝日( Asahi), 共同(Kyodo) and 毎日(Mainichi) come out with pro-LDP projections while pro-LDP media houses like 産経(Sankei),  日本経済(Nikkei) and 読売(Yomiuri) came out with anti-LDP projections.  This is mostly overcompensation for house effects and to some extend to lower expectations of their readers so after the election they can claim that the side they are for "exceeded expectations" which is what their readers want to read.  All of them will have initial projections but then adjust them over time which seems to be mostly about herding.

Asahi always seem to come out with their numbers first.  But the others will come out in a day or two.

In 2016 the initial Asahi projection was


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

The final result was

                 District        PR           Total
LDP              36           19            55
KP                 7             7             14
ORA               3             4              7
DP                21           11            32
PLP                0             1              1
SDP               0             1              1
JCP                 1            5              6
Pro-DP Ind     5             0              5

which meant that in 2016 Ashai overestimated LDP (and JCP as well) a bit relative to real results.

[jaichind]

Yomiuri PR poll (diff from a week ago)

LDP  36 (-4)
KP      6(+1)
JRP    7(+1)
DPP   3(+1)
CDP  10(--)
RS     ??
SDP    1(-1)
JCP     4(--)

LDP loses some ground to JRP but overall not much change.

[jaichind]

Kyodo poll on party support (not PR vote) with diff relative to its last poll right before the 2016 Upper House elections

LDP  35.9(+2.4)
KP     5.4(+0.4)
JRP    3.8(+0.5)
DPP   2.1
CDP   8.6
SDP  0.9 (--)
JCP   3.8(-0.2)

So the Center-Right parties (LDP KP JRP) are gaining relative to 2016.  JCP lost a bit of ground. I suspect DPP+CDP is most likely greater than the 2016 levels of support for DP but cannot be for sure.

It also has breakdown by prefecture

[jaichind]

共同(Kyodo) projections

          District    PR      Total
LDP         44      21        65
KP            7        7        14
JRP          4         5          9
DPP          2         3         5
CDP        11         9       20
RS            0         0        0
SDP          0         1        1
JCP           4         4        8
OPPN        2                   2  (Center-Left Opposition backed independents)



Which would produce

LDP+           79
Center Left   28
Third Pole      9
JCP               8

And when added to the class of 2016 it gives us

LDP+          150
Center Left   65
Third Pole     16
JCP              14

With anti-Constitutional revision bloc (Center Left + JCP) at 32.24% of the seats.  This means the anti-constitutional revision barely misses being able to block  constitutional revision.

Back in 2016 共同(Kyodo)'s first cut projection was

                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
Pro-DP Ind     6            0               6

The final result was

                 District        PR           Total
LDP              36           19            55
KP                 7             7             14
ORA               3             4              7
DP                21           11            32
PLP                0             1              1
SDP               0             1              1
JCP                 1            5              6
Pro-DP Ind     5             0              5

So again just like 2016 共同(Kyodo) which is anti-LDP has an even greater pro-LDP bias in its projections relative to anti-LDP 朝日( Asahi) which itself have a smaller pro-LDP bias.

[jaichind]

Both the 朝日( Asahi) and 共同(Kyodo) look fishy to me on the JCP front.  Both has JCP winning 4 district seat.  It seems to me that that is only possible if it beats out JRP in 神奈川(Kanagawa) for its 4th and final seat (which I agree with as I feel the same way) AND JCP beating out CDP for the second seat in  京都(Kyoto) which seems unlikely.  The only other place JCP can win its 4th seat is in 大阪(Osaka) where it can beat out JRP for the 4th and final seat.  But with 朝日( Asahi) and 共同(Kyodo) having JRP winning 3 and 4 district seats respectively this seems very unlikely.  I think both of them are wrong about 京都(Kyoto) but I will look forward to reading their prefecture by prefecture breakdown to see how they get to JCP to having 4 seats.

Also I suspect these projections are probabilistic projections and NOT a sum of medium projections (like UK exit poll projections) so they might not even have 4 seats where the medium case has JCP winning them but JCP having a lot of less than medium cases of winning a seat and they add up to 4.  We will see.

But the main narrative of these polls are clear.  They have LDP JRP and to some extent JCP over-performing at the expense of CDP-DPP with  共同(Kyodo) being more extreme on that front especially with the 1- member districts.

[jaichind]

Assessment of different media houses of the 32 1- member district before and after their processing of the mega poll

                                 Before                                 After
                        LDP  Tossup  Opposition       LDP   Tossup   Opposition
読売(Yomiuri)      16     13           3               17        12            3
日経(Nikkei)        16     14           2               19        11            2
朝日(Asahi)         20       4           8               20         5             7
產經(Sankei)       22       2           8               20         8             4
共同(Kyodo)        22       0         10               22         8             2


Depending on the media house it seems the maga poll shifted the projections toward LDP

[jaichind]

富山(Toyama) PR poll

LDP  36.4
KP     4.2
JRP    2.7
DPP   7.2
CDP   8.7
SDP   0.7
JCP    1.9



富山(Toyama) has a historical lean where LDP-KP PR vote is around 4.5% above the national average.  The response rate again is low but it seems fairly at par with recent polling nationally.   Main takeaway is the DPP is fairly strong here which seems to indicate that DPP is not doomed yet and that DPP could still do OK in rural areas.

[jaichind]

沖縄(Okinawa) PR vote poll

LDP   20.7
KP      7.3
JRP     2.1
PNHK  2.0
HRP    0.4
DPP    3.8
CDP    9.3
RS      0.8
SDP    6.4
JCP     6.9


LDP of course is much weaker in 沖縄(Okinawa) than rest of Japan.  What is key here is that PNHK actually has 2.0% of the vote. PNHK is overtaking HRP here as the main protest party.

[jaichind]

 朝日( Asahi) poll on how non-aligned voters will vote on PR when compared to the same poll in 2013 and 2016.  LDP around 40% CDP around 20% JRP and JCP around 10%

[jaichind]

石川(Ishikawa) poll on PR

LDP  43.8
KP     4.3
JRP    1.8
HRP   0.6
DPP   2.9
CDP   7.4
SDP   0.4
JCP   1.9



石川(Ishikawa) LDP-KP PR tends to votes around 6.5% greater than national average.  This poll seem pretty position for LDP

[jaichind]

鳥取(Tottori) and 島根(Shimane) poll on PR

鳥取(Tottori)
LDP  37.7
KP     8.1
JRP   
DPP   0.6
CDP 11.5
SDP   0.3
JCP   2.5

島根(Shimane)
LDP  45.1
KP     5.4
JRP   
DPP   0.9
CDP  6.0
SDP  2.3
JCP   2.2



Not sure why JRP is not polled. 

In 鳥取(Tottori) LDP-PR vote tend to be around 9% above national average while in 島根(Shimane) LDP-KP PR vote tend to be around 10.5%-11% above national average

[jaichind]

徳島(Tokushima) poll on PR

LDP  35.7
KP     5.3
JRP    3.0
PNHK 0.3
DPP   1.7
CDP   7.6
RS     0.1
SDP   0.5
JCP    2.7



LDP-KP PR vote tend to be around 5.5% higher than national average in 徳島(Tokushima)

[jaichind]

毎日(Mainichi) leaking polling on votes share some of which are fairly shocking

Overall JCP and PNHK seems to be doing way better than expected.  I suspect JCP and PNHK performance are exaggerated by this leaked vote share breakdown.

北海道(Hokkaido) (3- member)
LDP  29.8
CDP  28.4
LDP  17.4
JCP   14.0
DPP    7.7
PNHK  2.2
Very surprising that JCP is so much ahead of DPP and within striking distance of LDP

青森(Aomori)
LDP   48.7
CDP   47.5
PNHK  3.8
Very surprising how close CDP made this

岩手(Iwate)
LDP   48.0
OPPN 47.3
PNHK  4.7
Ozawa forces usually under poll so LDP still should lose. Still the LDP incumbent who is a Ozawa rebel is making a race of this

宮城(Miyagi)
CDP  51.3
LDP  44.3
PNHK 4.4
Surprising that the gap is this big. LDP should be slightly ahead here

秋田(Akita)
OPPN  57.9
LDP    39.6
PNHK   2.5
A shock that LDP can be so far behind in a place that should be in the bag for LDP.  The OPPN candidate has deep roots here which it seems is making a large impact

山形(Yamagata)
OPPN   51.9
LDP     45.0
PNHK     3.1
This should be neck-to-neck yet LDP seems well behind

福島(Fukushima)
LDP    56.1
OPPN  38.7
PNHK   5.2
This should be neck-to-neck but LDP incumbent seems to be well ahead as PNHK is doing fairly well

茨城(Ibaraki) (2- member)
LDP    54.6
CDP    20.3
JCP     12.0
JRP      6.2
PNHK   6.9
LDP lead seems surprisingly large.  PNHK result seems to be a shock

栃木(Tochigi)
LDP    54.0
CDP   43.6
PNHK   2.3
Looks about right

群馬(Gunma)
LDP    49.8
CDP    43.7
PNHK   6.5
Looks about right other than very strong PNHK showing

埼玉(Saitama) (4- member)
LDP    31.4
CDP   27.2
JCP    16.1
KP     15.6
DPP     5.3
JRP     2.8
PNHK  1.1
KP seems unusually weak and JCP seems very strong

千葉(Chiba) (3- member)
CDP   33.3
LDP    24.9
JCP    17.8
LDP   14.7
PNHK  8.1
Shock result with JCP beating out LDP for the 3rd and last seat.  It seems PNHK is eating into the LDP vote.  毎日(Mainichi)  still predicting LDP wins the third seat so even they do not believe their own poll

神奈川(Kanagawa) (4- member)
LDP   28.0
CDP   21.5
JCP    16.4
KP     13.5
JRP      8.4
SDP     4.7
DPP     3.4
PNHK   2.1
JCP seems very strong while KP and JRP seems to be underperforming

東京(Tokyo)
LDP   18.9
JCP    17.2
KP     14.1
CDP   10.7
CDP   10.0
LDP     9.6
PNHK   8.6
JRP     4.7
SDP    3.7
DPP    0.7
Very strong JCP performance while JRP and DPP performance is weak with PNHK eating into JRP and DPP vote

新潟(Niigata)
LDP   51.5
OPPN 46.2
PNHK  2.3
Very surprising that LDP is ahead.  The LDP incumbent had several embarrassing gaffs and he is lucky that the LDP did not just deselect him

富山(Toyama)
LDP   62.3
DPP   37.7
Looks about right

石川(Ishikawa)
LDP  60.5
DPP  39.5
Looks about right

福井(Fukui)
LDP   57.3
JCP   31.2
PNHK 11.6
PNHK result is a shock.  It seems PNHK are picking up the anti-LDP anti-JCP vote

山梨(Yamanashi)
LDP     48.7
OPPN   39.4
PNHK   11.9
This should be a tossup but the LDP candidate is strong while the OPPN candidate is weak.  It seems that PNHK is picking up a lot of the anti-LDP vote and giving a walk to LDP

長野(Nagano)
DPP   55.7
LDP   34.8
PNHK  8.7
LAB    0.8
Another PNHK shock.  DPP incumbent was well ahead but it seem PNHK here is picking up a lot of LDP support

岐阜(Gifu)
LDP   52.0
CDP   44.3
PNHK  3.7
Looks about right

静岡(Shizuoka) (2- member)
LDP   35.9
DPP   26.6
CDP   20.5
JCP   12.0
PNHK  5.0
LDP seems to be under-performing

愛知(Aichi) (4- member)
LDP  37.1
CDP  19.8
JCP   14.8
DPP  13.9
KP     7.3
JRP    4.2
SDP   1.2
PNHK 1.3
Shock surge for JCP which pushed KP out of the running.  KP had warned that they might be in trouble here but their result in this poll seems catastrophic

三重(Mie)
OPPN  53.3
LDP    43.1
PNHK  3.7
OPPN is a bit stronger than expected but LDP are the underdogs here

滋賀(Shiga)
LDP   48.9
OPPN 46.0
PNHK  5.1
This is suppose to be neck-to-neck which is what we are seeing

京都(Kyoto) (2- member)
LDP  52.5
JCP   23.4
CDP  18.5
PNHK 5.3
Very strong LDP performance and very weak CDP performance yielding second seat to JCP

大阪(Osaka) (4- member)
JRP    30.3
LDP    17.9
JRP    13.6
KP     13.2
JCP    11.9
CDP    9.5
PNHK  1.7
DPP    0.9  
Looks about right other than CDP under-performed and allowed JCP get ahead of CDP.

兵庫(Hyōgo)
LDP    31.2
JRP     21.1
KP      18.4
CDP    17.0
JCP      8.6
PNHK   3.7
JRP over-performed but CDP withing striking distance  

奈良(Nara)
LDP     51.0  
OPPN  45.9
HRP      3.1
LDP is under-performing here

和歌山(Wakayama)
LDP    71.2
OPPN  28.8

鳥取(Tottori)/島根(Shimane)
LDP    51.4
OPPN  30.9
PNHK  15.4
LDP is under-performing and support going to PNHK as the OPPN candidate has JCP background

岡山(Okayama)
LDP    65.4
CDP    31.9
PNHK   2.7
Looks about right

広島(Hiroshima)
LDP    34.3
OPPN  24.3
LDP    23.8
PNHK  8.8
JCP     7.7
Looks about right but PNHK is way over-performing

山口(Yamaguchi)
LDP    68.8
DPP    26.5
Looks about right

徳島(Tokushima)/高知(Kōchi)
LDP     53.2
OPPN   30.2
PNHK   13.7
LDP doing worse than expected but with OPPN has JCP background the vote LDP lost is mostly going to PNHK

香川(Kagawa)
LDP     50.6
OPPN   47.6
PNHK   1.7
A lot closer than expected

愛媛(Ehime)
OPPN   58.1
LDP     37.9
PNHK    1.7
A shock LDP blowout defeat

福岡(Fukuoka) (3- member)
LDP    34.2
KP      23.3
CDP    20.3
JCP      7.9
PNHK   6.8
Looks about right except for PNHK performance

佐賀(Saga)
LDP   55.9
DPP   44.1
Looks about right

長崎(Nagasaki)
LDP   52.5
DPP   40.4
PNHK  7.1
Again strong PNHK performance

熊本(Kumamoto)
LDP   50.7
OPPN 44.1
PNHK  5.2
LDP doing worse than expected

大分(Ōita)
LDP    50.5
OPPN  44.1
PNHK   5.1
Looks about right

宮崎(Miyazaki)
LDP    48.6
CDP   42.5
HRP     8.9
LDP doing worse than expected

鹿児島(Kagoshima)
LDP            49.0
OPPN          27.2
LDP rebel    23.8
LDP rebel cut into both LDP and anti-LDP votes

沖縄(Okinawa)
OPPN  71.9
LDP    24.6
PNHK   3.0
Blowout victory for OPPN which is unlikely.

[kaoras]

As much as I would love those results they are clearly not happening . I could see the NHK thing but why would the JCP surge so much now? Conditions seems less favorable now with CDP in the picture than in 2013-2016.

Is that poll/projection designed to shore up LDP vote?

[jaichind]

As much as I would love those results they are clearly not happening . I could see the NHK thing but why would the JCP surge so much now? Conditions seems less favorable now with CDP in the picture than in 2013-2016.

Is that poll/projection designed to shore up LDP vote?

I suspect they might be tainted by pro-PNHK elements in these leaks.  In 2016 1- member districts there are cases where HRP won 2%-3% of the vote which represents the protest vote.  This time most of that will go to PNHK.  I fail to see why the PNHK vote will be higher then that.  If you take out the PNHK bias there are clear pro-JCP biases which could be true.  Part of this could also be relative CDP and DPP under-performance which makes JCP stand out in places like 北海道(Hokkaido), 千葉(Chiba),  東京(Tokyo), 神奈川(Kanagawa), 愛知(Aichi), and 京都(Kyoto).  毎日(Mainichi) and 朝日( Asahi) do share the assessment that JCP is doing unusually well in places like It could be in district seats.  Not clear it is the the scale of the the leaked numbers.

[jaichind]

栃木(Tochigi) PR poll

LDP  44.8
KP     5.6
JCP    0.8
DPP   1.8
CDP 14.1
RS    1.7
SDP  0.8
JCP   2.0


The old YP vote aligned with 渡辺 喜美(Watanabe Yoshimi) plays a big role here.  It seems that this time that vote which went to HP in 2017 is now going to LDP.  RS support here is surprisingly strong