The following explanation of the allocation of the Upper House proportional seats and the subsequent post-electoral speculation are my answers to email questions that I received. Since the questions came from a long-time resident of Japan well versed in the Japanese economy and politics, I thought that my answers could be of benefit to some of my other readers as well. (Confession: people who do not know me well will be surprised (dismayed?) at how much of what I blog on the Upper House election I am learning on the fly.)
I, as a Japanese adult in good standing with the law of the land, have two votes in this Upper House election, one for the local seats (Tokyo is a multi-seat district, and gets five seats) and another for the proportional seats. I can cast my proportional ballot in two different ways.
For the proportional seats, each political party fields a national list of candidates. I am free to award my one proportional vote to any party or, alternatively, to any one of the individual candidates on any one of the party lists. After the voting is over, all the proportional votes that the individual candidates has received are attributed to the respective parties that listed those candidates in addition to the votes that the parties has each received in its own right. Then, the proportional seats are allocated in proportion to the number of votes each party has received. Finally, the seats that a party has won are allocated first to the candidate on its list who received the most votes, next to the one with the second most votes, and so on until the number of proportional seats that the party has won has been reached. By way of illustration:
Imagine that Japan has twenty eligible voters and the Japanese Upper House has four proportional seats, two of which are up for election, and that two political parties, A and B, are contesting this election with two candidates each, A1 and A2, and B1 and B2, respectively. On voting day the twenty-person electorate breaks overwhelmingly for A1, a golf dad, who gets twelve votes, while A2, ex-METI policy wonk, receives only two, and Party A itself only one vote in its own name. Party B, doing slightly better than Party A, receives two votes in its own name. Alas, its candidates are not able to emulate this institutional success; although Zenda ex-president B1 manages to gain three votes, B2 fails to receive a single vote. The final tally:
A: 1+13+2=15; B: 2+3+0=5.
A gets both seats, which go to A1 and A2, and B none, since 15/2 > 5, although, individually, Party B nominally outpolled Party A two to one and B1 beat A2 individually three to two. Of course, in the real world, 96 seats are contested, half of them every three years, which are enough to ensure that iniquities like the example above, where 25% of the votes go to naught, do not occur. But even in that crude example, it is easy to see how celebrity candidates who appeal to sizable constituencies that are independent of the party base can benefit the party twice over by this institutionalized coattail effect. A very poor vote getter or two may well slide in ahead of other more popular candidates in other parties if some of his/her co-candidates do exceedingly well. Thus the scramble to recruit golf dads, ex-presidents, ex-thug motivational speakers, and other B-list celebrities.
I believe that the steak analogy holds not only for the individual candidates but also for the LDP if there is enough time between the breaking of the pension scandals and the election date for the public to get over it. To push the analogy, recover from the bad experience with the e.coli on all those rare Del Monicos and also feel that they've had enough of that sautéed salmon (which came with its own little salmonella scare anyway). If "steak" is too charitable, think, the devil you know? Familiarity breeds consent?
But how much, you ask? That, I have no idea. You could do worse than ask Gerry Curtis. In fact, you can do much better than read this blog; Professor Curtis wrote the book on Japanese politics. Literally. Many times over. I heard he's in town.
It's difficult to say where Mssrs. Abe and Ozawa will be six months from now; if they survive the immediate aftermath, they should still be around; I think that mid-January is a little too early in the Diet schedule for the situation to become untenable in the Upper House. As for the election itself, Mr. Ozawa is too unpopular within his own party to be able to eat his words and stay on in the event of an outright ruling coalition majority as the immediate outcome. I say this is so even in the likely event that the DPJ beats the LDP one-on-one. Otherwise, Mr. Ozawa can plausibly claim that the opposition has beaten the coalition and I think stay on (and will stay on; he bleeds politics), even if the coalition does manage to cobble together a working majority. As for Mr. Abe, I am increasingly coming around to the view that he will stick around unless the coalition loses very badly. Figures like 39, 40 have been bandied about, and 40 does look as good as any for what is purely a mass psychology threshold.
But what are the chances of a working majority in the Upper House? Ah, the 64-seat question. Assuming 45 LDP and 14 New Komeito seats as the base case scenario, the coalition needs five more seats to reach the magic number. One and possibly both of the two Upper House defectors from the opposition, who are not up for election this time around, is assumed to be available at the right price. One "unaffiliated" conservative candidate is an overwhelming favorite to win and is a sure bet to join the coalition if he does. At that point, the New People's Party should be able to put them over the top. But will the PNP cooperate? I have become more and more skeptical of this possibility, given all the pro-opposition moves that it has been making with regards to the individual elections. In any event, the LDP will try extremely hard to pry loose a couple of opposition members or three. I do think that this post-election period will be a particularly problematic occasion to defect to the ins, but here I am edging precariously close to the realm of pure speculation.
That's about it for now.