Monday, June 28, 2010

Faith In Numbers (1)

In the dip in the popularity polls of the Kan Cabinet after the quite responsible shout out by Prime Minister Kan Naoto to the Liberal Democratic Party to stop the catcalls from the sidelines and instead join with the ruling coalition in reforming country's revenues base (i.e. - raise taxes) the nation's liberal news organizations are beginning to lose their marbles panicking.

Here are the latest seat projections for the various parties in the upcoming House of Councillors election from The Asahi Shimbun.

And here are the projections from the Tokyo Shimbun

According to the median view of the of the Asahi, the DPJ will end up with exactly the same number of seats -- 54 -- in the House of Councillors as it now controls. In the Tokyo Shimbun, the party will actually lose two spots, slipping to 52 seats. In both cases, the DPJ will have to go hat in hand post election in search of (a) coalition parther(s), as it needed to win 60 seats in order to have a 50%+1 majority in the House of Councillors.

Not having a majority in the upper house is not the end of the world, of course. The most important piece of legislation, the national budget, goes into effect 30 days after it is passed by the House of Representatives, no matter what the House of Councillors does with it. The DPJ holds a comfortable 307 out of 482 480 seats* majority in the House of Representatives - so the budget is safe. However, for all other legislation or the confirmation of officials, the DPJ will have to rely on the kindness of strangers to make it to the 122 votes it needs to get anything done on the legislative front.

There also so problematic issues of control of committee chairmanships in the House of Councillors should the DPJ fail to win an outright majority. These are not on a par, however, with the simple fact that no legislation is guaranteed passage without a 50%+1 majority of the House of Councillors members present -- which opens up some interesting opportunities in the case the DPJ fall just a few votes short of an absolute majority for sudden absences of independents from the chambers at crucial moments. Two of the seven independent members of the House of Councillors are up for reelection this year, meaning that at very least, in the next Diet session, 119 votes might be sufficient for passage of legislation which the DPJ really want to see passed.

Anyway, both the projected polls leave the DPJ far short of the majorities it needs. Just how they do this seems unsound.

First is the pass given to the microparties in both polls. There are not just five, as listed in both sets of results. There are eight seven. Two of the left: the Social Democratic Party and the Women's Party. Six on the right: People's New Party, Sunrise Party (Tachiagare Nippon), New Renaissance Party (Shinto kaikaku), The Spirit of Japan (Nihon Soshinto) and the Happiness Realization Party.

The larger the number of little parties, the greater likelihood they will zero each other out the d'Hondt system, which is the method used to assign seats in the proportional part of the election. In each of the last three House of Councillors elections a party had to secure 2.2% of the national vote to win even a single proportional seat. At present, only the SDP is polling greater than 2%.

To see how the winnowing process happens, check out the wonderful d'Hondt system calculator here.

Using the calculator, I have run a basic scenario, assuming conservatively that

1) just as many voters show up in 2010 as in 2007
2) skeptical non-aligned voters choose Your Party as the non-DPJ, non-LDP alternative
3) that the Communists and the Socialists stay put
4) that Your Party captures the business lobby vote
5) that the conservative microparties steal votes that would have gone to the LDP, and
6) some voters will vote for the LDP out of nostalgia or force of habit

The results of the simulation, which you can see here, assigns the 48 available proportional seats this way:

DPJ 19
LDP 11
New Komeito 6
Your Party 6
Japan Communist Party 3
Democratic Socialist Party 2
Sunrise 1

One can fiddle about and get New Renaissance a seat that the cost of one DPJ seat. However, failing a huge jump in turnout of folks boiling mad at the DPJ, these are likely to be the results we see in the morning paper on July 12.

Sic transit nunc gloria the microparties and their purported seats.

If my simulation can be faulted, it is for being too generous to the LDP. The party chose last year to run as the anti-Ozawa Ichiro party, choosing stances not based upon their logical coherence but on whether or not Ozawa was for them or against them. Other than anti-Ozawaism, the party failed to produce a convincing reason for its wavering supporters to stick around, the reason why Masuzoe Yo'ichi quit, dismissing "LDP" as the acronym for "Lousy Dumb Party."

Which brings up my second, much larger objection to the Asahi and Tokyo Shimbun projections: the presumption of viability of the LDP as a political force in the district elections - the subject of my next post.


*Kobayashi Chiyomi resigned from her seat the day after the close of the regular Diet session. The by-election for her Hokkaido #5 district seat will be in October. One other seat is also vacant.


Janne Morén said...

Well, your 19 proportional seats match Asahis median value, so as far as the DPJ goes it doesn't seem to materially change their fortunes.

Matching the previous seats won't be too bad either. Done well, it can create better, pragmatic, longer-lasting solutions than legislation formed out of one single ideological perspective. Another upside of coalition work is that you neutralize the fringe within your party at little cost to yourself. You'd "love to" impose a 100% tax on all capital profits, or sell the entire health-care system to the highest private bidder (or whatever), but unfortunately your coalition partners have yet to see the ideological light so you'll just have to postpone it for a later election cycle.

By the way, I'd hesitate to put Happiness Realization Party on the right; I'd rather put them on the "batshit loony" end of their own, private scale perpendicular to any direction existing in reality.

Anonymous said...

My prediction: Happiness Realization party takes all seats. The next day, Korea is unified by winged manatee angels.

Anonymous said...

A couple of quibbles.
The Women's party is not of the left, its a quasi-religious party devoted to the chairman of a cosmetics company. And they along with the Nihon Soushintou are included in the "Shoha", meaning that they didn't even have the support of the Happiness Realization Party. So that adds up to seven micro parties, I'm not sure who else you are counting.

Another thing is that the Lower House only has 480 seats to begin with, so you should be saying "out of 478 seats" not 482.

But what really should matter is that the district elections are completely different animals. The 16 prefectures that elect 2 people each will probably end up with one DJP member and one LDP member, since Ozawa's plan of taking both seats isn't going to work. On the other hand, only one party will win in the 29 districts that only elect one person, and by definition, those are the smaller, more rural prefectures.
This leads to these 29 prefectures having more influence as opposed to their size in the Upper House elections, sort of similar to the "swing states" in the U.S. Presidential elections.
So, you can't go simply by the poll numbers, you have to look into each prefecture's situation.
But I should wait until your next post before jumping to conclusions.

MTC said...

Anonymous -

Agreed. The necessary changes have been made.

As to the first error, absolutely no excuse, as I was looking at the official running tally at