Tuesday, November 25, 2014

Where Things Could Go Right For The DPJ

Today's Nihon Keizai Shimbun makes a big deal about the Democratic Party of Japan's running 90 fewer district candidates in this year's election than in 2012 and about 100 fewer than the LDP is running. (Link - J)

Pretty stark numbers, underlining the seeming the failure of Japan's single member districts to do as foretold and give rise to a two major party system.

Well, maybe.

In 2012, in a record low turnout election, a demoralized DPJ had candidates running in 264 districts. In 227 of those districts, the DPJ candidate was running against both an LDP candidate and at least one other non-Communist major party candidate.

Fast forward to today.

The Your Party, 69 candidates in 2012, is no more. Two of its members just joined the DPJ last week. In 2012 the Your Party received 5.2 million proportional seat votes.

The Mirai Party, 121 candidaters in 2012 and 3.4 million proportional votes, is no more. Mirai's remnant, the Life Party, is down to five legislators after two defected to the DPJ last week. Life will be running 17 candidates in this election.

The Japan Restoration Party ran 172 candidates in 2012 and received 12.2 million proportional seat votes. Its two fission products, the Japan Innovation Party and the Party of the Next Generations, are running 73 and 34 candidates, respectively. In the most recent polls, support for the JIP stood at about half the support for the DPJ. NexGen polls at 0%.

In 2012, the DPJ basically fought the election with both hands tied behind its back, having to defend a legislative program that was not even its own. There were non-LDP, non-DPJ viable national alternatives.

Not this time.

This year is much more a two party race, with the JIP having to choose whether to play the role of regional champion or national nuisance. It is the LDP and Abe who are running on their records.

The road of course is bumpy. DPJ and JIP candidates are running against each other in 24 districts. The DPJ is running against another non-Communist party in 22 more districts. In total the DPJ is running against non-Communist opposition in 53 districts.

Last time I looked, 53 out of 173 is a better ratio than 227 out of 264.

Hence the crippling importance of believable, inspiring leadership. The numbers are there for the DPJ to mess up Abe Shinzo's December 14 plans. A flag carrier the voters could believe in would have a huge impact, driving turnout and moving the marginal voter to wish to punish the LDP and Abe for their broken wheel of good fortune.

Whatever happens, the 2014 election will result in a great winnowing of the parties. NexGen and Life will collapse into nothingness; the Socialists may join the pair in final, pathetic oblivion. JIP will lose many of its proportional seats.

What will remain is the ruling coalition's two parties, a revived DPJ, a pugnacious but rump JIP and the Communists.

Not quite a two party system, but lookin' more and more like one.

Later - Check out the intelligent points from A. Sutter in the comments.

Later still - The Mainichi Shimbun has checked in with its English-language version of the story. (Link)


A.J. Sutter said...

That said, and despite the desire of some Japanese politicians to create a two-party system (e.g. Ozawa Ichiro's aspirations way back when to create a Westminster-style system in Japan -- not that that's been so 2-party recently), what's so good about a 2-party system?

E.g., Germany has been getting along pretty well with a multi-party system for most of the past six or seven decades.

One of the problems with a single-member district, first-past-the-post (FPTP) system is that many or even most peoples' votes are ignored. Close vote counts with lopsided results can contribute to tremendous polarization in a society. In such a system, a party could win a clean majority with just a bit more than 25% of the votes cast (depending on number of seats and number of voters). For example: suppose there were 480 seats and 60 million voters all perfectly apportioned at 125,000 voters per district. A party gets a majority if it wins 241 seats, to do which it needs only 62,501 votes per district. So the theoretical minimum it needs is only 15,062,741 votes (albeit precisely distributed), or 25.1%.

Party-list proportional systems such as Germany's result in a much better representation of the votes cast, especially when one uses a proportionality algorithm such as Sainte-Leguë. (Germany abandoned the D'Hondt algorithm for its national elections more than 30 years ago; we still use it here.) That gives German voters more of a voice -- and more interest in elections. Averaged over the past 5 national parliamentary elections, voter turnout was 70.3% in Germany vs 62.7% in Japan (and 44.3% in the last five 2-year election cycles in the US). No election system is free of arcane paradoxes, but FPTP's winner-takes-all character treats paradox as a virtue.

One reason Germany can function well is that their politicians have some maturity, and cooperate with each other; they even have grand coalitions from time to time. That might not describe the current lot of Japanese politicians, but cooperation and consensus aren't exactly alien to the Japanese national character. The Japanese legal tradition has already borrowed a lot from Germany. An election system that encourages politicians to emulate German parliamentarians would be a lot healthier than one that encourages them, especially when in opposition, to emulate the Republicans of the 112th and subsequent US Congresses.

now close to 60%-enfranchised Tokyoite in exile said...

As someone who happens to be tied to both countries – Japan and Germany – by birth, I couldn't agree more with A.J. Sutter's post if we are talking about the ideal electoral (and resulting party) system to properly represent current interest groups in Japanese, and arguably: any other modern society which, in my view, cannot be sensibly lumped into just two groups. And I also agree that the cultural/traditional preference for consensus-building over confrontation-based decision-making would be better suited by a multi-party system with its associated regular interparty negotiations and coalition-building process.

However, the electoral system that is currently in place is dominated by the FPTP/SMD vote. And in order to get that system to function for the voters, a two-party system is desirable: If the opposition parties come up with five different candidates, the government candidate wins by default and voters have no real choice at all. If there are only two truly competitive/serious (I don't mind a little Smiling Superman diversion in dull campaigns) candidates, then an FPTP election becomes at least a choice between two alternatives.

A thorough public debate on the best electoral system for the country is a good thing – and was, I think, somewhat neglected before the 1990s reform which was in no small part driven by Ozawa's ambitions. But, I'm not so sure if it would be a good idea to change the electoral system yet again in a fundamental way, or if it might not be better to get the existing system to work now that it is firmly in place and people have become used to it. To facilitate the functioning of the existing system, it would be helpful to further reduce or even abolish the proportional vote altogether, so that smaller parties would have to merge or be submerged into oblivion (Of course, Kōmeitō would be among the first to object). In contrast, a transition to a German-style system would instead require to expand the proportional vote/make it the decisive vote for the whole election (If my memory serves me right, one Komei proposal in the mini-electoral reform/district reform debate around summer [?] 2012 did entail a tie-up of some district seats with the proportional vote and was even picked up in the DPJ, but did not make it to the final three-party deal on tax hike/early elections/etc.).

Also related is the question of bicameralism (and Germany does have a – to put it simply [constitutional experts, close your eyes!]: federalist, quasi-upper house, though it is not directly elected). Contrary to the current tendency among some Japanese reformists who intend to scrap the upper house altogether, I think it might be a good idea to instead strengthen it and maybe make it a regionalist or even federalist chamber explicitly by name – there are significant urban/rural differences in Japan, especially in the beginning/upcoming demographic transition –, if the lower house turns towards proportional voting and a multi-party system where parties are used to resolve policy differences by finding some form of negotiated agreement. In an FPTP-based two-party system however, my impression is that bicameralism nowadays has a tendency to lead to unproductive blockades, because the two parties oppose each other fundamentally (in order to give the voters an apparently clear choice in [over-]professionalized [multimedia, quasi-permanent] campaigns for FPTP elections), and different control of the two houses then leads to unresolvable stand-offs as in recent US Congresses or to a lesser degree the "nejire" Diets between the 2010 and 2012 elections.

And then I talk to Italian and Japanese voters (modern history and political culture have a few interesting parallels), and wonder: Maybe the [dys]functionality of democratic systems is not related to electoral and party systems at all...?

Yet sleeping said...

The exiled Tokyoite wrote:

"the electoral system that is currently in place is dominated by the FPTP/SMD vote... a transition to a German-style system would instead require to expand the proportional vote/make it the decisive vote for the whole election"

This seems to assume that switching to a more proportional voting system would require big, disruptive changes. But the mechanics of the current 'parallel voting' system used in the House of Representatives are actually quite similar to the mixed member PR variants used in New Zealand, Scotland and Germany.

(The mixed member systems elect local members using FPTP, just as currently happens in Japan. However, when the proportional seats are allocated, any local members already elected for a party are taken into account and not doubled up. Deserve 20 members for your 40% of the vote and already have 18 local members? Only two more for you. This makes the overall result proportional - or at least more proportional.)

Because Japanese voters already choose a local member and a regional member when they vote, the practicalities of shifting to MMP ought to be relatively simple: No significant alterations to either electoral districts, the instructions given to voters, or the way party lists are managed.

a sleepy Tokyoite in exile said...

Yet sleeping does have a point from a technical perspective, and I am still pondering my personal view on this as a voter. But looking at the system as a whole, i.e. in terms of electoral strategy, party organization, the overall outcome of elections, etc., the proportional vote is not irrelevant, but certainly much less relevant than the district vote in the current parallel system. Look, for example, at the current candidate selection process: they are primarily selected for the districts, at least in the major parties who stand a chance in FPTP districts, the compilation of proportional lists is usually a secondary consideration – partly due to the dual candidacy/"sekihairitsu" system. And in the last three lower house elections, the district election was decisive enough so that even a very bad proportional result would have secured an assured majority for the winning party – 28% for the LDP in the 2012 proportional vote was not exactly what you would call a decisive victory even in proportional multi-party Germany. Considering all the implications, switching to a system where the proportional vote is dominant would constitute a quite fundamental change.

This also touches a – percieved or actual – weakness of the German system from a Japanese and probably even more so an Anglo-Saxon viewpoint. Although the German MMP system is explicitly designed to maintain the link between an MP and his constituency under an otherwise proportional party-list based system, it seems to be weaker than in Westminster-style systems or Japan. Certainly it can be said that, in contrast to Japan, campaign efforts in Germany are shifted more towards the proportional vote because, apart from minor deviations, it determines the overall composition of parliament (though it's called the secondary vote in German).

Personally, I would say that the benefits outweigh this and other disadvantages of a German-style MMP system and as far as pork-barrel politics is concerned, the stronger role of wider-area party organizations in choosing MPs may even be an advantage; but others may have strong objections to a system where the local district is no longer the primary unit to decide elections, but some perceivedly distant and anonymous proportional voting process which is then tied to a local vote in a – in my experience: for many people from countries with a majoritarian voting tradition – rather complicated manner. Remember, for example, even when the current parallel system was introduced, many people complained that you can no longer vote an MP out [because someone who loses his or her district may still be "resurrected" via the party list and keep a seat]. Though I would like to know if there are polls on how widely proportional list voting has become an accepted/desired part of Japanese elections by now or if many people think of it as an essentially unnecessary and unwanted appendix.

Yet sleeping said...

I wrote: No significant alterations to... the way party lists are managed.

Tokyoite in Exile wrote: [In] the current candidate selection process... they are primarily selected for the districts... switching to a system where the proportional vote is dominant would constitute a quite fundamental change.

Well, without going into too much detail, I think that's probably debatable. The basic orthodox strategy would still be to run your stronger candidates in single member districts (as I believe happens now), and to rank these same candidates highly on the party list as a contingency (as I believe happens now).

Why? Because the big parties still need to run candidates in the SMDs (they're still there to win, and they help harvest your proportional votes). For many candidates, their SMDs would still be safer or more achievable targets. Being publicised as the local candidate for a small geographic area makes it easier to attract a personal vote you can go out and work for yourself.

Tokyoite in Exile wrote: Although the German MMP system is explicitly designed to maintain the link between an MP and his constituency... [that link] seems to be weaker than in Westminster-style systems or Japan... probably even more so [from] an Anglo-Saxon viewpoint...

I can assure you that in the UK, Australia and New Zealand, the three systems I am most familiar with, both the mainstream media and political parties focus overwhelmingly on party leadership teams. The basic pattern of media coverage and campaigning is quite similar in each of these countries, even though their voting systems are, respectively, FPTP, a mix similar to Japan, and MMP. I believe Canada is pretty much the same. The prominent role of moneyed lobby groups in the US makes the comparison rather more difficult.

Incidentally, Michael, do you know off the top of your head whether changes to the voting system can be legislated directly (without special majorities etc)?

A.J. Sutter said...

Sleepy Tokyoite — the concluding thought in your last comment very generously supposes that Japanese voters have some sophistication about election systems. I teach law and politics at a university here to students who presumably are motivated by interest in those fields and who have taken classes in constitutional law, etc. — yet their awareness of election systems is roughly nil. The kind of sophisticated dialogue that has gone on in this thread is not often reproduced in the public sphere, apart maybe from a few scholars. But even some of them use very American and economistic “public choice”-style reasoning in their analysis, e.g. to defend dual candidacy.

Moreover, I’d expect that some people might be against proportional voting precisely because of the dual candidacy system, while being ignorant that there are alternatives. Similarly there is some support for getting rid of the Sangiin because people follow incumbent politicians’ misdirection and focus on the expense. And at a more wholistic scale, every year I have some students who say democracy is a bad system because Japanese politics are terrible and Japan is a democracy.

Japanese voters’ current attitudes have been formed in a Platonic cave built by Jimintou and a compliant press. You rarely hear about Germany being a role model, about the advantages of truly distinct bases of representation in parliamentary upper houses, or about the fact that Japan, while much better than say China, Zimbabwe or even Mexico (to name other countries with long-term one-party rule), is not really a democracy.

But: give a reasonably bright college sophomore a bit of Aristotle to read, and they figure out that last point pretty quickly. If there were more education and public discussion of alternative political systems, especially pointing out how different systems enhance or reduce representativeness and waste a smaller or greater number of votes, people might be much more open to a German-style system. Foreign-born residents like us may have some helpful role to play there.

BTW I have an essay in press (albeit not in a mass-market publication) that proposes a different way of re-purposing the upper house: to represent age-based constituencies. Current Japanese politics is tilted towards the elderly and away from the less numerous young —- with many burdens being placed on the latter. So how about allowing equal numbers of representatives for a small number of age-bands, e.g. 18-30 year-olds, 31-60 year-olds and 61 and older. The representatives themselves wouldn’t have to come from the same age-group; the special concern here would be to allow the young to have some more experienced advocates. As an afterthought to my essay, supermajorities should of course require more than 2/3 votes, such as 70%. I could be mistaken, but as I read the Kenpou, a system like this could be put in place through modifying the election law alone. It might be desirable to give this new Sangiin something closer to the absolute veto power of the US Senate, at least on certain matters, though that would require constitutional change. Should precedents be important, I’m not aware of any in modern politics, but there was some age-based differentiation in assemblies during the Roman Republic.