Recherche FFJ Research Statement J.A.A. STOCKWIN


Stability and Instability in Japanese Politics: Explaining Prime Ministerial Turnover

University of Oxford


For some years now, it has been common to analyse Japanese politics in terms of relative instability. I say “relative” because plainly there are important countries whose politics are far more unstable than the politics of Japan, but nevertheless in Japan there are significant indicators of instability that have attracted widespread interest and concern, both in Japan and elsewhere. The most striking of these is the rapid turnover of prime ministers that has been occurring. Between 2006 and 2011 Koizumi, Abe, Fukuda, Asō, Hatoyama and now Kan have successively occupied the prime minister's residence. That is six prime ministers in five years, giving an average tenure of less than one year. Indeed, since Koizumi stepped down in September 2006, not one of his successors has yet managed to stay in post for as much as a year. Over the same period there have been three British prime ministers, two French presidents, and two German Chancellors.
There are some other indicators of political instability as well. Prime ministers come to office with relatively high approval ratings, usually in the 50 to 60 per cent range, but these rapidly fall, so that towards the end of their tenure public opinion polls showed an approval rating of 26 % for Abe, 19% for Fukuda, 13% for Asō, 17% for Hatoyama,[1] and about 20 % currently for Kan.
Another indication is that voting patterns, which used to be remarkably stable and predictable, have become exceedingly changeable and difficult to forecast. Huge blocs of voters seem to change their votes at each election, mainly against the government in power at the time. The LDP was elected in a landslide in the House of Representatives in 2005, the DPJ almost exactly reversed that position in its favour in 2009, and if elections were held today it seems quite possible that the position would be reversed once again. Since the 1990s those declaring no party allegiance to public opinion pollsters have been a majority of electors, and there is anecdotal evidence that many electors have a low opinion of politicians in general, while perceiving that prime ministers are endemically weak.
Again, the mass media are relentlessly critical of politicians, they fill their pages with stories of money scandals and faction fighting within parties, and give little credit to politicians for dealing with the difficult policy choices they have to make. This no doubt fuels the general mood of scepticism that pervades the electorate.
Is political instability something new, or has it been there all the time?
Before advancing some hypotheses to explain political instability, I need to examine whether it is really a new phenomenon, or whether it has not been a pervasive feature of Japanese politics for many years. Kan Naoto is the 33rd Japanese prime minister since August 1945, whereas David Cameron is the 13th British prime minister over the same period. In Australia, Julia Gillard is the 12th prime minister since August 1945. It might seem to follow, therefore, that instability of prime ministerial tenure in Japan is nothing new. I argue against this, however, since some periods have seen remarkable political stability, This is not entirely a question of the length of time a prime minister stays in office, but it is worth noting that four of them have stayed in office for more than five years, and in one case (Satō Eisaku between 1964 and 1972) for over seven years.
Possible explanations for political instability
In this section, I want to examine five alternative hypotheses that might explain political instability in the recent period. They range from the very general to the very specific. I will critically examine each of them in my concluding section, when I shall also add a sixth hypothesis which I believe to be more persuasive than the others, though it does not necessarily negate their approaches.
1. Cultural factors
A feature of Japanese political parties that is often remarked is their tendency to divide into discrete and often well organised factions. In my own early work on the Japan Socialist Party I emphasised this aspect and examined it in some detail. It may well be argued that cultural factors underlie intra-party factionalism in the sense that Japanese society has a strong sense of in-group and out-group differences, which manifest themselves in all pervading factionalism, making it difficult for parties to remain united. Those emphasising this aspect today tend to be influenced by the work of the cultural anthropologist Nakane Chie.
Cultural factors undoubtedly underlie the ways in which politics is conducted in any country, including Japan. The particular aspect that I emphasised, namely factionalism, is important in the DPJ, as it was in governments formed over many years by the LDP. In particular the group of supporters of Ozawa often operate as a discrete group, and have caused problems for Kan and his Government. But this needs to be kept in perspective: Factions were much more cohesive under the earlier lower house electoral system based on multi-member districts, which forced candidates from the same party to campaign against each other. Today they are looser and more undisciplined. Moreover, Ozawa is widely regarded as a practitioner of 'old politics' in an era of political modernisation.
2. The change of government in September 2009
The sweeping victory of the Democratic Party on 30th August 2009 ended the long-established political structure centred on the LDP, government bureaucracy and key interest groups, and as such ushered in a new kind of politics having the potential to revolutionise the political system. Unfortunately, the new government did not prove itself particularly competent or capable of reforming the system in a rational or sustainable manner. Moreover, whereas the new government was determined to reduce the power of unelected bureaucrats, its early attempts to do so weakened its base without providing an effective alternative structure. In so far as the revised electoral system for the House of Representatives contributed to the DPJ victory in 2009 (15 years after its introduction), that too may be regarded as having contributed to political instability. This hypothesis, in other words, is based on the idea that the continuation of single-party dominance was the sine qua non for political stability.
However, the hypothesis that it was the change of government in 2009 that brought about political instability does not stand up to serious scrutiny. The old structure had been breaking down over a long period and by 2009 it was time for a change of government and perhaps also for a change of system. After all, Koizumi, the most successful recent LDP Prime Minister, said that he wanted to 'smash' the LDP. A new party in power should be able to renovate the system and give Japan a different sort of leadership. The failures of the DPJ Government may well be down to inexperience, and this suggests that without regular changes of government opposition parties inevitably lack the experience of governing.
3. “Nothing much has changed with the change of party in power”
Hypothesis No. 3 is the opposite of hypothesis No. 2 in the sense that it regards the alleged failure of change – not change itself – as the principal source of instability. According to this view, the DPJ is essentially another version of the LDP – admittedly a little to the left of it ideologically – and lacks a coherent alternative tradition. Hatoyama Yukio, the original Prime Minister of the new government, fitted all too neatly into the time-honoured LDP tradition of dynastic politicians, being the great-grandson of a pre-war politician, grandson of a former prime minister, son of a former LDP foreign minister and brother of a former LDP minister of justice. Not long after the Government was formed, two of its leading members, Hatoyama, the Prime Minister, and Ozawa Ichirō, the party's eminence grise, were embroiled in allegations of illegal financial transactions reminiscent of what used to happen under the LDP. Under the Kan Government also, at least three key ministers have been forced to resign for reasons that would have been familiar when the LDP was in power. Moreover, attempts under Hatoyama to branch out in new directions, especially in foreign policy and social welfare policy, had little success.
This hypothesis is not without substance, but despite appearances, the DPJ Government since late 2009 has worked hard to reduce wasteful expenditure, confront government and private interests, forcing them to trim back their favoured projects, and reduce the extent to which former government servants can find lucrative employment in government spin-offs and the private sector. The phenomenon of 'tribal politicians', who used to dominate particular policy areas, has been drastically scaled back. Since the change of government, many previously secret documents concerning Japan-US relations in particular have been published. Financial constraints have forced DPJ leaders to be less ambitious in providing social welfare than their 2009 manifesto promised, but progress in this field has not been entirely absent.
4. Absence of real leadership quality among Koizumi's successors
Hypothesis No. 4 puts the emphasis on the quality of leadership exhibited by successive prime ministers since Koizumi stepped down in September 2006. According to this, none of them have measured up to the very real leadership qualities exhibited by Koizumi himself, who undertook the cleansing of the banking system, gravely debilitated under a burden of non-performing debt, and presided over a reviving economy. In contrast to Koizumi, Abe placed primary emphasis on Constitutional revision, revision of the Fundamental Law on Education to emphasise patriotism and upgrading of the Defence Agency to the status of a ministry, but neglected the urgent economic tasks that his predecessor had pursued effectively. Fukuda, Abe's successor, essentially reacted to events, Asō was manifestly incompetent and Hatoyama eccentric. In other words, the system has not produced an outstanding leader since Koizumi, and this is the main source of the instability that plagues Japanese politics.
This hypothesis stands up reasonably well when we consider the immediate successors to Koizumi, namely Abe, Fukuda and Asō, different though they were from each other. But these were prime ministers in the old LDP style, essentially keeping the existing system turning over. When, however, we come to the DPJ prime ministers, Hatoyama and Kan, we need to emphasise the different kind of project that they were engaged in, namely to pioneer a different kind of politics, tackling head on some of the long-standing practices of the old government structure. When we recognise this, it seems that their performance has been rather better than it is often portrayed.
5. The House of Councillors is too powerful
Hypothesis No. 5 focuses on the constitutional powers of the upper house, the House of Councillors, which can veto bills reaching it from the House of Representatives so long as they are not the budget, a treaty or designation of a prime minister. Such a veto may be overruled by a two thirds majority of the lower house, but it unusual for governments to possess such an overwhelming majority. Where the majority in the House of Councillors has a different composition from that in the House of Representatives, this may result in what is now known as a nejire kokkai ('twisted Parliament'). There have been two recent instances of the nejire kokkai, the first between July 2007 and September 2009, when elections for half the upper house seats gave the DPJ and its allies an upper house majority, and the second in 2010, when the same thing happened but the other way round, making things difficult for the DPJ-led Government. This has forced governments to cut deals with other parties in order to get their legislation through. Even though the annual budget is constitutionally exempted from the two-thirds override rule, what are termed 'budget related bills' are not, and this creates a serious headache for governments in this situation.
There is much to say in favour of this hypothesis as an explanation of political instability. Both LDP governments between July 2007 and September 2009, and the DPJ Government led by Kan from July 2010, have been seriously obstructed in realising their legislative programmes by the ability of the House of Councillors to reject ordinary legislation sent to it by the House of Representatives. This is in effect an invitation to the opposition to reject legislation for the sake of scoring political and electoral points, rather than to engage in careful scrutiny of legislation.
At this point I want to introduce a sixth hypothesis, both broader in scope and in some ways more persuasive, than the previous five.
By way of a conclusion: hypothesis No. 6: Political instability is greater in periods of economic difficulty and less in periods of political prosperity
When we survey the history of Japanese politics since 1945, we find an interesting correlation between economic stability and progress, on the one hand, and political stability, on the other. The correlation is not perfect, but it is highly suggestive. Conversely, periods in which there has been economic stagnation also tend to be periods in which politics shows signs of instability. When we focus on length of prime ministerial tenure as an indicator of political stability or instability, we find that the longest-serving prime ministers were Yoshida Shigeru (1946-7 and 1949-54), Satō Eisaku (1964-1872), Nakasone Yasuhiro (1982-7), Koizumi Junichirō (2001-2006). Leaving aside Yoshida for a moment, Satō took the helm during the very rapid economic growth period of the 1960s, and stepped down shortly before the first oil crisis, which seriously damaged Japanese economic performance. Nakasone was Prime Minister during the second great spurt in economic growth, that of the 1980s, and Koizumi was in office at a time when the stagnant economy appeared to be returning to an upward path. Yoshida's case, admittedly, was somewhat different. His first term in office was in the chaotic period of the early Occupation, whereas his second, much longer term, took in the consolidation of the Occupation reforms, the San Francisco Peace Treaty and the return to national independence. These were times of rapid change and economic recovery, but could hardly be designated as a period of political stability. To some extent, Yoshida's position was sustained by the Occupation, and once the Occupation ended in April 1952, his power gradually slipped away.
Conversely, most of the 1950s, the 1970s, the period from the end of the 1980s to the end of the 1990s, and the period from 2006 to the present, have been times, both of economic uncertainty or stagnation, and times when prime ministers came and went like flowers after rain in the desert.
There appears to be a strong causal element in these correlations. In Japan as in other democratic countries, the electorate rewards governments when things are going well and punishes them when things are going badly. The mass media, with one eye on the electorate, are particularly tough on governments whose policies do not work out well, even when the failure of policies is brought about by difficult economic conditions rather than by lack of effort or perspicacity on the part of governments. Opposition parties also, quite naturally in terms of their aim of replacing the ruling party or parties in office, readily take advantage of a government's difficulties to disrupt its performance. Where institutions facilitate this, as with the excessive and disruptive powers of the House of Councillors, the government is further weakened and instability ensues.
In the case of the Kan Government, his party, whose instincts were progressive and reformist, had to preach the virtues of financial retrenchment forced on it by a stagnant economy and over-indebted financial system. As a result it rapidly lost the support of the mass of electors who had voted the DPJ into power in 2009. It is evident, however, that the Kan Government since the disaster of 11th March 2011 has moved into a different category from either 'stable' or 'unstable'. Rather, it has moved into a condition analogous to that of a war cabinet, in which confrontational politics is removed or drastically reduced and the business of government is placed on an emergency footing. How long this will continue, and whether it will eventually evolve into a stable long-term administration or an administration with a brief and chaotic life, is a matter for speculation and thus beyond the scope of this paper.
PS: This research note is a shortened version of the lecture that Professor Stockwin gave in Paris for Fondation France-Japon on May 3rd, 2011.
Reference: Stockwin Arthur (2008), Governing Japan: Divided Politics in a Resurgent Economy, Wiley-Blackwell, 4th Revised edition.

[1]Asahi, 21 February 2011.