Politics, No.3  Nov. 27, 2010


One of the most striking features of Japanese politics in recent years has been the succession of short-lived governments. Prime Minister Abe Shinzō took office in September 2006 and fell from power just a year later. This was followed by two more short-lived governments, led by Fukuda Yasuo and Asō Tarō. Both collapsed after barely a year in office. The trend has continued since the long-dominant Liberal Democratic Party fell from power and the Democratic Party of Japan took the reins in September 2009–the government of Hatoyama Yukio, the first DPJ prime minister, lasted just eight months, and the current administration under Kan Naoto already finds itself in a difficult position just three months after coming to office.

Not too long ago, however, Japan had a government that remained in office for five years and five months, under Prime Minister Koizumi Jun’ichirō (April 2001-September 2006). During this time, the government pushed through a series of bold policy reforms in areas long thought to be too politically sensitive to tackle, most notably the privatization of the postal services. These radical reforms were backed up by a clear change in political methodology, in which national policy was placed directly under the control of the prime minister and the Kantei, or prime minister’s office. Not everyone agreed with the Koizumi government’s policies or the political methods it employed, but insiders and observers alike were forced to admit that a major change had taken place.

Opinion differed, however, on the lasting impact of these changes. How permanent were the changes the Koizumi administration had wrought? Some commentators believed the changes should be regarded as a passing phenomenon due largely to the personal leadership of Koizumi himself, rather than a fundamental change in Japanese politics. Others claimed that Koizumi had merely taken advantage of changes adopted during the 1990s, including the overhaul of the electoral system and the reorganization of the central government, which provided a stronger role for the cabinet. According to this view, the situation would continue fundamentally unchanged even after Koizumi left office, now that a strong systemic foundation was in place. I shared this view myself at the time. (See my “‘Tsuyoi shushō’ wa nichijō to naru” [A ‘Strong Prime Minister’ Will Be Nothing Out of the Ordinary], Chūō Kōron, October 2006.)

Today, the relative merits of these two positions (we might call them the “individualist” and “institutionalist” schools of thought) would seem on the face of things to be quite clear. During its brief time in power, the Abe administration was widely ridiculed for bringing about the “collapse of the prime minister’s office,” while under the Hatoyama government, the priority given to the DPJ leadership over the cabinet was clear for all to see. It would appear that the individualists were right, and that the personal qualities of the leader are indeed more important than the structure of the political system in terms of determining how much authority a prime minister can wield. Japanese politics has not experienced the kind of sweeping changes the institutionalists predicted during the Koizumi years. This is probably the majority view at present.

But does this mean that the institutionalist argument is without merit in considering the role of the prime minister and cabinet in Japanese politics? Is it really true that Japan is unlikely to see another strong prime minister until someone like Koizumi happens to emerge with the right combination of character and leadership qualities? Certainly there was a good deal of unjustified optimism in the view of those who argued toward the end of Koizumi’s time in office that the system had changed fundamentally and irrevocably. But this does not mean we should dismiss the institutionalist argument entirely.

In this essay, I want to consider the conditions that might see another strong prime minister emerge in Japan, reexamining the argument from the institutionalist perspective. Preliminary to that, let us look back on the characteristics of the debate that took place during the Koizumi era and its main points of controversy.

The Institutionalist View

Essentially, the institutionalist argument that saw Koizumi as the first in a line of strong prime ministers rested on two main points.

The first was the idea that the new electoral system introduced in 1993 for elections to the House of Representatives had changed the behavior of political parties and Diet members. The new system involved a combination of single-seat constituencies and proportional representation. But members elected from single-seat constituencies made up a majority of the house, with only a limited number of candidates standing for election by proportional representation alone. In effect, this made the new system resemble a “first-past-the-post” system.

As a result, at the same time as strengthening the two-party system in the House of Representatives, there was a dramatic expansion of the influence of the party leaderships within the parties themselves, thanks to their ability to endorse candidates and allocate political funding. This was illustrated following the “postal dissolution” of the lower house in 2005, when, at Prime Minister (and LDP President) Koizumi’s behest, the LDP withdrew its endorsement from candidates who had opposed the government’s legislation to privatize the postal services and put up “assassins” to stand against them. In the case of the ruling party, the party leader is also the prime minister. It was this that led many people to think that the increased ability of the party leadership to control its Diet members would strengthen the position of the prime minister.

Another piece of evidence for the argument was the increased role given to the cabinet as part of the overhaul of the central government organs approved during the administration of Hashimoto Ryūtarō (1996-98) and implemented in January 2001, shortly before Koizumi took office. Throughout the postwar period, it had often been pointed out that the cabinet and especially the prime minister lacked sufficient appropriate advisory bodies under their direct supervision, and were therefore frequently subject to the dictates of bureaucrats when trying to turn policy priorities into legislation. This made radical changes in government policy all but impossible.

The Hashimoto reforms aimed to strengthen the cabinet by augmenting the power of the Cabinet Office and Cabinet Secretariat, which were made to operate as advisory bodies under the direct supervision of the prime minister and chief cabinet secretary. Combined with the skillful use of appointments to political posts, this made it possible to implement policy based on a core staff with a strong sense of loyalty to the prime minister. Koizumi brought in Keiō University professor Takenaka Heizō to head the newly created Council on Economic and Fiscal Policy, which was responsible for drawing up economic policies centered on structural reform and became a driving force of policies based on a coherent and consistent strategy.

The effects of these electoral and administrative reforms adopted in the 1990s weakened the influence of the special-interest legislators and bureaucrats who had dominated previous LDP governments and made it possible to shift the focus of government legislation away from previously powerful vested-interest groups, bureaucrats, industry representatives, and regional politicians. In their place came a new style of top-down policymaking led by the prime minister with the support of the advisory bodies under his direct supervision. Since these traditional networks had been a major factor behind the LDP’s long hold on power, Koizumi’s vow to “destroy the LDP” in the run-up to the LDP presidential elections in 2001 seemed to be no idle threat. And since these reforms involved irreversible changes to the political system, many commentators (myself included) believed that the new status quo would survive more or less intact even after a change of prime ministers, and that a strong prime ministerial presence would continue to be a factor in the post-Koizumi era.

The Trap the Institutionalists Failed to See

Two factors that were overlooked (or at least underestimated) by advocates of the institutionalist argument have turned the post-Koizumi era into something quite different from what was predicted.

One factor has concerned the development and training of new political talent within the parties. In a two-party system based largely on single-seat constituencies, the leaders of the two main parties are in effect running as candidates for prime minister, and they need to be able to wield the increased powers now held by the prime minister immediately if their party gains control of the government. The LDP, however, was unable to find a suitable new framework to replace its traditional approach, which had been based on jockeying for power among the factions within the party. It ended up conducting repeated leadership elections based on public opinion, which is not merely highly unstable but also notoriously difficult to measure accurately. The DPJ suffered from similar problems. The DPJ became the leading party in the National Diet before it had managed to develop a system that gave sufficient power to its representatives, with the party’s founding members rotating the important posts among themselves in a manner reminiscent of the genrō (elder statesmen) of the Meiji era (1868-1912). This was one factor that led to confusion.

Making it easier for the prime minister to ensure that his own intentions were reflected in policymaking and in the way the government was run meant that numerous political decisions were now concentrated in his hands. Clearly, this strengthened the authority of the prime minister. But it also put a considerable burden of responsibility on his shoulders. If a political leader and his party are not strong enough to stand up to the increased pressures and responsibilities inherent in the position of prime minister under the new system, the administration is destined to be short-lived. Classic examples of this can be seen in the cases of Abe Shinzō, who cited poor health when he resigned following the LDP’s defeat in the summer 2007 election for the House of Councillors, and Hatoyama Yukio, who was stymied by a lack of clear priorities and stumbled through a succession of gaffes and misjudgments before he too resigned.

Another factor was the House of Councillors. The members of this upper house are elected by a mixture of three different voting systems: single-seat races in some prefectural constituencies, multiple-seat races in others, and a further cohort of members chosen by nationwide voting according to an open-list proportional representation system. Each of these requires a different approach from both politicians and voters. This means that the House of Councillors operates according to a quite different logic from the House of Representatives. Add to this the house’s definition of itself as a “seat of good sense” and its virtual parity of power with the House of Representatives, and it is easy to see why control of the upper house has been a major issue virtually throughout the postwar era. Even Koizumi was careful to give due consideration to the views of leading LDP representatives in the House of Councillors, particularly those of the LDP’s upper house leader Aoki Mikio. (See Takenaka Harukata, Sangiin to wa nani ka [What Is the House of Councillors?], Tokyo: Chūō Kōron Shinsha, 2010.) When Abe’s government lost its majority in the upper house in 2007, leading to a “divided government” in which control of the two houses was split, the House of Councillors became an even bigger headache for the prime minister.

Koizumi was able to control the upper house only thanks to the deft use he made of Aoki and others, and by daring moves such as dissolving the House of Representatives when the upper house voted against his post office privatization plans. Commentators who believed the system had changed fundamentally perhaps failed to give sufficient consideration to the veto powers of the House of Councillors. Also, the development of a two-party system in the House of Representatives means that there is now little room for compromise in the case of a divided government–another respect in which the predictions of the institutionalist commentators have proved to be off the mark.

Strong Leader, Strong Government?

As we have seen, there is little definite evidence to support the position of those who argued that systemic changes were responsible for the strength of the prime minister during Koizumi’s time in office, and that a strong prime ministerial presence would continue to be a feature of the post-Koizumi era. There is no doubt that the prime minister exerted strong leadership during Koizumi’s time in office, but the individualists’ view that this was due more to Koizumi’s personal qualities as a politician than to anything else now seems persuasive.

Certainly it is reasonable to assume that if two political leaders operating under largely the same systems and conditions achieve radically different results, at least part of that difference must be due to divergent levels of political ability in the individuals involved. There is no shortage of evidence to support this view, and not just in the post-Koizumi era. Over the decades since the end of World War II Japan has had both long-lived administrations, such as those led by Yoshida Shigeru (1946-47 and 1948-54), Ikeda Hayato (1960-64), and Satō Eisaku (1964-72), and weak and relatively short-lived governments, such as those headed by Suzuki Zenkō (1980-82) and Kaifu Toshiki (1989-91).

Even so, I am not convinced that the individualist argument, which focuses exclusively on the political abilities of the leader, is in itself sufficient to explain what is happening. In particular, I think it is important not to lose sight of the following two points.

First, people’s concept of what being a powerful leader entails varies according to the system and conditions in place at a given time. During the years of the LDP’s long monopoly on power, for example, a leader was well on his way to success if he could secure abundant political funding and increase the size of his own faction, at the same time keeping a close eye on power relations between the various factions inside the party and nipping any conflict between them in the bud. This pattern of behavior was well suited to the multiple-member constituency system in place at the time, when there was almost no need for the prime minister to worry about the prospect of losing power to another party but plenty of need to keep tabs on competition between the various factions of his own party.

Perhaps the best exemplar of this tendency was Takeshita Noboru (1987-89). He too brought about major changes in political policy, pushing through a series of major reforms of the tax system, including the introduction of a consumption tax. These changes were introduced following a careful, thorough consideration of the situation that involved the ruling party, relevant government ministries, and business and industry. It was a process quite different from that employed by the Koizumi government. Similarly, Satō Eisaku managed to stay in power for so long thanks largely to his ability to keep the members of his party under control through skilled use of his appointment powers. This too was quite different from Koizumi’s approach. There is no doubt that both Satō and Takeshita were outstanding political leaders–but we also need to consider the system and conditions in place at the time that allowed them to exercise their talents.

The circumstances behind a prime minister’s resignation are another factor to bear in mind. In this respect, the leaders of the recent short-lived governments have been quite different from their predecessors. In the era of rivalry among the LDP factions, a prime minister would often face calls to resign from rival factions inside the party following major losses of parliamentary seats in an election or low levels of public support. This led to what were criticized at the time as “phony” shuffles of government that did little more than present the same material in a new wrapping. There is no doubt that this contributed to the longevity of LDP rule.

In order to protest openly against the party leadership today, however, the disgruntled party member needs to be ready to pay the price in terms of party endorsements and allocation of political funding. This means that criticism of the party from within generally tends to be weaker than it used to be. A case in point is the criticism of Ozawa Ichirō following allegations of improprieties involving political funding during the first half of 2010, which remained limited to parts of the party’s local organizations. Because of this, a leader is often able to limp on beyond his natural expiry date, postponing his resignation long after the point when his government has clearly lost all momentum. This was the case with Abe in 2007 and with Hatoyama earlier this year. One result of strengthening the power of the executive branch is that the prime minister now has a greater degree of autonomy when it comes to choosing the moment of his own demise.

Probability, Not Certainty

I hope the preceding discussion has made clear the influence that systemic conditions can have on a prime minister’s leadership. The prime ministers since Koizumi have clearly been operating under conditions quite different from those their predecessors faced. As the result of reforms that have changed the electoral system and strengthened the cabinet, they now enjoy powers both as party leader and prime minister that were only incompletely secured by previous systems. That so many prime ministers since Koizumi have proved incapable of using these powers is due to limiting factors in areas that were not affected by the reforms of the 1990s, such as the development of talent within the parties and the presence of the House of Councillors. In other words, it is not merely possible to discuss the question of prime ministerial leadership from the institutionalist perspective; it is necessary to do so.

It cannot be denied, however, that many proponents of this argument (myself included) tended to overplay the “strong prime minister” motif in the past. That the prime minister now operates under a system and conditions quite different from those that prevailed in the past has almost certainly increased the probability that policymaking and management of the government or ruling party will proceed in accordance with the prime minister’s intentions. But this is a probability, not a certainty. The mere fact that the prime minister is now supplied with a more promising system and more promising conditions does not by itself necessarily guarantee strong leadership.

Likewise, merely providing the systemic conditions for a prime minister to exert strong leadership does not imply successful policies. As I mentioned above, there is a greater possibility today that the prime minister will be able to transform his opinions and decisions into political policies. But the decisions themselves are not produced by the system itself: They depend entirely on the political wisdom and insight of the prime minister and his advisors. Improving the way that up-and-coming politicians are trained within their parties may increase the possibility that a candidate capable of making the right decisions will emerge in the future–but of course this too is nothing more than a probability.

Nevertheless, considering the conditions under which the prime minister is likely to exert strong leadership is not altogether meaningless. Perhaps the chief weakness of the individualist approach is that it necessarily deduces the qualities of strong political leaders a posteriori by looking at examples from the past. The problem with this approach is it is impossible to know how relevant this information is to the present time, or how relevant it will be in the future.

In the early days after Koizumi took office, not many commentators predicted a bright future for his administration–Koizumi was much more of a “lone wolf” politician than his predecessors. Following the “postal dissolution” of the House of Representatives in 2005, of course, no one was left in any doubt about his leadership qualities–but it was the institutionalists who were the first to point out Koizumi’s potential as a strong leader in terms of the way he managed his own party. The greatest significance of the institutionalist argument may therefore be this: It is able to make predictions and forecasts based on a logical position when a new type of leader emerges, even if these predictions must necessarily be probabilities rather than certainties.

The Possibility of a Strong Prime Minister

Ultimately, a strong prime minister is one who can make use of the systemic conditions in place to reach appropriate decisions in the areas of policymaking and government management. This may seem a rather obvious, even mundane, thing to say. However, we have no alternative but to take this as our point of departure in considering what form prime ministerial leadership might take. In the remainder of this article, I would like to think in more concrete terms about the conditions that might see a strong prime minister emerge in Japan in the future.

First of all, in order to take advantage of the systemic conditions already in place, a prime minister needs to make full use of his authority both as leader of his own party and as prime minister. In recent years, and particularly since the DPJ came to power, there has been so much emphasis put on “Kantei-led politics” and “politician-led policymaking” that people have all but forgotten about the authority the prime minister ought to wield as head of his own party. Under the Hatoyama administration, the party leader’s authority was for all practical purposes delegated to Ozawa Ichirō in his position as DPJ secretary general, with the party apparently ready to swallow whole any decisions he made. But within a parliamentary cabinet system, the prime minister’s authority stems at least in part from his position as the leader of the majority party. It is not surprising that a prime minister who renounces this authority should struggle to steer the rudder of government.

There is nothing wrong with appointing the ruling party’s secretary general or the chairman of its Policy Research Committee to an important post, if the prime minister decides this is necessary in order to keep a firm grip on the party. Along with the cabinet, the party leaders are there to assist and advise the prime minister. Together, these two groups form the core executive. It is wrong to think of the prime minister’s office and the ruling party as being mutually opposed when it comes to exerting leadership. The DPJ did away with its Policy Research Committee for a time–but a political party with no forum for debating policy is a strange thing indeed.

What is important is that the wishes of the prime minister as party leader should be clearly reflected during the decision-making process within the governing party. If the prime minister sets the basic direction on policy and legislation is proposed that accurately reflects these views, then it should be possible for concrete ideas to emerge from the leading party during this process.

It is likely that serious consideration will need to be given to the state of the House of Councillors, including major systemic reforms. I have already discussed this subject elsewhere, so I will not go into details here. Suffice it to say that reforming the House of Councillors does not simply mean strengthening the power of the prime minister. Instead, the aim of any reforms should be to protect the integrity of the bicameral system and to make the House of Councillors a house that accurately reflects the views of minority interests and ideas. (See my “San’in no ichizuke, shōten ni” [Focusing on the Status of the House of Councillors], Nikkei, July 16, 2010, p. 31.)

More difficult is the question of how to develop the decision-making ability of up-and-coming politicians. The status quo has long been criticized for its failings, and there has been no shortage of politicians with the potential to become leaders who have been enfeebled by a lack of cultivation and wider accomplishments beyond politics. These arguments have become stronger recently. There is also something to the argument that insight and discernment are more important than English-language skills or Internet savvy when it comes to evaluating our potential leaders.

It should not be forgotten, however, that voters and the media play a large part in deciding what is demanded of politicians in any given age. And it is voters and the media who have recently come to value language skills, Internet literacy, and an affable manner during regular interactions with journalists, dismissing as unfit for leadership any politician who happens to be lacking in these areas. The first step must be to reevaluate these criteria.

At the same time, we need to improve the ability of aspiring young politicians to make decisions based on substance and conviction. This will include making it easier for them to get the advice they need, when they need it. What aspects of Japan’s politicians today are excessive, and where are they lacking, looked at historically and in comparison with politicians in other countries? What do we need to do to improve the situation? What kind of environment should we aim to provide, and who should serve as advisors? The individualist argument will make its presence felt strongly in this context.

The recent succession of short-lived governments and the declining confidence in Japanese politics both domestically and internationally are not due entirely to the inadequacies of Japan’s politicians and political parties. They also reflect a decline in the ability to consider and debate politics and the ability of voters to take these arguments on board. Above all, the failure to adequately resolve the Koizumi-era debate about the advent of a “strong prime minister” has become a major weakness. Now is the time for the advocates of both positions to engage in a constructive dialogue to produce a breakthrough once and for all.

Translated from “‘Tsuyoi shushō’ no jidai wa sairai suru no ka,” Chūō Kōron, October 2010, pp. 178-85. (Courtesy of Chūō Kōron Shinsha) [November 2010]