Jump to ContentJump to Main Navigation
Growing Democracy in JapanThe Parliamentary Cabinet System since 1868$

Brian Woodall

Print publication date: 2014

Print ISBN-13: 9780813145013

Published to Kentucky Scholarship Online: September 2014

DOI: 10.5810/kentucky/9780813145013.001.0001

Show Summary Details
Page of

PRINTED FROM KENTUCKY SCHOLARSHIP ONLINE (www.kentucky.universitypressscholarship.com). (c) Copyright The University Press of Kentucky, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in KSO for personal use. Subscriber: null; date: 09 April 2020



(p.1) Introduction
Growing Democracy in Japan

Brian Woodall

University Press of Kentucky

Abstract and Keywords

This chapter surveys the analytical topography and introduces the conceptual framework that informs this study. Concepts from institutionalization theory and historical institutionalism are fused to create a unique lens through which to assess the process whereby Japan’s cabinet system evolved and the factors that molded its distinctive form and functions. It is posited that Japan’s cabinet system was transformed at seven critical junctures. In prewar times, significant changes followed the 1868 Meiji Restoration, the advent of party-led cabinets, and their violent demise with the “May 15th Incident.” In postwar times, American occupation planners orchestrated a dramatic reconfiguration, while significant changes followed the emergence of the “1955 system” that ushered in the Liberal Democratic Party’s protracted rule, the “shocks” of the early 1970s, the advent of coalition cabinets in 1993, and the emergence of “Twisted Diets” following the 2007 upper house elections. Consequently, Japan’s present system is the product of a developmental process that has resulted in the inability to establish a properly functioning system of cabinet government. A central lesson gleaned from this analysis is that growing democracy is not easy, and, in this regard, the Japanese case offers crucial lessons for understanding the challenges and disappointments that confront today’s developing countries.

Keywords:   Japanese politics, March 2011 disasters, cabinets, cabinet government, parliamentary system, Westminster model, comparative politics, institutions, institutionalization, historical institutionalism

After being in Government one realizes that Parliament has a somewhat more peripheral role to perform…. Parliament is, in fact, not the centre of government. The centre of Government is the Cabinet.

—British cabinet minister quoted in Donald D. Searing, Westminsters World (1994), 329 (italics added)

I must have attended nearly ninety Cabinet meetings. They lasted an average of ten minutes and all I did was sign documents.

—Former health minister and future prime minister Kan Naoto, quoted in Aurelia George Mulgan,

Japan’s Failed Revolution (2003a), 156

System Failure

On the afternoon of March 11, 2011, an apprehensive nation looked to Prime Minister Kan Naoto and his cabinet for leadership and reassurance in the aftermath of a series of cascading disasters. Unleashed by the most powerful temblor ever to hit the quake-prone country, the catastrophe began with the Eastern Japan Great Earthquake, one of the five strongest in recorded history. Hundreds of aftershocks followed, and millions of households and businesses were left without electrical power. The earthquake produced an enormous tsunami that propelled a lethal wall of water ten kilometers inland. Nearly twenty thousand people died and more than three hundred thousand were displaced. Hundreds of thousands of homes, buildings, roads, bridges, railway lines, and other elements of critical infrastructure were destroyed or damaged. In addition, the earthquake and tsunami disabled the cooling systems at the Fukushima Daiichi Nuclear Power Plant, resulting in hydrogen explosions and visible damage to three reactors. These natural and man-made disasters produced a vast, heart-wrenching humanitarian crisis that was painstakingly reported by the domestic and international media. The survivors desperately needed shelter, drinking water and food, bathroom facilities, and medical care. And, with a late winter storm approaching, no time could be wasted before (p.2) launching a large-scale, coordinated rescue and recovery effort. Government spokespeople called on citizens to remain calm and downplayed reports of a nuclear catastrophe.

But the leadership displayed by Kan and his cabinet was ridiculed by the mass media. One day after the earthquake, Kan assessed the stricken nuclear power plant from a helicopter (evoking criticism that his visit delayed crucial venting at the plant) and reportedly asked officials on the scene, “What the hell is going on?” (Guardian, November 2, 2011). Then, for the next few days Kan and his government’s spokespeople more or less read from a script written by the Tokyo Electric Power Company, the utility company that operates the plant. The Kan government repeatedly assured the nation that the situation was under control, even after the order was issued to evacuate tens of thousands of people living within a twelve-mile radius of the plant. The Kan government was also criticized for its humanitarian response. “It’s been a week, and there’s still been no government help,” said a resident of one of the hardest hit areas, expressing the exasperation of many victims (Los Angeles Times, March 18, 2011). In response, Chief Cabinet Secretary Edano Yukio conceded, “In hindsight, we could have moved a little quicker in assessing the situation and coordinating all that information and provided it faster” (Bloomberg BusinessWeek, March 18, 2011).

The fears of a distrustful public were substantiated when dangerous levels of radioactivity began to appear in milk and vegetables produced in the Fukushima area, and when the media reported that beef cattle fed contaminated hay had been shipped around the country. Then, one month into the crisis, Japan’s nuclear safety agency retroactively raised the event at Fukushima Daiichi from “Level 5” to “Level 7,” putting it on par with the Chernobyl catastrophe (Energy News, April 11, 2011). In the weeks that followed, opinion surveys showed that most citizens wished to do away with nuclear power. Meanwhile, Kan and various cabinet officials repeatedly contradicted themselves by calling for “de-nuclearization” and then asserting that the country could not do without nuclear energy. On August 30, with a sub-20 percent approval rating, Kan and his ministers resigned en masse amid criticism of the inept handling of the multiple disasters.

The first question raised by these events is this: Why did Prime Minister Kan and his cabinet ministers fail to use their powers to galvanize and reassure the nation following the catastrophic sequence of events? Indeed, the (p.3) Kan cabinet’s inept handling of the March 11 triple disaster is surprising, given the fact that the cabinet’s disaster management machinery had been bolstered following the Murayama cabinet’s bungled response to the 1995 Kobe earthquake that claimed nearly sixty-five hundred lives.1 A similar question could be posed concerning the succession of Japanese prime ministers and their cabinets who failed to lead Japan out of the “lost decades” (ushinawareta junen) that followed the bursting of the bubble economy.2 In fact, the Constitution of Japan vests executive power in the cabinet, and even a cursory reading of the historical record reveals that this “supreme law of the nation”—which, ironically, was authored in large measure by Americans who knew little about parliamentary democracy—aimed to establish a parliamentary cabinet system inspired by Britain’s “Westminster” model (Ward 1956, 1000; Stockwin 1999, 38; McNelly 2000, 65–66). Moreover, as we shall see, the executive leaders of the Japanese polity do, in fact, possess powers and prerogatives that are similar to those of their counterparts in other advanced parliamentary democracies.

It is tempting to blame Kan’s leadership for the flawed response to the calamities of March 11, which is exactly what the mass media did. Yet the scope and complexity of the disasters were unprecedented and would have proved a daunting challenge for any leader. The case can be made, therefore, that the flawed response was the result of failure of the cabinet system, not shoddy prime ministerial leadership. This draws attention to two additional puzzles. Why, after more than six and a half decades of parliamentary democracy, has cabinet government failed to set root in Japan? And what forces conspired to give Japan’s cabinet system its characteristic form and function, unflatteringly described as an “un-Westminster” system commanded by prime ministers who are “among the weakest democratic leaders in the world” (George Mulgan 2003a, 140; Krauss and Pekkanen 2010, 280)?

To solve these puzzles, we must trace the development of Japan’s parliamentary cabinet system. In this introductory chapter, I survey the analytical topography and introduce the conceptual framework that informs this study. I draw upon concepts derived from institutionalization theory and historical institutionalism to create a unique lens through which to assess the process by which Japan’s cabinet system evolved and the factors that molded its distinctive form and functions. I contend that Japan’s cabinet system was transformed at eight historical junctures. In prewar times, significant changes followed the 1868 Meiji Restoration, the advent of party-led (p.4) cabinets in 1898, and their violent demise in 1932. In postwar times, the Allied occupation produced a dramatic reconfiguration, while significant changes followed the emergence of the “1955 system” that ushered in the Liberal Democratic Party’s (LDP) protracted hegemony, the “shocks” of the early 1970s, the advent of coalition cabinets in 1993, and the emergence of Twisted Diets following the 2007 House of Councillors elections. The substantive chapters of this book trace the evolution of Japan’s cabinet system and, in so doing, “deal at length with history, and with time in history” (Braudel 1982, 26). Through this discussion, it will become clear that the present system is the product of a developmental process that has resulted in Japan’s inability to establish a properly functioning system of cabinet government.

Organization of the National Executive

The executive branch of the modern democratic polity is organized in one of three broad forms—presidential, semi-presidential, and parliamentary. In a presidential system, legislative and executive powers are separated, while the symbolic and political functions of the executive are fused in a single office. Among the thirty-four advanced industrialized polities that make up the Organization for Economic Cooperation and Development (OECD), the United States, Mexico, the Republic of Korea, and Chile are governed under presidential systems. Ministers in America’s presidential system go by the title of “secretary” and are appointed by the president with the consent of the Senate. But the U.S. cabinet has no formal constitutional role; rather, it is an informal, extra-constitutional advisory organ that is not directly responsible to the legislature. The cabinet is responsible to the president, and, as is the case with cabinets in most presidential systems, it is a relatively weak organ. Indeed, as the authors of a popular U.S. government textbook observe, “unlike in England and many other parliamentary countries, where the cabinet is the government, the American cabinet is not a collective body. It meets but makes no decisions as a group…. The cabinet is made up of directors, but it is not a board of directors” (Lowi and Ginsburg 1998, 242–243). For this reason, meetings of America’s cabinet have been described as “vapid non-events in which there has been a deliberate non-exchange of information as part of a process of mutual non-consultation” (Weisband and Franck, quoted in Cronin 1980, 253). Hence, (p.5) the U.S. cabinet is not a “formally integrated institution,” but “more a collection of a dozen or so secretaries who sometimes meet together in sessions called by the president, and at his whim” (Cohen 1988, 23).

In a semi-presidential system, executive powers are divided between a popularly elected president and a prime minister. The president is endowed with considerable constitutional authority, while the prime minister also wields substantial power and is subject to the confidence of Parliament (Shugart 2005, 324). Semi-presidential systems can be subdivided into two types, premier-presidential and president-parliamentary. The French Fifth Republic (1956 to the present) is the best known and most widely emulated premier-presidential model, but somewhat similar systems also can be found in Portugal, Poland, Lithuania, and the Ukraine. The classic example of a president-parliamentary system was that of Germany’s Weimar Republic (1918 to 1933), and this type of executive system is found in Russia, the Republic of China (Taiwan), Georgia, Armenia, Peru, and Mozambique. Among OECD countries, only France, Portugal, and Poland are governed under semi-presidential systems.

In a parliamentary system, the executive’s symbolic and political functions are carried out by separate officeholders. The symbolic functions are the domain of a “head of state,” in effect a ceremonial figurehead, as is the case with the hereditary monarchs in Britain, Sweden, and elsewhere, as well as with India’s president.3 The political functions are entrusted to a “head of government.” In Britain’s widely emulated “Westminster system,” the head of government, known as the “prime minister,” is the leader of the majority party or coalition in a popularly elected Parliament. This officeholder is selected by the Parliament, not through direct popular elections. Executive power is concentrated in the prime minister and “cabinet,” in essence a committee composed of “ministers” who head up the major departments of public administration. With few exceptions, ministers are recruited from the ranks of elected members of Parliament. The expectation that ministers should be elected MPs helps to ensure a “democratic character” to the cabinet, whose members are “accountable to the people’s elected representatives” (House of Commons 2010, 6). Because a single party or coalition controls Parliament as well as the executive, centralized and disciplined party control under strong prime ministerial leadership becomes the norm. Parliamentary cabinets make decisions as a collectivity, and their members are collectively responsible to the legislature. The prime minister and the cabinet serve at the pleasure of Parliament, which at any (p.6) time can remove them by passing a non-confidence resolution or rejecting a confidence resolution.

All together, twenty-seven of the thirty-four OECD countries can be classified as parliamentary democracies: the United Kingdom, Australia, New Zealand, Canada, Ireland, Belgium, the Netherlands, Luxembourg, Denmark, Norway, Sweden, Finland, Estonia, Iceland, Germany, Austria, Switzerland, Italy, Spain, the Czech Republic, Slovakia, Hungary, Slovenia, Israel, Greece, Turkey, and Japan. Indeed, since 1947 Japan has been governed under a parliamentary system, although the emergence of “party cabinets” in the 1920s and 1930s appeared “to be bringing a relationship between the Diet and the executive which resembles that of the British parliament to the cabinet” (Quigley 1932, 199). Although Japan’s brief prewar drift toward democracy ended in disaster, the postwar polity is founded on the institutions of parliamentary democracy.

Japan’s Cabinet System

Japan boasts all of the classic features of a parliamentary democracy. The head of state is the emperor, who as “the symbol of the State” performs certain ceremonial duties (Constitution of Japan, Article 1). The head of government is the prime minister, who must be an elected member of the Diet, as is also the case with a majority of cabinet ministers (Articles 67 and 68). The prime minister as well as all cabinet ministers must be “civilian” (Article 66). In the event that the cabinet fails to retain the Diet’s confidence, the cabinet must resign en masse unless the House of Representatives is dissolved within ten days (Article 66). The prime minister and cabinet are collectively responsible to the Diet in the exercise of executive power, which has been interpreted to mean that all cabinet decisions should be unanimous (Article 66; Administrative Management Bureau 2007, 117; Tanaka 1976, 41). To secure unanimity, the prime minister is empowered to appoint and dismiss ministers (Article 68). To facilitate the candid expression of views, no official public minutes are kept at cabinet meetings, which only state ministers and three designated government officials—two deputy cabinet secretaries and the director general of the Cabinet Legislation Bureau—are allowed to attend.4

The cabinet is responsible for performing a variety of functions. It administers the law, supervises the government bureaucracy, manages (p.7) foreign affairs, concludes treaties (with subsequent approval of the Diet), prepares and presents the budget to the Diet, enacts cabinet orders, and decides on amnesty, commutation of punishment, reprieve, and restoration of rights (Constitution of Japan, Articles 72 and 73). All laws and cabinet orders must be signed by the relevant ministers and countersigned by the prime minister to signify that they have assumed responsibility for the faithful implementation of those decisions (Article 74). A cabinet minister is the supreme authority within each government ministry and agency, supervising the agency’s officials, submitting bills and government ordinances to the cabinet for approval, and approving all significant departures in agency policy and procedure. The minister is empowered to appoint and dismiss agency officials below a certain rank (Park 1986, 1). The prime minister and cabinet occupy the apex of the central executive, and are tasked with coordinating and securing administrative uniformity among the various organs of government and bearing ultimate responsibility for all executive activities of the state (Administrative Management Bureau 2007, 116–117). The powers and perquisites that go along with ministerial appointment exude an irresistible appeal for Japanese MPs, which has given rise to the term “minister disease” (daijinbyo) to describe the phenomenon (Kuroda and Miyagawa 1990, 6). The celebrity status accorded ministers is attested to by the “tent villages” (tento mura) that the press corps sets up outside the Prime Minister’s Official Residence whenever a new cabinet is formed.5

The official view of the Japanese government is that “the Cabinet, Ministries, Agencies, and public corporations function as one organization, at the top of which exists the Cabinet. … Consequently, it is natural that the agencies and corporations which take care of national administration should be systematically organized under the Cabinet” (Administrative Management Bureau 2007, 116; italics added). While the ministers attract the lion’s share of media attention, these busy leaders could not perform their multifarious functions without the assistance of a large, professional support staff. Indeed, a minister must simultaneously don multiple hats—those of government representative, public agency head, front-bench party leader, and member of Parliament. The Cabinet Office assists with overall administrative coordination and a variety of other concerns, including the handling of official documents. The Cabinet Secretariat arranges cabinet meetings, provides safekeeping for official seals, and supports the chief cabinet secretary in sundry other ways. The Cabinet Legislation Bureau (Naikaku hosei kyoku, or CLB)—staffed entirely by elite career officials seconded by government ministries—offers legal opinions and assists in drafting legislative bills, cabinet orders, and proposed treaties (Asahi Shimbun Globe, June 14, 2010). In addition, the Security Council of Japan and other cabinet-related councils and committees provide for focused policy deliberation. Taken together, these organs—and their mutually supportive and interdependent interactions—constitute Japan’s cabinet system.6


(p.8) Figure I.1. Japan’s Cabinet System (2nd Abe Cabinet, December 26, 2012)

(p.9) The Comparative Perspective

Cross-national comparisons help us understand the extent to which Japan’s premiers and cabinet ministers are adequately equipped to shoulder their weighty administrative burdens. As noted earlier, Japan’s cabinet serves at the pleasure of the Diet, which can topple it with a simple majority vote in support of a non-confidence resolution or by rejecting a confidence resolution. Similar constitutional conventions are in place in Britain, Canada, and Australia. By way of contrast, a majority of German legislators must first elect a successor before the Bundestag can dismiss a chancellor (Kommers 1997, 191; Lancaster 1997, 317). In the Japanese case, non-conidence resolutions have succeeded on only four occasions, most recently in 1993 when the Miyazawa cabinet was ousted, injecting a brief hiatus in the LDP’s legislative hegemony.7 Likewise, Japan is one of many advanced countries that require or permit cabinet ministers to be recruited from the ranks of elected members of Parliament.8 Except for the occasional non-MP appointed to a ministerial post, Japanese cabinet ministers are entitled to hold a seat and to vote in Parliament, as is the case in Britain, the Commonwealth countries, and elsewhere.9 Such is not the case in Holland and Norway, where ministers are obliged to relinquish their parliamentary seats, nor in Sweden, where ministers may hold parliamentary seats but cannot vote (Andeweg and Bakema 1994, 66; Strom 1994, 43; and Larsson 1994, 172).

The Japanese prime minister’s formal powers are comparable to those of heads of government in other established parliamentary democracies. As noted earlier, Japan’s prime minister is empowered to appoint and remove cabinet ministers, which also holds true for the British, Danish, Icelandic, Irish, and Swedish heads of government. In contrast, the Dutch, Finnish, and Norwegian heads of state lack the power to select or sack cabinet ministers unilaterally (Andeweg and Bakema 1994, 59; Nouisiainen 1994, 98–99). Unlike the British prime minister, Japan’s prime minister is not empowered to create departments of state, as this requires Diet legislation. However, as part of the 2001 central government restructuring, the Japanese premier is now able to create ministerial portfolios known as “minister of state for special missions” (tokumei tanto daijin) (Headquarters for the Administrative Reform of the Central Government 2000). Nevertheless, while there has been some evidence of “presidentialization” of the prime minister’s role or the emergence of a “kantei” (the Japanese (p.10) equivalent of the White House) style of diplomacy, few, if any, would argue that the actual influence of the Japanese premier even remotely compares to that exercised by the British prime minister (Krauss and Nyblade 2005, 368; Shinoda 2007; Thayer 1996, 71).

Japan’s cabinets are similar in size to those in other parliamentary democracies.10 The Cabinet Law specifies that the number of ministers of state shall not exceed fourteen, except under special circumstances, in which no more than four additional ministers may be appointed. Including the prime minister, therefore, the maximum number of ministers in a Japanese cabinet is eighteen. In December 2010, the average membership size of the twenty-nine OECD member countries governed under parliamentary systems was just under nineteen members (Central Intelligence Agency 2010). The Israeli and Canadian cabinets were the largest, with thirty-five and thirty ministers, respectively, while the smallest were the Icelandic (twelve ministers) and Hungarian (eleven ministers) cabinets. Meanwhile, the Swiss Federal Council—which is not responsible to the elected assembly, and, therefore, not a parliamentary cabinet—was composed of just seven members. As Klimek and his colleagues discovered, there is a high significance of negative correlation between a cabinet’s size and its performance.11 With a few exceptions, cabinets that number more than around twenty members surpass the critical “Coefficient of Inefficiency,” resulting in poorer performance as measured by the United Nations Development Program’s Human Development Indicator (Klimek et al. 2008).

By law, Japan’s cabinet performs its functions through “cabinet meetings” (kakugi) that are presided over by the prime minister, who sets the agenda for discussion, during which each minister is allowed to pose questions (Cabinet Law, Article 4). Three main types of cabinet meetings have come to exist. The bulk of cabinet-related business is conducted in biweekly “regular cabinet meetings” (teirei kakugi), while “special cabinet meetings” (rinji kakugi) are convened whenever necessary. When an immediate cabinet decision is required, officials of the Cabinet Secretariat are dispatched to secure signatures (kao) from each minister in what are referred to as “revolving cabinet meetings” (mochimawari kakugi), even though face-to-face meetings do not occur.12 Cabinet meetings take place in a special room in the Prime Minister’s Official Residence, with ministers seated around a round wooden table in a predetermined order.13 The prime minister’s seat is considered the center of the table, with the chief cabinet secretary on the right and the holders of senior posts—generally beginning with the foreign (p.11) or finance ministers—seated in alternating right-left order around the table. This arrangement is somewhat similar to that of the British cabinet, whose seating is arranged hierarchically, with the cabinet secretary on the prime minister’s immediate right and holders of senior posts on his or her left or across the table (Kavanaugh 2000, 247).

If time spent in cabinet meetings is any indication, Japan’s ministers are about as busy as their counterparts in other parliamentary democracies. Regular cabinet meetings are generally held each Tuesday and Friday at 10:00 A.M., but when the Diet is in session the meetings are convened at 9:00 A.M. in a special room in the National Diet Building. By way of comparison, the Irish cabinet also meets biweekly, but ministers are required to attend an annual retreat lasting two or three days. The British cabinet normally meets for about two hours each Thursday morning at Whitehall when Parliament is in session, although under Tony Blair cabinet meetings were said to last about an hour (ibid., 247). Cabinets in Australia, Austria, Belgium, the Czech Republic, and Denmark also meet once weekly. The Norwegian cabinet generally meets on Monday and Thursday, and again briefly in the presence of the monarch on Friday. Meanwhile, the Swedish cabinet meets weekly when the Riksdag is in session, but ministers who are in Stockholm are expected to meet each weekday for “lunch deliberations.” In addition, Swedish ministers are expected to attend a weekly lunch deliberation as well as a short annual retreat (Larsson 1994, 173–174).

Japanese cabinets endure about as long as their counterparts in other countries. According to the government’s official system, the thirty-two individuals who held the premiership from May 1946 (first Yoshida cabinet) through December 2012 (second Abe cabinet) formed a total of ninety-three cabinets, or just under three cabinets per government. The average life expectancy of these governments was 760 days, meaning that prime ministers rotated about every two years.14 In contrast, a study of government duration in nineteen Western countries from 1945 to 1999 found the Finnish (399 days), Italian (350 days), and Portuguese (327 days) cabinets to be the most ephemeral, while the Canadian (945 days), British (995 days), and Luxembourgian (1,122 days) cabinets were the longest-lived (Huber and Martinez-Gallardo 2004). Viewed in comparative context, therefore, Japan’s governments find themselves in the middle of the pack when it comes to life expectancy.

Beginning in the early 1970s, a hierarchy of ministerial portfolios (p.12) emerged in Japanese cabinets that bears a close likeness to those found in other parliamentary democracies. Because the Japanese prime minister is the head of government and something more than primus inter pares in cabinet decision-making, it is natural that that portfolio would be accorded high esteem. In addition, the holders of the finance, foreign, and international trade and industry portfolios are regarded as the crème de la crème of departmental ministers (Kato and Laver 1998; Ono 2012). In this regard, the ministerial pecking order in Japanese cabinets corresponds with the results of an expert survey of twenty-three countries, which found that the finance and foreign affairs portfolios ranked no lower than third among all portfolios, while the economy/industry portfolio also ranked near the top (Laver and Hunt 1992, 133–316).

Of course, a permanent, professional staff is essential to assist ministers in shouldering the onerous executive burden they are called upon to bear. In Japan, the Cabinet Office, Cabinet Secretariat, and Cabinet Legislation Bureau provide a large measure of this administrative support. In 2012, the Cabinet Office boasted a full-time staff of 2,202 national government employees, while the respective staffs posted to the Secretariat and CLB numbered 338 and 72 employees (Ministry of Internal Affairs and Communications 2012b). On the surface, the permanent staff supporting Japan’s cabinet—nearly 3,000 employees—would appear to be considerably larger than that employed in the British Cabinet Office, with its staff of 1,230 employees (Office for National Statistics 2010). However, if the employees posted to Britain’s other cabinet-related agencies are included—in other words, the Central Office of Information (970 staffers), National School of Government (250 staffers), and Office of the Parliamentary Counsel (80 staffers)—the difference is not significant.

At least on the surface, therefore, Japan’s premiers and cabinet ministers appear to be amply empowered, sufficiently supported, and similar in form and function to their counterparts in other advanced parliamentary democracies. In other words, it seems that Japan’s foremost executive leaders possess the requisite formal authority and resources to carry out the functions tasked to them by the Constitution and related laws. Why, then, did the Kan cabinet fail to provide strong, effective leadership after the disasters of March 11? And why did a succession of prime ministers and cabinets who occupied the executive helm throughout the “lost decades” fail to muster those powers to right the ship? Is this the result of flawed leadership or of the failure of the cabinet system?

(p.13) Institutionalization

A theoretical guide is necessary to understand the process by which Japan’s cabinet system came to be what it is. The question to begin with is this: To what extent has cabinet government become institutionalized in the Japanese context? Institutionalization is a ubiquitous process that has been observed in a variety of national, subnational, and supranational political organizations around the world (Meyer and Rowan 1977; Tolbert and Zucker 1996). It employs parsimonious concepts to explain a universal process not bound by geography, culture, time, or regime type.

Institutionalization is the process whereby an organization becomes well established and, in so doing, acquires value and stability as an end in itself (Huntington 1965, 394; Ragsdale and Thies 1997, 1282). An organization’s value is enhanced as it develops a distinctive identity, functions, and mode of operation. The degree of institutionalization is determined by the organization’s progression along four dimensions: complexity, coherence, autonomy, and adaptability (Huntington 1965). Complexity is reflected in the emergence of an increasingly intricate organizational structure, division of labor, and growth of specialized units. Coherence involves the establishment of boundaries that distinguish the organization and its human element from all other people and groups. Autonomy, in the case of a political organization, is the quality of being relatively free of constraint such that the organization has “authoritative control over policy outcomes, acceded to by those in other organizations” (Ragsdale and Thies 1997, 1282). Adaptability refers to the ability of an organization to perpetuate its existence through leadership succession and modification of its roles, functions, and procedures in response to external and internal challenges. Complexity and coherence relate to the organization’s internal dynamics, while autonomy and adaptability relate to its interactions with the external environment. An organization is not fully institutionalized until it has achieved high and enduring levels of development in each of these four measures.

There is no agreement concerning institutionalization’s causes. Perhaps the earliest formulation was Durkheim’s “density theorem,” which holds that “the division of labor varies in direct proportion to the volume and density of societies, and if it progresses in a continuous manner over the course of social development, it is because societies become regularly more dense and generally more voluminous” ([1893] 1984, 205; Polsby 1968, 164). This is implicit in Weber’s thesis that large, highly differentiated (p.14) bureaucracies, whose power derives from rational-legal authority, emerged as a consequence of the development of market economies and centralized states (Weber 1946, 204–209). It is also evident in Michels’s study of Germany’s Social Democratic Party, from which he derived his famous “iron law” that holds, “who says organization, says oligarchy” (Michels [1911] 1962). In fact, the expansion of the role of government has been a persistent theme in countries around the world since the nineteenth century, as reflected in the rise of the “Administrative State” and the “Welfare State” (for example, Ashford 1986). Hence, as a larger portion of the nation’s life came under the sway of centrally made decisions, “the agencies of the national government institutionalized” (Polsby 1968, 164).

Yet institutionalization is also driven by human choice and ambition. This is seen in Parkinson’s observations concerning increases in staff size and paperwork generated in Britain’s Colonial Office even as the overseas empire was contracting, and, therefore, should have resulted in the opposite. Indeed, Parkinson’s “laws”—officials “multiply subordinates, not rivals” and “make work for each other”—derive from the rational self-interested actions of bureaucratic actors rather than abstract organizational forces (Parkinson 1955). As Ragsdale and Thies observe, individuals affiliated with the organization employ their energies and seek to amass “resources to establish the organization, bolster its persistence, and make their activities more routine. The environment creates conditions for the organization to be taken for granted as it conducts specialized activities upon which other units grow to depend” (1997, 1283; also Zucker 1991, 105). As Searing found in his study of political roles in Britain’s Westminster system, the principal incentives that attracted ministerial aspirants were ambition, status, stimulation, and power (1994, 356–357). Ministers—the vast majority of whom are, after all, professional politicians—seldom see their careers advance when their agencies shrink in budget, staff size, or policy significance. Such is also the case with career bureaucrats, who have a vested interest in seeing their agencies grow.

Institutionalization theory provides conceptual tools with which to explain how Japan’s cabinet system developed. It draws attention to the frequently uneven progress of development along the dimensions of institutionalization and the consequences that follow. For example, the battle to privatize Japan’s postal system pitted the Koizumi cabinet against the “postal family,” a deeply entrenched subgovernment composed of influential LDP lawmakers, government officials, and postmasters that epitomized (p.15) Japan’s rigidly fragmented political structure. To adapt government policy to restore economic prosperity, it was necessary for the Koizumi cabinet to declare war on a traditional pillar of support for the ruling party that gave birth to the cabinet. “When institutions are better at holding the line than responding to change,” explains Kesselman, “political constraint and overinstitutionalization can occur” (1970, 25; Reed 1991). I explore the implications of uneven institutional development in the substantive chapters of this book, but for now it is necessary to ask how to identify a well-established cabinet system. Specifically, what would such a system look like in the Japanese context? On this score, the logical comparison is to Britain’s widely emulated Westminster system.

The Westminster Ideal

An established Westminster system is characterized by high and enduring levels of development in each of the four measures of institutionalization. Because ministerial appointment bestows prestige, power, and perquisites, cabinet portfolios are highly coveted and almost exclusively reserved for ruling party lawmakers who possess the requisite level of seniority and ambition. Because prime ministers believe that “parliamentary apprenticeships are prerequisites for understanding ministerial roles and performing them successfully,” they will distribute few if any portfolios to “lateral entrants” or insufficiently senior MPs (Searing 1994, 362; Rose 1971, 401). British ministers tend to serve nearly fifteen years in Parliament before entering the cabinet, and a similarly lengthy parliamentary apprenticeship seems reasonable to expect of Japanese ministerial aspirants (Searing 1994, 348; Heasman 1962b, 318). Assuming that elections are held every three years, novice ministers would be in their fifth terms. It follows that the majority of ministers will embark upon political careers by the time they reach their mid-forties and will emerge from “occupations of a kind that facilitate the pursuit of politics as a serious career,” such as law, media, academia, parliamentary staffer, and local elective politics (King 1981, 262–263). Few young people possess the “name brand,” established campaign organizations, or riches required to launch and sustain a parliamentary career. This puts the offspring or close relatives of MPs in a position of comparative advantage. And because “the only significant path to the top of the polity passes through the university,” cabinet ministers will tend to be well educated (Aberbach et al. 1981, 51).

(p.16) Complexity is a defining characteristic of an established Westminster system. The British system began as a small, nominal cabinet within the Privy Council and evolved into an expansive and specialized network of executive organs, multiple levels of ministerial roles, and a “pecking order within the Cabinet that is very well recognized by Ministers” (Searing 1994, 283–284). I would expect those at the top of the ministerial hierarchy to possess the basic attributes required of a cabinet minister mentioned in the previous paragraph, albeit in more concentrated dosages.15 It is reasonable to assume that the high-status ministers will have been elected to at least seven parliamentary terms in the dominant chamber of Parliament, amassed more than two decades of service as MPs, and previously held ministerial posts. In other words, the highest-status portfolios will not be given as gold watches simply to reward many years of faithful service. Also, the lion’s share of prestigious portfolios likely will be reserved for men—alas, “women have been found to hold fewer cabinet positions, and where they have been appointed, to be mainly allocated portfolios with ‘feminine’ characteristics and lower levels of prestige” (Krook and O’Brien 2012, 1). While the British cabinet evolved with little external influence, Japan’s modern cabinet system incorporated elements of foreign models. Thus, even though the postwar cabinet system did not materialize out of thin air, I would expect advancement toward institutionalization to be reflected in increasingly differentiated organizational structures and ministerial roles (Searing 1994, 344). This does not necessarily mean that cabinets will become larger and larger. On the contrary, I would expect that lawmakers would, through intuition or trial and error, discover that, as “Parkinson’s Law” and empirical assessment have shown, cabinets of more than around twenty members perform poorly (Klimek et al. 2008).

An established Westminster-style cabinet is the most powerful executive organ, and as such it should be able to act with relative autonomy vis-à-vis the legislature, government bureaucracy, and other potential competitors for executive primacy (Lijphart 1999, 10). The cabinet sets the general course for government policy, acts as the ultimate coordinator and arbiter of last resort in executive affairs, and oversees the various departments of government. In the words of a former British cabinet minister, the cabinet is at the “centre of the web” (Searing 1994, 366). The cabinet sets the agenda for legislative deliberation and sponsors the largest share of policies that are enacted into law. In addition, the cabinet oversees the government bureaucracy, referees inter-ministerial disputes, and reviews (p.17) and authorizes all major policy decisions (Kavanaugh 2000, 239). Ministers are the supreme authorities within their departments, over which they are individually responsible. At the same time, all ministers are subordinate to the cabinet, which, as a team, possesses powers that are greater than the sum of its parts. The prime minister has the power to appoint and dismiss ministers and, of necessity, plays a role that exceeds that of first among equals. Yet the cabinet’s “consent to major initiatives usually must be obtained, and not even the most determined Prime Minister could prevail against the opposition of his or her colleagues for long” (Thomas 1998, 12). In sum, a Westminster-style cabinet should be firmly entrenched in the center of the central state executive.

Finally, a well-established Westminster-style system is the product of an evolutionary process that requires “the capacity to respond to the challenges of time and changing environment; the dropping of old tasks and taking on of new ones and the resolution of succession problems” (Dominguez and Mitchell 1977, 173). For this reason, I perceive organizational age and iterated leadership succession as simple indicators of durability. Indeed, a defining feature of the Westminster system is the broadly embraced legitimacy and irregularity by which prime ministers and cabinet members assume their executive posts, only to be replaced by new prime ministers and cabinets. To endure and continue to fulfill their core functions, cabinets must adapt to the challenges posed by protracted economic downturns, mushrooming government deficits, public health concerns, environmental problems, demographic changes, and natural and manmade disasters. In addition, they must deftly respond to international structural changes (for example, the end of the Cold War), tension in foreign relations, territorial conflicts and challenges to national sovereignty, and homegrown as well as international terrorism. Likewise, they must effectively respond to structural changes or challenges in the political order (such as the advent of coalition governments) and major corruption scandals.

In sum, a well-established Westminster system boasts high and enduring levels of development in each of the four dimensions of institutionalization. The complexity and coherence dimensions relate to the cabinet system’s internal orientation, while the autonomy and adaptability dimensions relate to its external orientation. A Westminster system presumes the existence of competitive legislative parties whose members are sufficiently disciplined and fearful of punishment at the hands of party leaders (p.18) that their votes seldom deviate from government proposals. To carry out its functions, the cabinet must exhibit a high degree of cohesiveness, as reflected in well-developed criteria for recruiting members and promoting leaders, collectivity norms, and highly differentiated organizational structures. It is important to keep in mind that the cabinet is a collective body whose members jointly make decisions for which they are “equally and jointly responsible,” while allowing “all departmental perspectives and competing claims to be filtered through the specific issue being debated” (James 2002, 6; Buckley 2006, 4; also Rhodes et al. 2009, 120–132). As the foremost executive organ of government, the cabinet must be endowed with sufficient powers and resources to operate with relative autonomy in conducting executive affairs and in adapting to environmental challenges. In its external orientation, therefore, the cabinet system must be capable of effecting strategic choice in executive decision-making (Hrebiniak and Joyce 1985, 340). By distilling the dimensions of institutionalization into two synthetic variables— cohesiveness (coherence + complexity) and strategic choice (autonomy + adaptability)—we can derive a typology of cabinet forms.

Four types of parliamentary cabinet systems emerge, corporatist, confederate, figurehead, and Westminster. The corporatist cabinet exerts a high degree of strategic choice, although it lacks cohesiveness as a result of internal divisions or an inability to control the ruling party or coalition. The confederate cabinet presents a cohesive executive, but its degree of freedom in effecting strategic choice is constrained by a fragmented or unstable policy-making environment. The lack of cohesion on the part of the corporatist cabinet and the fragmented or unstable policy-making environment that confronts the confederate cabinet effectively rule out the possibility of cabinet government in either case. On the other hand, as the foremost organ of executive decision-making, the Westminster cabinet is endowed with sufficient powers and prerogatives to enable it to respond flexibly and tactically to virtually any challenge that may arise. In addition, because its ministers double as leaders of a centralized and disciplined majority party, “it can confidently count on staying in office and getting its legislative proposals approved” (Lijphart 1999, 12). In contrast, the figurehead cabinet clearly is not the most powerful organ of government, even if laws and conventions say that this should be so; it functions mainly to rubber-stamp decisions made elsewhere. Moreover, because the strategic choice of the prime minister and cabinet is constrained—owing to a coalition arrangement or to a decentralized dominant party—these cabinets cannot count on staying in office or getting their legislative proposals approved. These handicaps severely constrain the cabinet’s adaptive capabilities in response to external or internal challenges.


Figure I.2.Typology of Parliamentary Cabinet Systems


(p.19) I argue that a transitional fifth type—a disjoined cabinet—stood at Japan’s executive helm from 1993 through 2013. This is not a distinct cabinet type, but, rather, reflects the chaotic and unsettled state of Japan’s cabinet system since the structural changes and political reforms of 1993 and 1994. While, on occasion, the disjoined cabinet exhibits forceful, coherent executive leadership (for example, the Koizumi cabinet’s campaign to privatize the postal system), just as often its actions are inconsistent and ineffective (such as the Kan cabinet’s handling of the Fukushima Daiichi crisis). The hopes of the citizenry, whose expectations had been elevated by the 2001 government restructuring and the changes promised by the (p.20) Democratic Party of Japan (DPJ) when it became the party in power, were dashed by the inability of the disjoined cabinets to provide forceful leadership under coalition arrangements and Twisted Diets. While Japan’s cabinet system has assumed different forms during different historical periods, it has never developed into a fully functional Westminster-style cabinet system despite meaningful steps in this direction (Estevez-Abe 2006; Krauss 2007; Mochizuki 2007).

Explaining Institutional Change

While institutionalization theory helps to explain how Japan’s cabinet system evolved, it cannot explain why it came to assume its characteristic form and function. Indeed, since institutionalization is a ubiquitous process, we might logically expect all parliamentary systems patterned after the Westminster model to exhibit similar organizational forms and performance levels. Yet, despite outward similarities, Japan has never evolved a fully functioning Westminster-style cabinet system. I argue that the distinctive organizational structures, roles, and relationships that together form Japan’s cabinet system were forged in a matrix of laws, ordinances, political structures, and unwritten codes of conduct. But where do institutions come from? The answer is found in the spurts of institutional innovation that cluster around critical junctures in history when “institutional configurations are upended and replaced by fundamentally new ones” as well as in the gradual evolution of established institutions (Mahoney and Thelen 2010, 2).

North’s definition of the term institution—a “humanly devised constraint” that structures behavior—provides a foundation upon which to build a conceptual framework (1990, 3). Formal institutions are embodied in rules, laws, and constitutions (ibid., 4). In the case of Japan’s cabinet system, important formal institutions include the Constitution, Cabinet Law, National Government Organization Law, and the political party system. During the half-century of LDP dominance, intraparty politics and cabinet decision-making became so intimately intertwined that it was impossible to say where one began and the other ended (Laver and Shepsle 1994b, 7; also Calder 1988, 446). A consequence of the failure of the opposition to oust the LDP was the systematic exclusion of organized labor and environmental, women’s rights, anti-nuclear, and other interest groups that tended to support the opposition (Steinhoff 1989). Of course, this situation could (p.21) have been remedied had the opposition unseated the LDP, but this did not occur from 1955 until 2009. Informal institutions are embodied in behavioral norms, conventions, and unwritten codes of conduct, such as the seniority system for ministerial aspirants and the unanimity rule in cabinet decision-making (North 1990, 4; Rose 1971, 401, 411). Taken together, these formal and informal “rules of the game” constitute what North terms the “interdependent web of an institutional matrix” that patterns behavior and molds organizational forms (North 1990, 3, 95; italics added).

While the Northian definition explains the static effects of institutions, it does not account for their dynamic aspect. Institutions should be viewed as political instruments that are “fraught with tensions because they invariably raise resource considerations and invariably have distributional consequences” (Mahoney and Thelen 2010, 8). In other words, “dynamic tensions and pressure for change are built into institutions,” which embody “political legacies of historical struggles” that invariably dictate an unequal distribution of resources and produce cleavages that later form coalitions for change (ibid., 7, 14; Pierson 2000, 258–259). Those interests that are disadvantaged by existing arrangements represent potential change agents that will constantly be on the lookout for opportunities to redress their grievances. As Mahoney and Thelen explain, “where institutions represent compromises or relatively durable but still contested settlements based on specific coalitional dynamics, they are always vulnerable to shifts” (Mahoney and Thelen 2010, 8). For these reasons, I define an institution as a humanly devised constraint that structures behavior and carries power-distributional consequences.

While much attention focuses on the abrupt bursts of institutional innovation that cluster around critical junctures, most institutional change is gradual and incremental. A critical juncture is a watershed event, or “key choice point,” that leads to the reconfiguration of institutional arrangements or structural patterns (Mahoney 2001, 6–7). It can appear as an immediate response to an external shock, such as the Great Depression or the Second World War. It can also result from a primarily internal stimulus, such as the Iranian Revolution that overthrew the Pahlavi regime in 1979 or the toppling of long-standing dictatorships in Tunisia and Egypt in the “Arab Spring.” But such exogenous shocks occur only rarely, while institutions constantly undergo incremental change. For example, the electoral realignment in the United States that began in 1932—when FDR’s Democratic Party took control of the executive and legislative branches, and, (p.22) more or less, maintained its dominance for the next three decades—facilitated a succession of incremental changes that fundamentally transformed American politics. In this way, the cumulative effects of gradual changes, as well as the occasional exogenous shock, create openings for disadvantaged interests to seek to improve their situation by pressing for change.

Mahoney and Thelen propose a framework for identifying and explaining the different forms of institutional change. Four types of institutional change are identified, displacement (“the removal of existing rules and the introduction of new ones”), layering (“the introduction of new rules on top of or alongside existing ones”), drift (“the changed impact of existing rules due to shifts in the environment”), and conversion (“the changed enactment of existing rules due to their strategic redeployment”) (Mahoney and Thelen 2010, 15–18). Each type of change can be produced gradually, but displacement episodes are also associated with shocks that result in abrupt institutional reconfigurations that produce path-dependent legacies. Particular types of change agents tend to be associated with particular types of change: insurrectionaries are associated with displacement, subversives with layering, renegades with drift, and opportunists with conversion (ibid.,22–27).16

Building upon this framework, I argue that Japan’s cabinet system was significantly recast at eight historical junctures. Two of these resulted in reconfigurations of institutional and structural arrangements. The 1868 Meiji Restoration and the American-led Allied occupation of Japan (1945–1952) were exogenous shocks that resulted in the displacement and recasting of a broad array of political institutions and socioeconomic structures. The six remaining tipping points were precipitated by the accumulation of gradual changes and structural shifts. The changes that followed the establishment of Prime Minister Ōkuma Shigenobu’s “party cabinet” in 1898 and the advent of coalition cabinets following the LDP’s brief fall from grace in 1993 reflect drift, in the sense that existing rules were changed in response to environmental shifts. Similarly, the changes that followed the 1932 assassination of Prime Minister Inukai Tsuyoshi and the “shocks” of the early 1970s—precipitated by actions or threats on the part of U.S. president Richard M. Nixon and the Organization of the Petroleum Exporting Countries (OPEC)—resulted in the layering of new rules on top of or alongside existing ones. The changes that followed the emergence of the “1955 system,” during which the LDP enjoyed parliamentary dominance, can be understood as the conversion of existing rules to serve the needs of the perpetually ruling party. A somewhat similar state of affairs has accompanied the changes that emerged in the post-2007 era of “Twisted Diets,” in which the ruling party in the lower house does not hold a majority in the upper house. (p.23)

TABLE I.1 Institutional Change in Japan’s Cabinet System, 1868 to 2013

Event (Year)

Source of Change

Change Agent

Institutional Change

Cabinet Type

Meiji Restoration (1868)





ŌOkuma’s “Party Cabinet” (1898)





“May 15th Incident”(1932)





Defeat & Occupation (1945)





Creation of “1955 System” (1955)





Nixon & OPEC “Shocks” (1971~)





Coalition Governments (1993)





“Twisted Diets” (2007)






In sum, my approach brings together conceptual tools from political sociology and historical institutionalism. I believe that analytical eclecticism is necessary to explain how Japan’s cabinet system evolved, why it came to assume its characteristic form and functions, and what consequences this holds for Japan’s system of democratic governance and its foreign relations. As Katzenstein and Okawara explain, “theory and policy are both served better by eclecticism, not parsimony” (2001/2002, 185). By combining different paradigms, I am knowingly jumping headfirst into what Evans approvingly describes as the “eclectic messy” center of comparative politics, where getting it right is the analyst’s primary objective (Evans 1996, 2). My central argument is that Japan’s cabinet system was forged through an ongoing process of institutionalization filtered through and interacting with a matrix of key institutions that were periodically transformed as a result of gradual change and the occasional shock.

(p.24) The Context of this Study

In recent years, Japan’s cabinet system has begun to attract scholarly attention. Thayer identifies three “archetypes” in the evolution of Japan’s cabinet system—“the imperial cabinet, the predominant party cabinet, and the coalition cabinet” (1996, 71). Assessing Japan’s 2001 government restructuring, Shinoda sees little meaningful change with regard to the functions of the cabinet, which continues to play an insignificant role, while “actual decisions are made at regularly scheduled meetings of the top career officials of ministries and agencies and then rubber-stamped at subsequent cabinet meetings” (2005, 805). George Mulgan argues that “Japan does not have cabinet government; it has party-bureaucratic government,” by which she means that the cabinet is subordinate to the long-ruling LDP and government officialdom (2003a, 129). Krauss and Pekkanen argue that the LDP never became the strong, centralized ruling party required in a Westminster system. This is because vote mobilization was in the hands of the candidate-support organizations (kōenkai) of party backbenchers, the distribution of offices was the domain of the bosses of the intraparty factions, and the party’s Policy Affairs Research Council performed the role of veto player in policy-making (Krauss and Pekkanen 2010, 5–6, 279–280; also Krauss 1989, 51–52).

While most pundits doubt that cabinet government will arrive on the Japanese scene, at least anytime soon, some offer reasons for hope. Shimizu maintains that the changes made under the Koizumi government ensure the eventual emergence of cabinet government (2005, 404). Estevez-Abe shares this view, pointing out that Koizumi’s bold decision to obtain a popular referendum on his postal privatization bills “cast the die in favor of a Westminster system that centralizes power in the hands of the party leadership and prime minister” (2006, 633). She concedes that, historically, Japan’s cabinet “simply rubber-stamped decisions made by others,” but argues that an “increasingly centralized party structure of the LDP and the strengthening of the Cabinet will push Japan in the direction of a British-style parliamentary democracy” with a “Westminster system” (ibid., 633, 651). Krauss argues that, as a result of the institutional changes following the LDP’s brief hiatus from power in 1993, “Japan has moved somewhat closer to the British model” (2007). Of course, these scholars were writing before the LDP’s temporary expulsion from the helm in 2009, and much of the progress toward enhancing the cabinet’s role taken under the Koizumi (p.25) cabinet was reversed by the LDP- and DPJ-led governments that followed. For these reasons, it is time to take a hard look at the nature, capabilities, and limitations of Japan’s parliamentary cabinet system—which is what this book is about.

Rival Approaches

Theories purporting to explain who rules the roost in the affairs of Japan’s central state executive fall into three broad categories, dominant actor theories, fragmented state theories, and core executive theories. To date, the debate among learned observers has primarily focused on which political actor dominates executive decision-making. Laver and Shepsle identify six general models of executive decision-making—bureaucratic government, legislative government, prime ministerial government, party government, ministerial government, and cabinet government (1994b, 5–8). Each of these models has attracted advocates among students of Japanese executive decision-making.

Bureaucratic government holds that “the elite bureaucracy … makes most major decisions, drafts virtually all legislation, controls the national budget, and is the source of all major policy innovations in the system” (Johnson 1982, 20; also Pempel 1974, Campbell 1977, Inoguchi 1983, Amyx 2004, and Vogel 1996). This view is bolstered by the image of a Japanese cabinet minister as a mere “figurehead who [is] manipulated and controlled by the career officials” (Park 1986, 8). In addition to being a traditional elite, Japan’s career civil servants carry out their duties under the oversight of, at most, six political appointees per ministry. Because ministers rely on bureaucrats for policy expertise, they are prone to become the bureaucrats’ “lap dogs” (Takenaka 2008, 51, 81; Iio 2007). Bureaucratic influence is ensured by the Supreme Court-like role of the Cabinet Legislation Bureau, which examines all draft bills, treaties, and cabinet orders, and whose opinions are regarded as authoritative and difficult to overturn. As Samuels argues, “No administrative agency of the Japanese state enjoys higher prestige or greater independence than the CLB” (2004, 2). Bureaucratic influence is further ensured through the deputy chief cabinet secretary for administrative affairs (Naikaku kanbo fukuchokan), a senior ex-bureaucrat who chairs the biweekly meetings of administrative vice ministers (jimujikan) from each ministry at which the agenda for cabinet meetings is set (Johnson 1995b, 221; Tanaka 2000, 5). Yet the bureaucracy’s (p.26) relative influence has diminished in recent years as a result of liberalization, privatization, scandal, and the fact that fewer ex-officials are opting to pursue “second careers” as MPs.

Party government presumes the existence of a strong executive whose members are “subject to the discipline of well-organized political parties” (Laver and Shepsle 1994b, 7). This model found fruitful application in the works of rational choice theorists, who posit that “real bureaucrats … administer in the shadow of the LDP,” by which they mean that through a series of delegations LDP backbenchers strummed the tune to which cabinet ministers obligingly danced (quotation is from Ramseyer and Rosen-bluth 1993, 120; also McCubbins and Noble 1995). Beginning in the early 1960s, the LDP began requiring that all policy proposals be submitted to its Policy Affairs Research Council for “prior approval” before being sent on to the cabinet and then to the Diet (Krauss and Pekkanen 2010, 20; also George Mulgan 2003a, 140; Holliday and Shinoda 2002, 101). Yet, because of the LDP’s essentially decentralized structure, this veto role actually served to undermine the relative influence of the prime minister (who doubled as party president) and central party leadership, and, in so doing, impeded the emergence of a Westminster-style cabinet system (Krauss and Pekkanen 2010, 279–280).

The four remaining models can be accounted for succinctly. Ministerial government denotes a system in which “individual ministers … are able to have a significant impact on policy in areas that fall under their jurisdiction” (Laver and Shepsle 1994b, 8). This thesis echoes in Campbell’s discovery that, in the politics of Japanese budgetary policy, “if collectively the Cabinet is not an important … actor, individual ministers have often assumed prominent parts” (1977, 151). Prime ministerial government posits that “the cabinet is a mainly residual organization” in policy-making, ministers are “agents of the … prime minister’s will,” and bureaucrats are “confined to implementing rather than making policy” (Elgie 1997, 222–223). While recent reforms and the increasing importance of television have enhanced the prime minister’s role, Japan’s premiers continue to be regarded as passive leaders who serve mainly as “consensus articulators” (Hayao 1993, 202; Angel 1988–1989, 600; Iio 2007). Legislative government assumes that the cabinet mechanically implements policies made by the legislature (Laver and Shepsle 1994b, 6). The policy gridlock and rapid turnover in governments dictated by the post-2007 Twisted Diets provides fodder for advocates of this model.

(p.27) Finally, cabinet government denotes a parliamentary system with a strong executive that provides tactical direction to government policy, and whose leaders are chosen by Parliament from among its members and bound together by norms of collective responsibility and solidarity. While this viewpoint has attracted a few advocates—including an ex-bureaucrat who opined that the cabinet is now “the main battle ground for policy-making” (quoted in Asahi Shimbun Globe, May 31, 2002)—most observers believe that the cabinet is “more an aggregation than a real institution, its membership turns every year or so … and ministers are normally more concerned with their individual political affairs than with advancing the Cabinet as such” (Campbell 1977, 151).

While they provide insights, these generic models do not account for factors specific to the Japanese case. For example, what about the role of the “policy tribes” (zoku) in shaping executive decision-making? The influential MPs who are members of the various tribes exercise considerable sway over policy outcomes in agriculture, postal affairs, public works, and so on (Muramatsu and Krauss 1984 and 1987; Sato and Matsuzaki 1986; Inoguchi and Iwai 1987; Schoppa 1991; Krauss and Pekkanen 2010). While service as a cabinet minister often affords entrée to a policy tribe, not all zoku politicians are incumbent ministers. The influence of these policy tribes contributed to the “political capture” that afflicted policy-making, especially in the latter years of the LDP’s hegemonic reign (Samuels 2013, 181). And what about the role of the LDP’s intraparty faction bosses in allocating ministerial posts? In fact, factional “balance” was a critical concern in the allocation of portfolios under the “1955 system” (Thayer 1969, 31–35; Curtis 1988, 86–87; Krauss and Pekkanen 2010, 132–134). This diminished the prime minister’s leadership role, and on occasion produced cabinets that seemed to be under the “remote control” of faction bosses. Finally, how do we account for the external pressure (gaiatsu) applied by the United States, which, after all, remains the guarantor of Japan’s national security and a major destination for its exports? It goes without saying that American views strongly influenced Japanese executive decision-making during the occupation period (Dower 1999). Much like the proverbial elephant in the room, explicit or implicit pressure from the United States is difficult for Jap-anese policy-makers to ignore, especially when it comes to bilateral trade and security issues (Schoppa 1997; Pyle et al. 2010; Woodall 1996, 129–132).

Fragmented state theories come in a variety of forms. The most extreme variant, the “stateless nation” thesis, argues that “there is no supreme institution (p.28) with ultimate policy-making jurisdiction. Hence there is no place where …the buck stops. In Japan, the buck keeps circulating” (van Wolferen 1989, 26, 5). “Hollow state” theory holds that “government is becoming so fragmented that pulling business together at the centre is now an almost impossible task” (Holliday and Shinoda 2002, 108). Various case studies confirm the fragmented nature of Japanese executive decision-making, especially during the 1990s. In his research on Japanese budgetary politics, Campbell observed signs of fragmentation in the 1960s in the form of “subgovernments made up of agencies, specialized LDP politicians, and interest groups—sometimes in alliance, sometimes with much internal division” (Campbell and Scheiner 2008, 99; also Campbell 1977 and 1984). Subsequent studies drew attention to the ways in which a fragmented political structure shaped policy outcomes in a variety of arenas, including education policy, tax reform, public works, postal services, and the response to the banking crisis (for example, Hayao 1993; Schoppa 1993; Kato 1994; Woodall 1996; MacLachlan 2004; and Amyx 2004).

Finally, core executive theories focus on the organs and structures that “pull together and integrate central government policies, or act as final arbiters within the executive of conflicts between different elements of the government machine” (Dunleavy and Rhodes 1990, 4; also Holliday and Shinoda 2002; and Krauss 2007). This approach explicitly recognizes the multiplicity of actors that, of necessity, must be involved in executive decision-making in an advanced parliamentary democracy. In a study comparing the core executives of Britain and Japan, Holliday and Shinoda argue that the cabinet is not a “key player” in the core executive of either country. They posit that Japan’s core executive is composed of the “Prime Minister, the Chief Cabinet Secretary and his three deputies in the Cabinet Secretariat, the key leaders of the LDP, plus top officials in the supporting offices. Those offices are the Prime Minister’s Office, the Ministry of Finance and the Ministry of Foreign Affairs” (2002, 98).

While there is merit in each of these explanations, I have come to see things in my own way. As I demonstrate in the substantive chapters of this book, the relative influence of specific actors has varied over time in response to domestic and external crises, institutional change, and tectonic shifts in political, economic, and social structures. Moreover, I argue that by the early 1990s a fragmented policy-making environment severely limited the options that prime ministers and cabinets could pursue in attempting to guide the country out of the gloom and policy lethargy that defined (p.29) the “lost decades.” While core executive theory draws attention to power relations among a number of actors, I believe that its most fruitful application is as a conceptual device for assessing shifts over time in that power balance.


By focusing on the evolution of Japan’s cabinet system, the core chapters of this book highlight the capabilities and limitations of a crucial component of Japanese parliamentary democracy. Throughout this book, I seek to solve two central puzzles. First, why has cabinet government failed to take root in Japan? This is puzzling because postwar Japan has been governed under Westminster-style parliamentary institutions and yet has not managed to nurture cabinet government. In other words, why has Japan failed to produce cabinet government in practice despite having a parliamentary system in form? Focusing on the manner and degree to which the cabinet system has become institutionalized reveals much about the capacity of Japan’s political executive—what it can and cannot do, as well as how it lives up to and falls short of its intended place in a parliamentary system. In this regard, the Westminster ideal provides benchmarks against which to assess the degree to which the Japanese cabinet system has progressed toward institutionalization.

This calls attention to a second puzzle. Why has Japan’s cabinet system assumed its characteristic form and function? Since institutionalization is a ubiquitous process, why is the Japanese cabinet system distinctive? The question draws attention to the shaping influence of institutions, which requires that we “take time seriously” by accounting for the significant roles played by context and history in determining institutional change in the “longue durée” (Pierson and Skocpol 2002, 685; Braudel 1982, 25–54). It also highlights the distributional consequences of institutions and the role of critical junctures and tipping points as strategic openings for institutional reconfiguration.

While the domestic implications should be obvious, understanding Japan’s cabinet system is relevant to academics, policy-makers, and business leaders in other countries. For one thing, the findings presented in this book illuminate the dimly understood inner workings of the cabinet system of an important non-Western parliamentary democracy. Perhaps this intensive analysis of the Japanese case might offer lessons for reformers in Japan as well as for those who seek to build democratic institutions in countries such as Iraq and Afghanistan. Or, perhaps someday the Japanese experience may help to guide the establishment of democratic regimes in Cuba, Nigeria, China, and the new regimes born during the Arab Spring. At the very least, Japan’s struggles with this fundamental component of parliamentary governance should serve as a cautionary tale for those who believe that parliamentary institutions in form equate to parliamentary government in practice. In fact, growing democracy is not easy, and in this regard the Japanese case offers crucial lessons for understanding the challenges and disappointments that confront today’s developing countries. (p.30)


(1) . As a result of the 1995 Kōbe earthquake, the posts of Minister of State for Disaster Management and Deputy Chief Cabinet Secretary for Crisis Management were created to coordinate disaster response policies.

(2) . The “lost decades” refers to the prolonged economic stagnation and absence of effective political leadership that followed the bursting of the “bubble economy” in the early 1990s. It left Japanese financial institutions buried under a mountain of nonperforming loans and produced a succession of governments that proved unable to enact fruitful policy solutions.

(3) . The symbolic powers of Japan’s head of state are constitutionally defined, while in Britain those powers developed through historical convention. Australia’s constitution grants extensive powers to its head of state, although those powers are not actually used. Aurelia George Mulgan kindly pointed this out to me.

(4) . Members of the press corps were allowed to observe cabinet meetings in December 1985, September 1993, and April 2002 (Naikaku seido hyakunen shi henshū iinkai 1980, 22 and 24).

(5) . Tent villages became a fixture on the scene around the time of the formation of the Okada cabinet in July 1934 (Naikaku seido hyakujūchōnen 1995, 130).

(6) . For purposes of this study, the Imperial Household Office, Fair Trade Commission, Financial Services Agency, and the National Personnel Authority are not considered part of the cabinet system.

(7) . Of the forty-nine non-confidence motions submitted between May 1947 and May 2013, only four were approved—December 23, 1948 (second Yoshida cabinet), March 14, 1953 (fourth Yoshida cabinet), May 16, 1980 (second Ōhira cabinet), and June 18, 1993 (Miyazawa cabinet).

(8) . Australia, Austria, Canada, the Czech Republic, Denmark, Finland, Germany, Greece, Hungary, Iceland, Ireland, Italy, Korea, New Zealand, Poland, Spain, Turkey, and the United Kingdom also permit ministers to retain their parliamentary seats.

(p.236) (9) . By traditional convention, British ministers are expected to be MPs (Rose 1971, 401, 411). Many of the isolated instances in which non-MPs have been awarded portfolios occurred during wartime, and came with the expectation that these individuals would win a seat in the House of Commons in a subsequent by-election. Peacetime examples are rare. In October 1964, Prime Minister Harold Wilson gave the foreign secretary portfolio to Patrick Gordon Walker and the technology portfolio to Frank Cousins even though neither held seats in the Commons. Walker and Cousins agreed to accept peerages, and were expected to win seats as “carpetbaggers” in by-elections held several months later. Cousins emerged victorious, but Walker was defeated and had to surrender his portfolio (Brazier 1997, 64–65). More recently, in October 2008, Prime Minister Gordon Brown gave the business secretary portfolio to Peter Mandelson, a veteran of Tony Blair’s cabinet who did not currently hold a seat in Commons, but Mandelson was immediately elevated to the House of Lords (Telegraph, October 3, 2008). I am grateful to Arthur Stockwin for pressing me to clarify this point.

(10) . Britain’s “government” consists of about one hundred members who are nominated by the prime minister and appointed by the monarch; the “cabinet” itself consists of only about fourteen of these individuals (Curtis 1997a, 68). In contrast, Japan’s government in late December 2012 consisted of seventy-four individuals—nineteen cabinet ministers, three deputy chief cabinet secretaries, twenty-five state secretaries, and twenty-seven parliamentary secretaries.

(11) . A study of Canadian provincial governments found that larger cabinets complicate decision-making and erode teamwork (White 1994, 262).

(12) . The types of cabinet decisions are explained at “Naikaku seido to rekidai naikaku,” www.kantei.go.jp/jp/rekidai/1-2-5.html; accessed May 23, 2013.

(13) . Since May 7, 2002, cabinet meetings have been held in a room on the fourth floor of the Prime Minister’s Official Residence (Kantei), except during parliamentary sessions, when they are held in a special chamber in the National Diet Building. Prior to this, the cabinet met in a room on the second floor of the old Kantei, a Frank Lloyd Wright-inspired structure that opened in March 1929. In the closing days of the Pacific War, cabinet meetings were held behind two-meter-thick reinforced walls in the National Defense Telephone Bureau (Naikaku seido hyakujuchonen 1995, 99).

(14) . The term “government” denotes the continuous period from the appointment of a prime minister until his or her dismissal.

(15) . As Heasman observed, “the relative importance of any office … [depends] in large measure on the influence and character of the incumbent as well as on conditions of the time” (1962a, 309).

(16) . Mahoney and Thelen (2010) associate symbionts with institutional drift.