Analyzing Accountability Relationships in a Crisis: Lessons From the Fukushima Disaster

Abstract

The Fukushima Daiichi nuclear disaster became the worst nuclear accident since the 1986 Chernobyl disaster. Many studies have examined the so-called “first-order causes” of the Fukushima disaster, such as economic interests and lax regulations. However, studies examining the disaster have paid little attention to how it escalated during the response phase, that is, “second-order causes.” This study examines the unresolved question using an analytic frame of accountability relationships. The results demonstrate how crisis management organizations of government faced cross-pressures within a web of accountability relationships while dealing with the disaster. In particular, these organizations’ responsiveness to hierarchical accountability had a negative effect on the political accountability relationships. It is the contribution of this research to specifically identify multiple and complex relationships between the types of accountability. Previous studies have mostly treated professional accountability as the single dependent variable. In contrast, this research argues that the other accountabilities—hierarchical and political—can be dependent variables as well. More importantly, inconsistent with the findings of previous studies, the economic pressure of political accountability had no effect on professional accountability relationships as a result of the heroic efforts of a nuclear plant manager.

Keywords

accountability relationships the Fukushima disaster crisis management centralized decision making nuclier power village

The Fukushima Daiichi nuclear disaster was a catastrophic incident that happened on March 11, 2011, involving the Fukushima number 1 nuclear power plant. The disaster occurred when the plant was destabilized by a gigantic tsunami caused by the Tōhoku earthquake. It resulted in the meltdown of three of the plant’s six nuclear reactors, and the worst nuclear disaster since the 1986 Chernobyl incident. The National Diet of Japan Fukushima Nuclear Accident Independent Investigation Commission (NAIIC) announced that the nuclear disaster was “man-made” and that its immediate causes were all preventable. It was stated that the plant operator, Tokyo Electronic Power Company (TEPCO), regulators including the Nuclear and Industrial Safety Agency (NISA) and the Nuclear Safety Commission (NSC), as well as the Ministry of Economy, Trade, and Industry (METI), the government body promoting the nuclear power industry, all failed to enforce fundamental safety requirements in advance of the disaster (Kurokawa et al., 2012). In turn, these actors, the critical parts of the so-called “Nuclear Power Village,” were not adequately prepared for disasters such as enormous earthquake and tsunami that occurred in Fukushima due to its collective solidarity (which resembles that of “Iron Triangle”) over promoting nuclear energy (see the succeeding section on the Village for details).

However, the more serious issue regarding the Fukushima disaster resides in the fact that as the crisis escalated, the Prime Minister’s office and the aforementioned actors were ineffective in “preventing or limiting the consequential damage” at the Fukushima number 1 plant (Kurokawa et al., 2012). Figure 1 shows the devastating results of polls that asked people’s opinions regarding Prime Minister Kan and his government’s crisis management concerning the Fukushima disaster.¹ As time passed, an increasing number of people indicated that they believed Kan’s leadership was not effective, an increase from 69% in April 2011 to 75% in May 2011. In addition, the percentage of those who did not approve of Kan’s response to the Fukushima disaster increased from 61% in April 2011 to 73% in May and June 2011. This rise in the disapproval rate was accompanied by a decrease in the percentage of people who indicated that they trusted the government report on the radiation released from the Fukushima nuclear plant, which dropped from 32% in April 2011 to 26% in May 2011. All poll results reflect how crisis management of the Kan administration was not effective during acute stages of the Fukushima nuclear disaster.Surprisingly, studies examining the Fukushima disaster have paid little attention to how it escalated during the response phase or why the response failed; that is, the so-called “second-order causes” (Rijpma & van Duin, 2001). Indeed, previous studies have examined the “first-order causes” of the Fukushima disaster, focusing mainly on discovering causes that “accumulate but go unnoticed [during an incubation period], until a trigger or precipitating event occurs” (Rijpma & van Duin, 2001, p. 144; Nakamura, 2012; Nakamura & Kikuchi, 2011). In this regard, the present study fills the gap in the literature by addressing the following questions: Did primary actors involved in the management of the Fukushima crisis manage the crisis accountably? For example, was the meltdown of the three nuclear reactors of the Fukushima plant avoidable or could it have been minimized during the response stage? That is, the study determines whether, and to what extent, primary actors were “accountable jugglers” (Radin, 2002) during the crisis and examines the impact of such behaviors on the outcomes of crisis management. As emphasized by the NAIIC, which stated that the primary actors of the Fukushima disaster “lacked a sense of responsibility to protect people’s lives and society” (Kurokawa et al., 2012), it is critical to examine “second-order causes” of the Fukushima disaster based on the framework of accountability relationships.² Toward this aim, this study adopts “the most common and significant” types of accountability relationships in recent public administration literature, first suggested in Romzek and Dubnick (1987): hierarchical, professional, political, and legal accountability.

Figure 1.

Poll results regarding Kan and his administration’s crisis management.

This article is divided into four parts as follows: The first part presents first-order causes such as unpreparedness for huge tsunamis, economic interests, and lax regulations. The second part illustrates reasons why attention was focused on the behaviors of the Prime Minister and his advisory configurations³ in their political contexts. The third part conceptualizes four types of accountability relationships mentioned above, and provides an in-depth analysis of the Fukushima disaster. The fourth part concludes with a summary of the case analysis and provides important implications for future crisis management efforts.

First-Order Causes by the “Nuclear Power Village”

It is apparent that the Fukushima disaster has a starting point that may date back many years before the actual onset of the accident. The “nuclear power village” holds direct accountability concerning the first-order causes of the case: lax government regulations, strong economic interests at the expense of safety, and secrecy. The nuclear power village, a closely knit decision-making network, was composed of the Diet (traditionally Liberal Democratic Party [LDP] politicians), government officials (particularly from the NISA), staff members of TEPCO’s headquarters, and other actors. In Japan, there is a long-standing custom in which public officials in nuclear regulatory agencies are hired for executive positions in electronic power companies soon after retiring from government jobs, which is referred to as amakudari.⁴ In fact, the vice president position at TEPCO had long been filled by a retired METI official. Furthermore, there was another tradition for the conservative LDP to receive financial contributions from these power companies, which was critical for the party to win elections. In turn, there is little doubt that government regulation was lax and even overwhelmed by the economic interests of power companies (Nakamura & Kikuchi, 2011, p. 897).

In 2008, TEPCO knew that the Fukushima nuclear plant was vulnerable to a great tsunami. However, it decided not to build a higher seawall due to financial considerations. In addition, nuclear utility companies have been hesitant to retire older reactors because of the high cost of decommissioning reactors (estimated at $2.5 billion per reactor; Kingston, 2012). In fact, cost considerations have negatively affected the management of the Katrina disaster in the United States as well. Despite repeated warnings about storm risks, the Bush administration sharply reduced the Army Corps of Engineers budget for levee repairs and renovations around New Orleans by more than half in the years preceding Hurricane Katrina. The budget cuts produced a direct causal link between Bush and the Katrina disaster (Boin, ‘t Hart, McConnell, & Preston, 2010, p. 719; Blumenthal, 2005).

Nuclear power experts shared the thought that ordinary people would not be able to understand the subject well anyway (Asahi Shinbun, 2011, March 16b)—secrecy. In turn, the nuclear village nurtured solidarity and group-thinking mentality such that critics were ostracized (Kingston, 2012). As a consequence, the members of the nuclear village successfully constructed the myth of “safe, cheap, and reliable” nuclear power. Overall, insights into the first-order causes can shed light on why the Fukushima disaster occurred. However, they cannot explain why the response to the accident failed once it was triggered. This study suggests that the accident escalated during the response phase because of ineffective accountability relationships by Prime Minister Kan and his crisis advisory committees (i.e., configurations).

Centralized Decision Making in a Crisis

It is a well-known argument in the crisis management literature that government decision-making bodies become highly centralized under a national crisis (Boin & ‘t Hart, 2003; ‘t Hart, Rosenthal, & Kouzmin, 1993).⁵ There are generally two reasons for this high degree of centralization. First, the requirements for prompt decision making and urgent responses are hardly reconcilable with bureaucratic, time-consuming policymaking structures (Bronner, 1982; Cohen, 1979; Perrow, 1967; Rosenthal, 1986, p. 106; ‘t Hart et al., 1993, p. 14). Therefore, government responses to a crisis involve ad hoc adaptation through the centralization (‘t Hart et al., 1993). Second, a national crisis or disaster incites a pervasive desire among people for a plebiscitarian leader and his or her show of decisional resolve (Lowi, 1985; Wolfenstein, 1967). In short, a number of crisis management scholars have focused on the leadership of chief executives and their advisory systems in times of acute emergencies (e.g., Boin & ‘t Hart, 2003; Boin et al., 2010; Lodge & Wegrich, 2012; Masters & ‘t Hart, 2012; ‘t Hart, Tindall, & Brown, 2009). Accordingly, this study focuses on the composition and modus operandi of configurations surrounding Prime Minister Kan at the time of the Fukushima crisis.

Political Contexts

It is important to illustrate how the party leadership of the LDP as well as the Democratic Party of Japan (DPJ) historically made efforts to control the bureaucracy. This illustration should provide a better understanding of the underlying context of Japan’s political situation during the crisis. Traditionally, cabinet ministers had been helpless to control autonomous bureaucrats in Japan (Mulgan, 2000a) because bureaucrats exercise considerable control over both information and technical expertise. The pathology of uncontrollable ministries was uncovered through the disclosure of numerous financial scandals involving bureaucrats in the 1990s. In response to harsh public criticism, Prime Minister Hashimoto of the LDP, a major conservative party in power since 1955 except for some brief periods, radically reorganized Japan’s central government under the “Hashimoto Administrative Reform” in 1998 by strengthening the power of the prime minister and the cabinet while weakening that of bureaucrats.⁶ Pempel (2000) regarded this transformation as part of a “regime shift.” The reinforcement of cabinet leadership was further facilitated by Prime Minister Junichiro Koizumi (LDP), an exceptional leader in Japanese politics with great fortitude and resolve, and continued through two consecutive administrations of the DPJ, a centrist political party (2009-2011). As soon as he took office, Yukio Hatoyama (2009-2010), the first DPJ prime minister, declared the start of a cabinet-centered government (Nakamura, 2012, p. 251). Similarly, Kan (2010-2011), the second DPJ prime minister and the chief commander during the Fukushima crisis, led the context of the public’s desire for strong political leadership to reform conservative bureaucrats. Both DPJ administrations were under the party platform of governance led by politicians (so-called “seiji shudo”). The platform was exemplified by the direct command of political leaders in the field without going through relevant agencies.

In addition, these two DPJ administrations were markedly different from previous LDP administrations in terms of the level of experience and capability to govern bureaucracies. This lack of prior experience in government led to a gap between DPJ politicians and professional bureaucrats.⁷ In turn, professional bureaucrats were estranged from party leadership and forced to wait until they received cues from the chief executive and his political appointees to take action (Nakamura, 2012, p. 252).

Accountability in Crisis

Conceptualizations

It is crucial for governments to effectively deal with multiple accountability issues, particularly in the face of crises, during which they “come front and center” (Romzek & Ingraham, 2000, p. 244). In general, issues concerning accountability are raised when events go wrong, when consequences of government’s actions or inactions are perceived to be deleterious to the public as a whole or to certain groups or individuals within it—“accountability” tends to have “negative connotations” (Gregory, 2003, pp. 557-558). Accountability was originally referred to as answerability to some authority for one’s behavior or conduct. Thus, the original or “core accountability” has commonly regarded the process in which “one side, that calling for the account, seeks answers and rectification while the other side, that being held accountable, responds and accepts sanctions” (Mulgan, 2000b, p. 555). Although the word “accountability” is now commonly used in public administration literature, the usage of another synonymous concept, “responsibility,” was more dominant in the past. In the classic debates between Carl Friedrich and Herman Finer, both scholars conceptualized the term “responsibility” differently—the former emphasized the inward responsibility of public employees to their professional knowledge and values whereas the latter reasserted the criticality of responsibility to external directions from their political masters (Finer, 1941; Friedrich, 1940). Accepting the two scholars’ ideas on “responsibility,” Mosher (1982) suggested two types of responsibility: “objective responsibility,” which addresses the formal structure within which the chain of principal–agent relationships work, and “subjective responsibility,” which focuses on a psychologically oriented moral obligation for the individual to explain and justify their actions. Consequently, the notion of the former is compatible to Finer’s assertion on responsibility (or “core accountability”) whereas that of the latter is synonymous with Friedrich’s views. The Friedrich–Finer debate (in turn, Mosher’s two sorts of responsibility) was often described as a discussion on the primacy of different types of “accountability,” internal or external (Guy Peters, 1995, p. 318; Harmon & Mayer, 1986, pp. 47-49; March & Olsen, 1995, pp. 165-7). In turn, “accountability” and “responsibility” were frequently used synonymously and interchangeably (Gregory, 2003, p. 557; Mulgan, 2000b, p. 558).

However, “responsibility” is now increasingly confined to its “more internal aspects of official activity”—that is, Friedrich’s inward liability of the individual to his or her moral values (Mulgan, 2000b, pp. 557-558) or Mosher’s subjective responsibility. It may be asserted that “responsibility” is a part of “accountability” (Mulgan, 2000b, p. 558). On the other hand, other scholars such as Uhr (1993) and Gregory (2003, p. 564) regard accountability and responsibility more dichotomously. Mulgan (2000b, p. 558) perceives this view as “a more promising development in the territorial jousting between ‘accountability’ and ‘responsibility’” and argues they are two sides of the same coin, indicating its negative and positive end respectively. For example, Uhr (1993, p. 4) illustrates “[a]ccountability is about compliance with authority, whereas responsibility is about empowerment and independence . . . . If accountability is about minimizing misgovernment, responsibility is about maximizing good government.”

But, the word, accountability, bears many different aspects, and thus still remains “an elusive phenomenon” (Thomas, 1998, p. 387) or “chameleon-like term” (Mulgan, 2000b, p. 555). Romzek and colleagues (e.g., Radin, 2002, pp. 17-19; Radin & Romzek, 1996; Romzek & Dubnick, 1987; Romzek & Ingraham, 2000) defined accountability in a broader sense to facilitate its application to the public sector.⁸ According to these studies, accountability is associated with the means by which government units and their public officials manage various expectations derived within and outside of the unit. These studies provide a robust analytical framework of accountability: hierarchical, legal, professional, and political accountability. First, hierarchical accountability systems focus mainly on internal controls of government units based on the priorities of those at the top of the hierarchy. Close supervision or its surrogate means such as rules, regulations, standard operating procedures, and organizational directives are examples of this type of control. Under such systems, obedience to these control measures is expected. Hierarchical accountability is intimately linked with Mulgan’s (2000b, 2003) “accountability as control,” particularly within a public organization or government structure, which aims to guarantee public servants are properly constrained in accordance with the wishes of their superiors.

Second, unlike hierarchical accountability systems, legal accountability systems involve external controls and the scrutiny of government units and their activities. Two relatively autonomous units, namely the government unit⁹ and its legal overseer, are involved in legal accountability systems based on a principal–agent agreement. Legislative oversight hearings, fiscal audits, and court reviews of administrative practices are typical examples of this type of accountability. In addition, external peer review can be another example of legal accountability (see Romzek & Ingraham, 2000, p. 245: review of pilot’s knowledge and skills by external pilots, called “black hats”). Legal accountability is highly compatible with Mulgan’s (2000b, 2003) “accountability as control” as well, but in an external sense to outside agencies like legislatures. Both legal accountability and accountability as control (in the external sense) generally occurs at the later stages of evaluation and correction in the “policy-cycle” literature. Although legal accountability is an ex post facto measure, it can influence individual behavior beforehand. That is, every individual acknowledges that any action (or inaction) he or she takes could potentially become the target of disciplinary investigation by, for example, review boards containing fellow professionals.

Third, in professional accountability systems, individuals are expected to base their decision making or judgments on internalized professional skills and knowledge, where intimate external control is inappropriate. Therefore, government units trusting them to do the best job possible defer to their expertise and experience. In this respect, professional accountability for independent discretionary behaviors is being used to mean “the purely personal exercise of judgment and adherence to internalized standards” (Mulgan, 2000b, p. 560). Professional accountability, in turn, requires individuals to utilize the sense of “personal responsibility” or what Mosher (1982) called “subjective responsibility” which is based on the conscientious performance of duties (Harmon & Mayer, 1986, p. 49; Hart, 1968, pp. 212-214; Mulgan, 2000b, p. 560). There is an ethical obligation on the individual to explain and justify his or her decisions and actions honestly as well as openly (Gregory, 2003, p. 560). Sanctions are also inward in that the individual faces his or her own sense of professional malfeasance or guilt (Mulgan, 2000b, p. 560). Thus, for individuals in the process of accommodating professional accountability, professional norms, such as standards specific to the job in question, are inseparable from general moral values, such as honesty, integrity, and fairness.

Finally, political accountability involves relationships between a government representative and his or her key stakeholders. That is, being responsive to concerns of key external constituents such as the general public, elected officials, interest groups, and even future generations is a key feature of political accountability systems. While political accountability is relevant to all personnel, it plays a more dominant role at higher levels of authority (Romzek & Ingraham, 2000, p. 245). Political accountability is conceptually identical to so-called “accountability as responsiveness” (Mulgan, 2000b, pp. 566-569) in which public agencies and officials are supposed to be responsive to their political stakeholders and the public. While hierarchical and legal accountabilities stress the coercive role of internal and external pressure respectively, political accountability (or accountability as responsiveness) points more broadly to the government units’ general compliance with popular demands.¹⁰

Accountability Relationships in Conflict

Individuals work under cross-pressures within a web of accountability relationships that include different behavioral standards against which their effectiveness can be evaluated (Romzek & Ingraham, 2000). Thus, expectations from different accountability relations are frequently in conflict, and in turn, individuals often resort to one accountability relationship while neglecting others. In particular, hierarchical and political accountabilities become more stressed than professional accountability, especially when events go wrong (Gregory, 2003; Romzek & Dubnick, 1987, p. 562). For example, whether individuals have obeyed commands frequently takes precedence over technical considerations as whether their decisions reflect sound exercise of discretion (professional accountability) is often downplayed (Romzek & Ingraham, 2000, p. 250). The perspectives of professionals and their moral choices are often ignored or overridden by top-down directives due to the executive-branch distrust of their judgment and autonomy (Radin, 2011, p. 109; Gregory, 2003, p. 566). Similarly, managers who abide by professional standards—“I used my best judgment”—may confront sanctions resulting from failure to follow hierarchical standards—“You didn’t follow the rules” (Romzek & Ingraham, 2000, p. 243). On the other hand, the pursuit of personal and professional values is sometimes linked with reducing political and hierarchical accountability and establishing more trust in the independent discretion of public servants (Sinclair, 1995, p. 233).

Hierarchical accountability can often contradict political accountability in that complying with organizational rules may prevent a public official from giving his or her clients what they want, or may encourage safety at the expense of production results and low costs. These are major concerns for elected officials and interest groups. On the contrary, it is also natural for officials who are under chronic resource strains and time pressure for rapid economic growth to focus on production results rather than to comply with rules and regulations. In this respect, one of the main aims of the recent reform movement (i.e., new public management) has stressed political accountability relative to hierarchical accountability. That is, “client focus,” the major catch phrase of the movement, encouraged public officials to be more responsive to the needs of citizens and less concerned with following the instructions of their bureaucratic superiors or deferring to a set of procedures (Mulgan, 2000b, p. 568). In addition, as the interests of stakeholders vary (i.e., due to political pluralism), different political accountability relationships can be in conflict with each other—for example, the preferences of ruling and opposition parties to which public managers are supposed to respond are generally heterogeneous.

In summary, governments operate under one or two types of accountability relationships on a daily basis, but have to cope with all four types in times of crises because all types are typically triggered in such situations. Thus, problems of accountability become wicked—messy, complex, and paradoxical (Gregory, 2003, p. 566)—especially during crises. In turn, conflicting and overlapping accountability demands may result in the overall frustration of individual accountability relationships (Mulgan, 2003, p. 108). The different accountability pressures result in public officials playing one side off against the other, remaining only partly (not fully) accountable. However, it is also important to note that political accountability is the driving force behind all types of accountability, including professional accountability, in that it is imperative for government units to respond to demands from their ultimate superior—the general public—under contemporary democracies.¹¹ Accordingly, the activity of accident investigation boards (legal accountability) may reflect an attempt to be responsive to the public (political accountability) by taking steps to assure such a misfortune does not occur again.

Case Analysis

Methodology

The assessment of four types of accountability relationships is straightforward. First, the issue of hierarchical accountability relationship will be identified when a member of the Nuclear Emergency Response Headquarters (NERH) or the Government-TEPCO Integrated Response Office (GTIRO)—see “advisory configurations” section below—complies or does not comply with the chain of command. The issue of hierarchical accountability relationship will be also recognized when a component of the advisory configurations follows relevant rules. Second, the problem of professional accountability relationship will be considered when decisions are made by a member of the advisory configurations based, or not based, on expert knowledge or professional consciences. Third, the issue of political accountability will be assessed when a component of the advisory configurations considers or actually responds to the (potential or practical) interests of its external stakeholders. Finally, the matter of legal accountability relationship will be recognized when the conduct of the advisory configurations during acute stages of the Fukushima disaster is reviewed by their legal overseer, namely the NAIIC.

The case analysis is based on published inquiries, monographs, media coverage, investigative reports, and scholarly judgments. In this regard, the data used in this study are close to secondary data as they were collected for other purposes. Although interviews with important actors, such as Prime Minister Kan and former TEPCO CEOs, in the Fukushima disaster could be tailored to the purpose of this research, secondary data analysis has the merit that can greatly reduce researcher’s costs (O’Sullivan & Rassel, 1995, p. 236). Thus, the analysis of secondary data enables researchers to conduct studies which are otherwise infeasible (O’Sullivan & Rassel, 1995, p. 236). To take a balanced perspective, this study used two different data sets. That is, this research analyzed not only media coverage and the Diet’s official investigation report that focus on mainly dysfunctional behaviors by government and TEPCO, but also sources that may provide alternative perspectives, such as Prime Minister Kan’s memoirs and monologue of top bureaucrats.¹² The analysis applies to events from the onset of the Fukushima nuclear plant incident on March 11, 2011, until Prime Minister Kan’s resignation speech on August 26, 2011. The events during this period merit close analysis because they can shed light on various stages of the crisis: the acute stage (March 11-19, 2011), the subsequent political battle, and nuclear energy policy reforms by the incumbent leader.

Advisory Configurations

According to ‘t Hart et al. (2009, p. 479), advisory configurations refer to “the cluster of formal and informal advisors, as well as the relevant components of their ‘back offices’ who on a de facto basis provide counsel to a leader during the immediate response phase of an emergency episode.” Advisory configurations of Prime Minister Kan were composed largely of two ad hoc institutions: the NERH and the GTIRO.¹³ The NERH was established following the prime minister’s declaration of a nuclear emergency at 19:03 on March 11, 2011, at the prime minister’s official residence (Kantei). The Nuclear Emergency Preparedness Act (NEPA) mandates that the prime minister create the NERH to cope with a nuclear emergency situation. Key members of the NERH included the minister of economy, trade, and industry (Kaieda, vice director-general), the chief cabinet secretary (Edano), the deputy chief cabinet secretary (Fukuyama), the deputy chief cabinet secretary for crisis management (Ito), and the director-general of the NISA (Terasaka) as well as the fellow/TEPCO vice president (Takekuro, dispatched from the company to advise the NERH). These key members were chaired by Prime Minister Kan (director-general). In addition, under these key members of the NERH there were many practical teams, including an informal technical team composed of alumni from Prime Minister Kan’s alma mater.

Second, Prime Minister Kan established the GTIRO at TEPCO’s headquarters on March 15, 2011, with himself as chairman. Before establishing this office, Kan felt that the company was not responsive to the NERH (i.e., Kantei) in terms of the management of the Fukushima crisis. He wanted to put TEPCO under government control by stationing Kaieda, the minister of METI, as one of the two vice chairmen (with TEPCO’s president, Shimizu, as the other vice chairman). Therefore, the GTIRO was a subordinate unit of the NERH regarding the Fukushima crisis management. In short, both the NERH and the GTIRO were more heterogeneous than cohesive because members of each ad hoc institution simply lacked the time to create a reasonable level of rapport (‘t Hart, 1990, p. 140). In particular, members of the NERH came from the METI, the NISA, and TEPCO and were inherently heterogeneous to those at the official residence. In addition, nuclear experts from these organizations worked implicitly for the interests of the “nuclear power village,” and members of the NERH and the GTIRO were in turn close to Heclo’s (1977) notion of “a government of strangers.”

Accountability in the Fukushima Disaster

When the great earthquake occurred at 14:46 on March 11, 2011, on the northeastern side of Japan, Prime Minister Kan was attending a committee on accounts in the Upper House of the Diet to answer questions from opposition party members. Consequently, all primary members of Kantei and the NERH were fully staffed and at work in Tokyo. However, on the day of the incident, two of the top TEPCO executives were out of town: Masataka Shimizu, the president, was in Osaka, and Tsunehisa Katsumata, the chairman, was in China.

Hierarchical accountability

The NERH command structure mirrors the typical hierarchical accountability relationship in that it focuses on the chain of command and the expectation that rules (i.e., the NEPA) and top-down directives are foremost considerations. The director-general of the NERH (prime minister) holds strong authority, including the power to direct not only government agencies but also operators of nuclear plants (i.e., TEPCO) when necessary. It appears that the members of the NERH were, at least at the outbreak of the Fukushima nuclear accident, well aware of the primary manuals for a nuclear power crisis: the NEPA. At 15:42 on March 11, 2011, immediately after the waves of two gigantic tsunamis hit the Fukushima Daiichi nuclear plant, TEPCO reported to the NISA based on Article 10 of the NEPA that a “specific event” occurred. At 16:45, TEPCO further reported to the regulatory agency based on Article 15 of the NEPA that the water injection into the emergency core cooling system was impossible in two units of the Fukushima Daiichi nuclear power station. In response, Kaieda, the minister of the METI (the overseer of the NISA) reported these events to Prime Minister Kan and requested the issuance of a declaration of a nuclear emergency situation (DNES) and the establishment of the NERH. Prime Minister Kan then announced the DNES to the public at 19:03 and established the NERH at Kantei. However, it took more than 2 hr for the government to issue the DNES after being informed by TEPCO about the emerging situation, which clearly limited an agile government response. This was because the prime minister was very inquisitive about the situation and Kaieda had to answer all questions in detail. Prime Minister Kan had a tendency to figure out what was happening at the accident site in detail and control it. He believed that a leader must check the site directly and make decisions based on this information, which was relevant for political accountability, particularly the DPJ’s party platform, called “seiji shudo.” In other words, Kan’s pursuit of the party platform (i.e., political accountability) delayed the issuance of the DNES (i.e., the completion of hierarchical accountability).

In the same vein of the DPJ platform, Prime Minister Kan arrived at the site by helicopter at 07:12 on March 12. Kan then asked the vice president of TEPCO, Sakae Muto, why the ventilation of the first reactor was delayed. However, the vice president mumbled without offering a specific explanation.¹⁴ With this response, Kan lost patience and shouted at him, which clearly signaled the government’s centralized leadership. Furthermore, the prime minister’s direct involvement in the management of TEPCO—the pursuit of political accountability—confused the existing chain of command—hierarchical accountability—under which field employees were governed by bypassing the middle level (i.e., government bureaucrats). In turn, one employee of the firm complained that “Kantei forces us to report everything. But, originally we were supposed to report solely to the NISA” (Asahi Shinbun, 2011, March 16a).

According to NEPA requirements, an offsite center such as the local nuclear emergency response headquarters must be established near the site of the nuclear accident. However, this requirement did not take into account the coexistence of multiple emergencies from earthquakes and tsunamis, among others. Therefore, this agency did not function at all until at least 5 days after the outbreak of the nuclear accident. The general director of the offsite center (the vice minister of the METI) and other supporting members from the local government could not arrive at the site because of the traffic jam caused by the earthquake and subsequent tsunami. As a result, the local nuclear emergency response headquarters was established at the Fukushima prefectural office on March 15, 2011. As a result of the paralysis of the offsite center during the first several days, there was a long delay before the government could grasp the status of the accident. That is, the NERH’s preoccupation with (flawed) manual—hierarchical accountability—delayed the establishment of the offsite center, and subsequently, its initial response. The prime minister should have acted more resiliently by establishing the offsite center at an alternative site at the outset to avoid any difficulty in information gathering and crisis decision making.¹⁵

As another example, in the middle of pumping seawater into a nuclear reactor at the Fukushima Daiichi power plant as a last resort to prevent it from melting down on March 12, the TEPCO headquarters ordered the chief of the Fukushima nuclear plant, Yoshida, to stop pumping because they did not get approval from the prime minister (Imanishi & Asahi Shinbun Publications, 2012, p. 51). Fortunately, Yoshida, who was faithful to his professional expertise in the acute situation, only pretended to comply with the order and continued pumping. This incident reflects the inflexible hierarchical structure caused by Kan’s centralized leadership that could have disrupted the timely frontline responses during the national crisis—TEPCO headquarters’ compliance with the hierarchical accountability might have compromised Yoshida’s pursuit of professional accountability.

Professional accountability

Relationships of professional accountability were carried out to the extent that the government deferred to nuclear expertise or technocrats’ judgments. Due to the highly technical nature of nuclear power, the government recognized the need for nuclear experts in managing the Fukushima disaster. However, Prime Minister Kan was not equipped with reliable technical advisory mechanisms at the NERH. The director-general of the NISA, Terasaka, was expected to provide technical advice to Prime Minister Kan during the acute stages of the Fukushima disaster, but he could not do so (Kan, 2012, p. 17) because of his lack of nuclear expertise (i.e., he was a generalist who majored in economics at the University of Tokyo).¹⁶ Perceiving the top officials to be unreliable, Prime Minister Kan assembled an informal technical team consisting of alumni from his alma mater (Tokyo Institute of Technology) to provide him with necessary information and advice.¹⁷

Furthermore, the NISA was not technically prepared for addressing the nuclear power plant accident as it was an agency established for merely inspecting safety at nuclear plants and other energy-related facilities. Similarly, other agencies such as the Japan Self-Defense Forces (JSDF), the Fire and Disaster Management Agency (FDMA), and the National Police Agency (NPA) were not equipped with any administrative function directly related to nuclear accidents. For example, it was the Central Nuclear Biological Chemical Weapon Defense Unit (CNBC) of the JSDF that took an active role in addressing the Fukushima disaster, although this unit was intended to deal with nuclear attacks, not nuclear accidents. In addition, immediately following the gigantic tsunami that hit the Fukushima Daiichi nuclear plant, it lost all power. Therefore, a power supply car was urgently needed to cool the nuclear reactor and thus avoid a meltdown. The NERH attempted to transport power supply cars from various places to the accident site. At around 21:00 on March 11, 2011, it was reported to the headquarters that a power supply car had arrived at the plant site. However, the car was not helpful because its plug did not fit the nuclear plant. This clearly indicates the NERH’s lack of technical expertise in coping with nuclear power accidents. In short, Prime Minister Kan and his advisory configurations such as the NERH and other supporting agencies were not supplied with reliable technical expertise to which they could defer to deal with the nuclear disaster—professional accountability relationships.

While it is true that Kan was not under favorable conditions to accommodate professional accountability expectations, he himself (perhaps unconsciously) impaired such expectations being accountable to the DPJ’s party platform—meeting the political accountability relationship. As mentioned earlier, Kan arrived at the nuclear accident site in the morning of March 12. However, his direct supervision interrupted, and even confused, field operations. The chief of the Fukushima Daiichi nuclear plant, Masao Yoshida,¹⁸ later mentioned that “We could not cope with the accident for about two hours because I was attentive to meeting the Prime Minister when he visited the site on March 12” (Imanishi & Asahi Shinbun Publications, 2012, p. 39). The prime minister’s centralized leadership obstructed the settlement efforts in the field, and his professional accountability relationship with it.

With the escalation in radiation exposure at the Fukushima nuclear plant site, the NERH ordered the evacuation of residents within a 3 km radius of the site at 21:23 on March 11, 2011. Soon after this order, Masao Yoshida, the chief of the plant, reported to the NERH that the first unit was in a very serious condition. Therefore, the headquarters expanded the evacuation area from a 3 km radius to a 10 km radius at 5:44 on March 12. However, after an explosion at the first unit of the Fukushima nuclear plant at 15:36 on March 12, the evacuation zone was further widened to a 20 km radius at 18:25 on the same day. These frequent expansions of the evacuation zone were particularly confusing to residents who were within the 3 km radius because they had to move three times. Later, Prime Minister Kan acknowledged the burden imposed on these residents (Kan, 2012, p. 83). These frequent changes in the evacuation zone reflected the unreliability in professional judgments of the headquarters. However, a critical information-gathering system for obtaining such judgments was not available until March 15, 2011. In other words, the headquarters did not know about the existence of the System for Prediction of Environmental Emergency Dose Information (SPEEDI), which is used not only for the measurement of radioactive releases but also for weather and topographical data to predict how far radioactive substances can travel after being released into the atmosphere. The decision makers at Kantei could not use this sophisticated system to set the evacuation zone for at least 5 days following the outbreak of the accident, which implies that the evacuation zone was arbitrarily set. In consequence, uneasiness (eventually distrust in the government) spread across the rescued. A refugee stated that “Everyone (thousands of people) here in the shelter are frightened and anxious because nuclear radiation is invisible” (Asahi Shinbun, 2011, March 14).

Unfortunately, the anxiety of refugees became reality—areas outside the 20 km radius of the site and additional evacuation zones were not radiation-free. On March 15, excessive levels of radioactivity were measured at locations beyond the 20 km radius of the plant (Asahi Shinbun, 2011, March 15).¹⁹ This indicates that the evacuation zone set by the government was based not on reliable scientific knowledge or information, but on other factors such as political considerations. The government defined the zone in a very conservative manner because of a wider zone potentially inducing panic among local residents and thus escalating the public’s criticism of politicians. In summary, although it is absurd that the NERH did not recognize the existence of the critical equipment, SPEEDI, at the onset of the nuclear disaster, the government might have gradually and consciously widened the evacuation zone (without the basis of objective information) so that a potential nation-wide chaos could be averted. In this regard, the government’s concern about the political accountability relationship with the public restrained itself from being accountable to science-based professionalized decisions on establishing the evacuation zone.

Globally, various governments and organizations, particularly the U.S. government, responded to professional accountability expectations. On March 17, the U.S. government officially urged American citizens living within 80 km from the plant to evacuate (Asahi Shinbun, 2011, March 17a). The government stressed that the 80 km zone was based on a careful scientific evaluation, and guidelines of the U.S. Nuclear Regulatory Commission (NRC) and the Department of Energy. Following the U.S. recommendations, the governments of South Korea, the United Kingdom, Australia, New Zealand, and Mexico recommended their citizens within an 80 km radius from the disaster site to evacuate (Asahi Shinbun, 2011, March 18). The German government even instructed its citizens living in Tokyo and Yokohama to take refuge in more distant areas such as Osaka and other countries (Asahi Shinbun, 2011, March 17b). Consequently, the professional accountability of the Japanese government in establishing the evacuation zone was substantially invalidated by foreign governments’ recommendations.

Political accountability

Political accountability relationships entail expectations and concerns of key stakeholders and responsiveness to them (Romzek & Ingraham, 2000, p. 245). Whether management members during the Fukushima crisis were responsive to these stakeholders was closely linked to opportunities for their future promotions (Romzek & Ingraham, 2000, p. 245). Therefore, individuals in advisory configurations of Prime Minister Kan sought to avoid situations in which they would have to inform external stakeholders that they were unable to meet their assigned goals. For these advisory configurations, there were three key stakeholders, namely the public, party politicians, and the nuclear power village. The distribution of accurate information to the public is essential in crisis management, but the government did not follow this fundamental rule at the critical moment (Nakamura & Kikuchi, 2011, p. 895). Soon after the explosion and white smoke were seen at the first unit of the Fukushima nuclear plant at 15:36 on March 12, Edano, the chief cabinet secretary, held a press conference to explain the situation to the public. In his press conference, Edano did not clearly define the incident by using conclusive words such as “explosion,” instead framing it ambiguously as an “explosive event” (Asahi Shinbun, 2011, March 13b). To make matters worse, the NISA kept repeating the statements “We are checking it,” “Take refugee just in case,” and “There is no immediate risk to health” without providing exact information on the situation at the time (Asahi Shinbun, 2011, March 13a). This led people to believe that the government was concealing critical information regarding the incident. In fact, the cabinet was hesitant to release information on the accident, and even censored it. For example, immediately following the explosion at the plant (but before the NISA’s ambiguous information, as indicated earlier), a press secretary at the NISA publicized that it caused a critical meltdown (Nakamura & Kikuchi, 2011, p. 895). However, the NERH (i.e., Kantei) held that the secretary’s announcement was premature and misleading, indicating that no meltdown had occurred.²⁰ Since the announcement, the press secretary no longer appeared in public, sparking a rumor that he was transferred to another section. An official of the NISA commented that “We cannot make the decision to release information ourselves. We are waiting for the cabinet’s approval” (Asahi Shinbun, 2011, March 16a). This indicates that political accountability (i.e., reporting exact information) to the public was prevented by hierarchical accountability to the cabinet.

The prime minister did not issue an apology for inconveniencing the public in his first press conference, which was held a couple of days after the occurrence of the nuclear disaster. Instead, he illustrated that the Fukushima nuclear disaster was due to the gigantic tsunami far beyond what could be imagined. This statement further triggered severe criticism from the media and public. The public interpreted his statement as an excuse for the man-made disaster. The only apology was made during his final press conference on August 26, a week before his resignation. He stated that “I feel very sorry for the insufficient information provided to the public, the ineffective communication of my thoughts to the people, and the fact that things did not go smoothly due to the twisted Diet during the national crisis” (Kan, 2012, p. 185). However, his apology was too late to gain public sympathy.²¹ Thus, it appears that he failed to meet political accountability to the public.

The party system is one of the key objects of political accountability, particularly for the office of the prime minister. This office carries a greater obligation to the party arrangement than the office of the president because of the dependence on political support from parliamentary allies to remain in power (Masters & ‘t Hart, 2012, p. 761). Therefore, it was necessary for Prime Minister Kan to faithfully follow the party platform (DPJ): governance led by politicians (so-called “seiji shudo”). Kan’s responsiveness to this platform appeared as centralized political leadership and top-down directives; that is, a political accountability relationship. However, Kan’s leadership was not effective, as illustrated earlier (see hierarchical and professional accountability sections above). His political accountability to the DPJ platform negatively affected not only hierarchical but also professional accountability relationships.

The prime minister’s commitment to the party platform also did not guarantee the party’s political support for his leadership. When the leadership is negatively perceived by the public, it is a good opportunity for rival politicians to levy blame against the incumbent government. With the rise in the level of public distrust in the Kan administration during his crisis management of the Fukushima disaster, it was reported in the middle of May 2011 that Japan’s main opposition party (LDP) considered submitting a no-confidence motion against Kan. Immediately after this report, the public was also informed that Ichiro Ozawa, a rival political leader of the DPJ, was considering backing the motion if put forward by the LDP (Reuters, 2011, May 27). Dissenters such as Ozawa within the ruling party supported the motion because they worried that Kan’s poor ratings would undermine their chances of winning future elections. In this regard, the ruling party closely resembled “Groupthink Type I” in that, in the face of a perceived threat, fringe actors such as Ozawa and his followers attempted to disassociate themselves from decisions of the cabinet (i.e., the NERH; ‘t Hart, 1990, p. 192, 202). However, the more relevant reason resided in “a deepening rift” between Kan and Ozawa. The rift was evident ever since Kan defeated Ozawa in the DPJ presidential election in September 2010 and excluded him and his supporters from key cabinet posts, taking an “anti-Ozawa path” (The Japan Times, 2011, June 2). On June 2, the motion was rejected because of opposition from other DPJ members. Perceiving unstable political support within his party, particularly during the national crisis, Prime Minister Kan might have considered a coalition cabinet. According to a media report issued on May 19, 2011, Kan requested the president of the LDP, Tanigaki, to join in the crisis management cabinet. However, Tanigaki rejected the idea, criticizing it as being Kan’s hidden intention to prolong the life of the cabinet (Kan, 2012, p. 137). In summary, once the prime minister’s political accountability to the public deteriorated, his political accountability to the other stakeholders such as ruling and opposition party members became helpless.

Despite its invisibility and implicitness, the nuclear power village was one of the most influential stakeholders in political accountability relationships during the acute stages of the Fukushima disaster. It was Yoshida’s idea (the chief of the plant) to inject seawater into the nuclear reactor of the Fukushima plant, the last resort to avoid a meltdown. However, the village members (e.g., those at the headquarters of TEPCO and the NISA) were more concerned about economic interests of the nuclear industry and their jobs and thus were hesitant to fully support Yoshida’s idea. In fact, pouring seawater into the reactor meant decommissioning it (or losing $2.5 billion) because the salt in seawater would render it unusable. Even Prime Minister Kan appeared cautious about adopting this approach. He asked Haruki Madarame, the chairman of the NSC, whether it could cause “recriticality.²²” Surmising Kantei’s attitude concerning the last measure, the headquarters of TEPCO (i.e., the GTIRO) ordered Yoshida to stop pumping, as mentioned earlier. However, the headquarters’ true intention was to pursue economic interests of the village, not to rescue the nation from the catastrophe. In short, the NERH (particularly those at the headquarters of TEPCO and the NISA) tacitly pursued political accountability to the nuclear power village (i.e., its economic interests), which could have interrupted Yoshida’s urgent effort to avoid a meltdown—his professional accountability.

Furthermore, being responsive to the culture of the village, executive members of TEPCO (at the GTIRO) undermined the government’s political accountability to the public. The village members’ responsiveness to their norm, that is, secrecy, was closely associated with their future career prospects. Although an enormous amount of status information was reported from the nuclear plant site to the headquarters each day, only 1% to 10% of it was communicated to the public (Imanishi & Asahi Shinbun Publications, 2012, p. 94). An executive member at TEPCO headquarters later mentioned that “[i]f all the information about the site was reported to the government, it would become a great trouble. . . . What about my position and career advancement? Do you understand?” (Imanishi & Asahi Shinbun Publications, 2012, p. 95). Partly because executive members of TEPCO hid the truth from the cabinet, the NERH could not deliver details of the accident to the public, undermining its political accountability. In this regard, political accountability to the public (i.e., providing them with critical information) and to the nuclear power village (i.e., keeping secrecy) were in conflict.

Legal accountability

Legal accountability relationships are usually carried out when accidents occur or when charges of inappropriate behaviors are filed (Romzek & Ingraham, 2000, p. 245). Prime Minister Kan and his advisory configurations including the NERH and the GTIRO were under the expectations of legal accountability relationships to the extent that their conduct during acute stages of the Fukushima disaster was reviewed by an external task force, namely the NAIIC. Oversight activities of the NAIIC represented the accommodation of the legal accountability of Kan and his formal, as well as informal, advisory configurations. On October 30, 2011, the NAIIC Act was enacted, creating an independent commission composed of a chairman and nine other members appointed by the National Diet. With its investigative powers, the NAIIC investigated the background and causes of the Fukushima disaster through more than 900 hr of hearings and interviews with 1,167 people (Kurokawa et al., 2012). In particular, it organized a total of 19 Commission Meetings in which a number of witnesses, including Prime Minister Kan and his primary advisors (e.g., Edano, Kaieda, Terasaka, and executive members of TEPCO), were in charge of investigative hearings. In its official report, the chairman of the NAIIC (Kiyoshi Kurosawa) concluded that the Fukushima accident “cannot be regarded as a natural disaster. . . . Its effects could have been mitigated by a more effective human response” (Kurokawa et al., 2012). To ensure that a disaster like Fukushima would never be repeated, the NAIIC made a series of recommendations, including the following (Kurokawa et al., 2012, pp. 22-23):

A crisis management structure of the government “must be established with a consolidated chain of command.”

Existing rules should be “rewritten in order to meet global standards of safety,²³ public health, and welfare.”

A new regulatory organization (which is supposed to resolve any conflict of interest between the regulation and promotion of nuclear power) “should keep up with the latest knowledge and technology.”

“Information should be disclosed with public health and safety as the priority, instead of government convenience.”

The decision-making process of the new regulatory body “should exclude the involvement of electric power operator stakeholders.”

“The government must set rules and disclose information regarding its relationship with the operators.”

“New rules must be created that . . . set criteria to determine whether reactors should be decommissioned.”

Recommendations 1 and 2 aim to reform the “chain of command” and “rules,” respectively, in terms of the hierarchical accountability relationships (# 19 and 20²⁴). Recommendation 3 enhances the conditions of professional accountability (# 21). Furthermore, Recommendation 4 improves the political accountability relationships with the public (# 22). By contrast, Recommendations 5, 6, and 7 aim to break troublesome political accountability relationships with interest groups (i.e., the nuclear village: # 23). Consequently, the outcomes (i.e., recommendations) of the legislative investigation by the NAIIC address problems associated with the three accountability relationships (hierarchical, professional, and political accountability) during the crisis.

Discussion

This section discusses interrelationships between hierarchical, professional, and political accountability expectations, which will be summarized in schematic form (see Figure 2). First, it is important to elucidate factors that negatively influenced hierarchical and professional accountability expectations. Prime Minister Kan’s faithful following of the DPJ party platform (strong political leadership) and the emergency manual (NEPA) institutionalized highly centralized decision making reflecting his micromanagement and direct control of field operations (# 1 and 2 respectively). However, his centralized leadership—indication of his political accountability to the party platform—not only confused the NERH’s hierarchical accountability (i.e., broke the chain of command and delayed the issue of DNES: # 3), but also obstructed its professional accountability (i.e., his visit to the field delayed Yoshida’s efforts to cope with the accident: # 4).

Figure 2.

The interrelationships between hierarchical, professional, and political accountabilities.

Had Yoshida’s professional commitment (i.e., reliance on expertise and morality) not overridden the order to stop injecting seawater into the reactor (i.e., TEPCO headquarters’ compliance with hierarchical accountability and their implicit pursuit of political accountability to the nuclear village), this might have undermined their professional accountability to the field’s settlement efforts (the dotted # 5 and 6 respectively). In addition, the NERH’s preoccupation with hierarchical accountability to the flawed manual (NEPA), which appeared as the delayed establishment of the offsite center, suspended its professional accountability (i.e., initial response to the nuclear disaster: #7). The NERH’s quick resort to hierarchical accountability in compliance with rules reflects the administrative reality which stresses a risk-averse, rule-oriented approach to accountability during a crisis (Romzek & Ingraham, 2000)

Furthermore, TEPCO’s political accountability to the secrecy of the nuclear village led to concealing critical information from the cabinet, which in turn obstructed the NERH’s professional accountability in regard to establishing appropriate evacuation zones for affected residents (# 8). However, the NERH’s apprehension about maintaining political accountability to the public in the face of national crisis (in turn, concern about potential social disorder) also contributed to compromising its professional accountability—the invalid evacuation zones (# 9).

Second, it is necessary to illustrate how the government’s political responsiveness to the public deteriorated during the crisis. Prime Minister Kan’s command and control leadership style prevented bureaucrats from releasing important information on the status of the nuclear plant to the public without his permission. In other words, his establishment of the top-down hierarchical structure interrupted the cabinet in its smooth political accountability relationship with the public (# 10). Further, his micromanagement reinforced TEPCO’s secrecy such that they concealed critical information about the plant status from the cabinet. Without specific information, the NERH had no option but to make ambiguous (or general) statements to the public. These facts indicate that TEPCO’s political accountability to the village (i.e., secrecy) was strengthened by its hierarchical accountability to the prime minister’s micromanagement in which they did not report negative and critical information about the field due to the fear of repercussions (# 11). As a result, TEPCO’s political accountability to the interest group contributed to the government’s ambiguous public statements—the cabinet’s ineffective political accountability relationship with the people (# 12). In this regard, TEPCO and the cabinet served heterogeneous stakeholders, the nuclear village and the public, respectively. That is, being responsive to one stakeholder, namely the nuclear village, weakened accountability expectations from the other stakeholder, namely the public.

In addition, government officials’ (particularly the director-general of the NISA) lack of expertise and technical arrangements in dealing with the nuclear accident, which undermined their professional accountability, disturbed them in having the harmonious political accountability relationship with the public by releasing ambiguous information about the field (# 13). The lack of specific government information and no apology from Prime Minister Kan resulted in widespread public distrust of the government—deteriorated the government’s political accountability relationship with the public. Other factors such as a broken chain of command, the delayed establishment of the offsite center, the field’s slow response, and frequent changes in the evacuation zone further reduced the level of trust in the government. That is, negative impacts on both hierarchical and professional accountability relationships of the government worsened its political accountability relationship with the public (# 14, 15, and 16).

Under the heightened level of public distrust in the government, the prime minister’s efforts to be responsive to his party platform and to invite the opposition party to the cabinet’s crisis management failed. Under this circumstance, politicians, particularly in rival or opposition parties, used it as an opportunity to show their ability to control the incumbent government, “to demonstrate their toughness as a countervailing power,” to inflict damage upon the opposing camp, and to bring about some cleavage within the ruling party (Boin, ‘t Hart, Stern, & Sundelius, 2005, pp. 108-109). The nonconfidence motion against the Kan cabinet by the LDP and Ozawa, a dominant rivalry politician in the ruling party is a representative example. Another example includes the president of the LDP, Tanigaki, criticizing Kan’s invitation to join the crisis management cabinet on account of his hidden intention to extend the life of the cabinet.²⁵ In short, it might be quite natural to observe that the prime minister’s crumbled political accountability to the public—ultimate political superior—could not invite support from their surrogates—the ruling and opposition parties (# 17 and 18, respectively). In other words, Kan and his cabinet’s efforts to build political accountability relationships with both parties failed.

Figure 2 not only simplifies but also integrates what this section has discussed, which focuses on the predominantly negative interrelationships between hierarchical, professional, and political accountabilities. The numbers in Figure 2 are identical to those in parentheses illustrated so far. The thickness of each arrow indicates the frequency influence from one accountability to the other. In Figure 2, hierarchical accountability pressures appear to have negative effects on political accountability relationships with the highest frequency. But, it should be noted that most of the negative effects on political accountability relationships originated from meeting the DPJ platform—political leadership (i.e., # 1). Further, a circular shaped arrow toward political accountability indicates the political accountability expectations have negative impacts on themselves. Likewise, hierarchical accountability expectations negatively influence themselves as well. Interestingly, political accountability relationship exert a negative influence on hierarchical accountability expectation with the least frequency. Moreover, it is noteworthy to mention professional accountability relationships do not have any impact on hierarchical accountability expectations. The least frequency influence on hierarchical accountability may imply the relatively inflexible (i.e., officially institutionalized) nature of its elements: rules and chain of commands.

In short, from the analyses of Figure 2, it is clear that this study fills gaps in the literature. That is, while previous research mostly treated professional accountability as a dependent variable analyzing the negative influences of hierarchical and political accountabilities on professional accountability (e.g., Romzek & Dubnick, 1987; Romzek & Ingraham, 2000), this study suggests a more integrated perspective that incorporates not only these influences but also interactions, particularly between hierarchical and political accountabilities. In this regard, this research argues that all three accountabilities can be dependent variables. Nevertheless, it is obvious that the final goal of all accountability relationships is to satisfy political accountability expectations of the public—the ultimate political master.

Conclusions

It is clear that the Fukushima disaster occurred as a result of the nuclear village’s long-standing practices such as unpreparedness for nuclear crises, prioritizing economic interests, and use of secrecy. However, these first-order causes alone cannot explain why responses to the disaster during acute stages failed. This study’s results demonstrate how crisis management organizations of the government, including the NERH and the GTIRO, faced cross-pressures within a web of accountability relationships while dealing with the Fukushima nuclear disaster. Results indicate that various components of these accountability relationships undermine one another. It is the contribution of this research to specifically identify multiple and complex relationships between the types of accountability. Previous studies predominantly focused on limited interactions between types of accountability stressing how professional accountability can be compromised (Romzek & Dubnick, 1987; Romzek & Ingraham, 2000).

However, the more valuable contribution of this study may be in its discovery that economic (or production) pressures of political accountability (i.e., from the nuclear village) did not affect professional accountability relationships (although it could have done so regarding TEPCO headquarters’ order to stop injecting seawater into the reactor). This result is noteworthy in that the literature on crisis management attributes disasters in complex, high-risk systems to economic pressures, suggesting that these pressures undermine professional accountability relationships (e.g., Perrow, 1984; Romzek & Dubnick, 1987; Romzek & Ingraham, 2000; Vaughan, 1999). Although this argument generally holds, it may not always be the case. The Fukushima nuclear disaster case contrasts the Challenger case in which the pursuit of bureaucratic and political accountability detracted NASA from professional accountability. It is possible to find a professional accountability mechanism that is not susceptible to economic and political pressures, as exemplified by Yoshida’s heroic efforts to prevent the meltdown of the Fukushima nuclear reactor. It was clearly due to his unshakable belief in professional accountability (i.e., expertise and moral values) during the crisis. He resisted the act he considered to be seriously damaging to the public—TEPCO’s order to stop pouring seawater. His disobedience to the directive from TEPCO (hierarchical accountability) and the economic interests of the nuclear village (political accountability) deserves praise in that managers generally accept their supervisors’ order out of fear of the repercussions, hopes of maintaining favorable relationships, or personal advancement concerns (Mulgan, 2000b, p. 567). That is, organization members are often compensated for obedient conformity rather than critical objection (Gregory, 2003, p. 560).

Then, what factors led Yoshida to disregard the potential penalty from the TEPCO headquarters and individual interests, and pursue professional accountability? First, he used individual discretion to ignore the hierarchical directive, partly because he might think that TEPCO headquarters did not fully understand the challenges and conditions under which he and first-line employees experienced. Romzek and Ingraham (2000, p. 250) suggest that such circumstances frequently lead managers to perceive administrative superiors as part of the problem and to dismiss their directives as “bureaucratic red tape.” Ironically (but fortunately), the headquarters’ lack of understanding the field’s response to the accident seemed to invite Yoshida’s objection to the order from superiors.

Second, Prime Minister Kan had strong confidence in Yoshida. When he visited the site on March 12, 2011, Yoshida clearly explained the current situation of the first reactor, unlike executives of TEPCO. Yoshida even eased Kan’s concern by stating that “We will ventilate the first reactor by organizing a band determined to die.” Later, Prime Minister Kan stated that meeting with Yoshida was the most important achievement of his site visit and Kan deferred to Yoshida’s expertise. Although hard to prove, Yoshida might have disobeyed TEPCO’s order to stop pouring seawater partly because of his presumed rapport with the prime minister. This implies the crucial importance of delegating authority to the field during acute stages of crises.

Lastly, it might be possible he owed professional accountability not only to his professional peers, but also to the public which led him to act independently of the obligation to his superiors. In other words, his professional conscience made him answer directly to the public and being less constrained to the chain of command (Mulgan, 2003, p. 30, p. 144). In this regard, Yoshida’s professional accountability was closely linked to serving the public interest—political accountability to the public—at the time of national crisis.

The challenge then is to empower crisis managers who can exercise their reasonable discretion (professional standards), such as in the case of Yoshida, instead of merely following commands (hierarchical standards) during acute stages of a crisis. For the time being, it can be suggested for crisis managers not only to be situated in decentralized structures but also to be equipped with professionalism closely linked to public interests. Acknowledging the fact that far more abhorrent crimes have been committed under the name of obedience than under the name of rebellion (Gregory, 2003, pp. 560-561), there are no doubts that producing such intuitive and public-minded managers is of the utmost importance.

Footnotes

Appendix

Author Note

In memory of Theodore J. Lowi (1931-2017).

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biography

Yoonho Kim is a professor in the Department of Public Administration at the University of Seoul, Korea. He holds a PhD in government from the Department of Government at Cornell University. His research focuses on organization theory, administrative reform, and comparative study of Korean and Japanese public administration.

References

Asahi Shinbun. (2011, March 13a). Ho ahn in · toh den “kakunin chu” renppatsu [NISA·TEPCO, firing “We are checking”]. p. 3.

Asahi Shinbun. (2011, March 13b). Kiki kanri chigu hagu [Incoherent crisis management]. p. 3.

Asahi Shinbun. (2011, March 14). “Mie nai kara hu ahn” [“We are anxious about radioactivity because it is invisible”]. p. 17.

Asahi Shinbun. (2011, March 15). Kakuchi de ijyoh chi kan soku [Unusual measurements in various places]. evening edition, p. 1.

Asahi Shinbun. (2011, March 16a). Ho ahn in, nibusa medatsu [NISA’s blunts standing out]. p. 3.

Asahi Shinbun. (2011, March 16b). Toh den ‘in ppei tai shitsu’ myaku myaku [TEPCO’s continuous ’predisposition to hide’]. p.3.

Asahi Shinbun. (2011, March 17a). Bei, jikokumin e hinan kankoku [U.S. advises its citizens to evacuate]. evening edition, p. 3.

Asahi Shinbun. (2011, March 17b). Ei doku mo tai hi unagasu [U.K. and Germany urged their citizens to evacuate as well]. evening edition, p. 3.

Asahi Shinbun. (2011, March 18). 80 kiro gai kankoku kan ei gou nadomo [S. Korea, U.K. Australia and other nations also recommended that their citizens living within a 80 km from the plant should evacuate,]. p. 5.

10.

Blumenthal

(2005, September 4). Katrina comes home to roost. The Guardian. Retrieved from https://www.theguardian.com

11.

Boin

‘t Hart

(2003). Public leadership in times of crisis: Mission impossible? Public Administration Review, 63, 544-553.

12.

Boin

‘t Hart

McConnell

Preston

(2010). Leadership style, crisis response and blame management: The case of Hurricane Katrina. Public Administration, 88, 706-723.

13.

Boin

‘t Hart

Stern

Sundelius

(2005). The politics of crisis management: Public leadership under pressure. Cambridge, UK: Cambridge University Press.

14.

Bronner

(1982). Decision making under time pressure. Lexington, MA: D. C. Heath.

15.

Cohen

(1979). Threat perception in international crisis. Madison: University of Wisconsin Press.

16.

Colignon

R. A.

Usui

(2003). Amakudari: The hidden fabric of Japan’s economy. Ithaca, NY: Cornell University Press.

17.

Finer

(1941). Administrative responsibility in democratic government. Public Administration Review, 1, 335-350.

18.

Friedrich

C. J.

(1940). Public policy and the nature of administrative responsibility. In Friedrich

C. J.

Mason

E. S.

(Eds.), Public policy (pp. 3-24). Cambridge, MA: Harvard University Press.

19.

Gregory

(2003). Accountability in modern government. In Peters

B. G.

Pierre

(Eds.), Handbook of public administration (pp. 557-568). London, England: Sage.

20.

Guy Peters

. (1995). The politics of bureaucracy. New York, NY: Longman.

21.

Harmon

M. M.

Mayer

R. T.

(1986). Organization theory for public administration. Boston, MA: Little Brown.

22.

Hart

H. L. A.

(1968). Punishment and responsibility. Oxford, UK: Oxford University Press.

23.

Heclo

(1977). A government of strangers: Executive politics in Washington. Washington, DC: Brookings Institution.

24.

Imanishi

, Asahi Shinbun Publications. (2012). Fukushima genbatsu no shinzitsu: Sai ko kanbu no dokuhaku [The truth of Fukushima nuclear plant disaster: A monologue of a top bureaucrat]. Tokyo, Japan: Asahi Shinbun Publications.

25.

Kan

(2012). Toden Fukushima genbatsujiko soridaijin to shite kangaetakoto [Tokyo electronic power company’s Fukushima nuclear accident: What I thought as prime minister]. Tokyo, Japan: Gentosha.

26.

Kingston

(2012). Japan’s nuclear village. The Asia-Pacific Journal, 10(37), 1-29.

27.

Kurokawa

Ishibashi

Oshima

Sakiyama

Sakurai

Tanaka

. . . Yokoyama

(2012). The official report of the Fukushima nuclear accident independent investigation commission: Executive summary. Tokyo, Japan: The National Diet of Japan.

28.

Lodge

Wegrich

(Eds). (2012). Executive politics in times of crisis. Houndmills, UK: Palgrave Macmillan.

29.

Lowi

T. J.

(1985). The personal president: Power invested, promise unfulfilled. Ithaca, NY: Cornell University Press.

30.

March

J. G.

Olsen

J. P.

(1995). Democratic governance. New York, NY: The Free Press.

31.

Masters

art

(2012). Prime ministerial rhetoric and recession politics: Meaning making in economic crisis management. Public Administration, 90, 759-780.

32.

Mosher

F. C.

(1982). Democracy and the public service (2nd ed.). New York, NY: Oxford University Press.

33.

Moynihan

D. P.

(2012). Extra-network organizational reputation and blame avoidance in networks: The hurricane Katrina example. Governance, 25, 567-588.

34.

Mulgan

A. G.

(2000a). Japan’s political leadership deficit. Australian Journal of Political Science, 35, 183-202.

35.

Mulgan

(2000b). “Accountability”: An ever-expanding concept? Public Administration, 78, 555-573.

36.

Mulgan

(2003). Holding power to account: Accountability in modern democracies. Basingstoke, UK: Palgrave Macmillan.

37.

Nakamura

(2012). Asian model of government re-examined in the aftermath of the global economic crunch: A Japanese perspective from the experience of the triple disasters in March 2011. International Review of Administrative Sciences, 78, 239-259.

38.

Nakamura

Kikuchi

(2011). What we know, and what we have not yet learned: Triple disasters and the Fukushima nuclear fiasco in Japan. Public Administration Review, 71, 893-899.

39.

O’Sullivan

Rassel

G. R.

(1995). Research methods for public administration. New York, NY: Longman.

40.

Pempel

T. J.

(2000). Regime shift: Comparative dynamics of the Japanese political economy. Ithaca, NY: Cornell University Press.

41.

Perrow

(1967). A framework for the comparative analysis of organizations. American Sociological Review, 32, 194-208.

42.

Perrow

(1984). Normal accidents: Living with high-risk technologies. New York, NY: Basic Books, Inc., Publishers.

43.

Radin

B. A.

(2002). The accountable juggler: The art of leadership in a federal agency. Washington, DC: CQ Press.

44.

Radin

B. A.

(2011). Does performance measurement actually improve accountability? In Dubnick

M. J.

Frederickson

H. G.

(eds.), Accountable governance: Problems and promises (pp. 98-110). Armonk, NY: M. E. Sharpe.

45.

Radin

B. A.

Romzek

B. S.

(1996). Accountability expectations in an intergovernmental arena: The national rural development partnership. Publius: The Journal of Federalism, 26, 59-81.

46.

Reuters. (2011, May 27). Japan powerbroker Ozawa ready to challenge PM. Retrieved from http://mobile.reuters.com

47.

Rijpma

J. A.

van Duin

M. J.

(2001). From accident to disaster: The response to the Hercules crash. In Rosenthal

Boin

Comfort

L. K.

(Eds.), Managing crises: Threats, dilemmas, opportunities (pp. 143-154). Springfield, IL: Charles C. Thomas.

48.

Romzek

B. S.

Dubnick

M. J.

(1987). Accountability in the public sector: Lessons from the Challenger tragedy. Public Administration Review, 47, 227-238.

49.

Romzek

B. S.

Ingraham

P. W.

(2000). Cross pressures of accountability: Initiative, command, and failure in the Ron Brown plane crash. Public Administration Review, 60, 240-253.

50.

Rosenthal

(1986). Crisis decision-making in the Netherlands. Netherlands Journal of Social Sciences, 2, 2103-2129.

51.

Rosenthal

‘t Hart

Kouzmin

(1991). The bureau-politics of crisis management. Public Administration, 69, 211-233.

52.

Samuels

R. J.

(2013). 3.11: Disaster and change in Japan. Ithaca, NY: Cornell University Press.

53.

Sinclair

(1995). The chameleon of accountability: Forms and discourses. Accounting, Organizations and Society, 20, 219-237.

54.

‘t Hart

. (1990). Groupthink in government: A study of small groups and policy failure. Baltimore, MD: Johns Hopkins University Press.

55.

‘t Hart

Rosenthal

Kouzmin

. (1993). Crisis decision making: The centralization thesis revisited. Administration & Society, 25, 12-45.

56.

‘t Hart

Tindall

Brown

. (2009). Crisis leadership of the Bush presidency: Advisory capacity and presidential performance in the acute stages of the 9/11 and Katrina crises. Presidential Studies Quarterly, 39, 473-493.

57.

The Japan Times. (2011a, May 13). Reactor 1 in worse shape than thought. Retrieved from http://www.japantimes.co.jp

58.

The Japan Times. (2011b, June 2). No-confidence vote for Kan Cabinet. Retrieved from http://www.japantimes.co.jp

59.

The Japan Times. (2013, July 9). Man who battled Japan nuke disaster passes away. Retrieved from http://www.japantimes.co.jp

60.

Thomas

P. G

. (1998). The changing nature of accountability. In Peters

B. G.

Savoie

D. J.

(Eds.), Taking stock: Assessing public sector reforms (pp. 348-93). Montreal, Quebec, Canada: McGill-Queen’s University Press.

61.

Uhr

(1993). Redesigning accountability: From muddles to maps. The Australian Quarterly, 65, 1-16.

62.

Wolfenstein

V. E.

(1967). Some psychological aspects of crisis leadership. In Edinger

L. J.

(Ed.), Political leadership in industrialized societies (pp. 155-181). New York, NY: John Wiley.