Taking the Heat or Shifting the Blame? Ethical Leadership in Political-Administrative Relations After Policy Failures

Abstract

Ministers are faced with intense account-giving pressures when things go wrong under their watch. While ministers carry hierarchical responsibility for the behavior of their top administrators, ministers have incentive to shift blame and scapegoat them after policy failures. This is likely since political-administrative relations in most parliamentary systems have become more managerial. However, there is little empirical research regarding the formal and informal accountability rules that govern ministerial and administrative accountability after failures. This article compares two parliamentary systems with ministerial responsibility: the Australian State of New South Wales (NSW) and the Netherlands. NSW political-administrative relations have become more managerial than Dutch ones. The comparison highlights two important ethical constraints for ministers: (a) formal rules guiding political-administrative relations; and (b) informal norms regarding appropriate ministerial account-giving. While hierarchically oriented accountability rules and informal (parliamentary) norms buttress Dutch top administrators, these buttresses are less patent in the NSW case. Subsequently, while Dutch ministers have multiple times taken “sacrificial responsibility” for policy failures, NSW top administrators often have to front the media and have taken the fall for their ministers. To conclude, formal accountability rules and informal rules-in-use make ministers take the heat instead of shifting the blame.

Keywords:

• ethical leadership
• ministerial accountability
• policy failures
• political-administrative relations
• scapegoating

MINISTERIAL AND ADMINISTRATIVE ACCOUNTABILITY

When policy or bureaucratic failures occur in a minister’s portfolio, the minister and senior administrators in the department are under intense media scrutiny and political pressure to explain their behavior (Aucoin, 2012; McGraw, 1990). Public leaders, whether political or administrative, will set the tone for the civil service, either by setting a “good example” of how to behave responsibly or by managing down in a way that is appropriate (Heres & Lasthuizen, 2012). Ministers “need to be seen ‘doing the right thing’” to maintain public confidence (Holland & Fleming, 2003, p. 5).

Ministerial responsibility—and the emanating “chain of accountability”—is a crucial, albeit contested, mechanism to ensure that ministers and administrators conduct themselves responsibly and ethically (Strøm, 2000). Senior administrators account for the conduct of their departmental staff to their minister, while ministers account for departmental actions and their own actions to Parliament, and increasingly to the media (Bovens, 2007).

However, what is ethical leadership in these policy failures? Public administration includes inherently contrasting moral values (Pesch, 2008). For example, ministers must choose between shifting blame toward administrators and taking hierarchical responsibility for problems created by “many hands” (Thompson, 1980). Indeed, ethical leadership is context dependent (Heres & Lasthuizen, 2012, p. 442). What is considered ethical is shaped by both formal rules and informal norms. The question of this article is how formal accountability mechanisms and informal norms affect ministers’ decisions to take responsibility or to shift blame.

While there has been extensive research on ethical behavior of ministers in integrity violations (Holland & Fleming, 2003) or formal accountability mechanisms (Michael, 2005), limited research exist in how rules and norms impact ministers’ decisions to take the heat or shift the blame in policy incidents. This article goes beyond a formal focus by exploring perceptions of the actors in the “accountability chain” (Goss, 2003).

The article will draw on public accountability literature, which distinguishes between accountability as answerability and as blameworthiness (Dubnick, 2003). The article then relates accountability to the shifting “public service bargain” between ministers and top administrators (Hood & Lodge, 2006). This article explores three interrelated empirical questions: How do formal accountability mechanisms shape ministerial-administrative accountability after policy failures? How do those in the “chain of accountability” perceive ministerial-administrative accountability when a minister is under attack? Further, is scapegoating more prevalent in a system with a managerial public service bargain? To answer these questions, this article compares two parliamentary systems with similar ministerial responsibility conventions: the Netherlands and the Australian state of New South Wales. While both systems have long-established conventions of ministerial responsibility, New South Wales has a more pragmatic managerial bargain between ministers and top administrators.

MINISTERIAL ACCOUNTABILITY AND POLITICAL-ADMINISTRATIVE RELATIONS

A central idea in public administration is that “ethical behavior requires the presence of external accountability mechanisms” (Dubnick, 2003, p. 406; Finer, 1941). Responsible government is an umbrella term that covers many interrelated forms of responsibility, of which accountability (as answerability) and blameworthiness are only two (Thynne & Goldring, 1981, p. 200). This study uses the term accountability to highlight the social aspect of this process (Dubnick, 2003). Accountability is a situated pressure on actors to provide answers for their behavior toward specific account-holders, debate with these account-holders, and accept possible punishment (Bovens, 2007; Mulgan, 2003). Policy failures can act as “entry points” to study how these accountability dynamics and subsequent sanctions work in practice (Olsen, 2017; Romzek & Dubnick, 1987).

Answerability and Blameworthiness

Policy failures trigger account-giving as answerability, in the sense that actors have to provide explanations for their actions to their hierarchical or democratic “principal.” Each principal has certain informal and formal ex post mechanisms to control the performance of the actor (Strøm, 2000). Officially, top administrators account to their minister, while ministers account to Parliament and increasingly to the media (Bovens, 2007). For example, journalists can interrogate ministers in interviews, parliamentarians can investigate ministers in inquiries, and the minister can demand performance reports from their civil servants. Meanwhile, top administrators are increasingly public account-givers in parliamentary hearings and even in media (Aucoin, 2012). Ministers and top administrators can explain in their accounts why failures are less harmful than claimed (justifications), or why they are not the responsible agent by shifting responsibility, and thus blame, to other actors (excuses) (McGraw, 1990).

Policy failures also trigger account-giving as blameworthiness when certain actors are singled out for blame for the failure (Dubnick, 2003). In this case, principals have certain ex post controls to punish their actors (Strøm, 2000). For example, ministers can demote their civil servants, the media can destroy the reputation of a minister, Parliament can force ministers to resign, and voters can elect different parliamentarians.

However, the problem of the “many hands,” the fact that a policy failure is often caused by many factors and individuals acting together, makes it nearly impossible to establish which individual is (causally) responsible and thus blameworthy for the failure (Pesch, 2008; Thompson, 1980). According to Bovens (2007, p. 458), “many hands problems” create dilemmas of “who should be singled out for accountability, blame and punishment.” Even if casual responsibility is difficult to pinpoint, accountability norms aim to find an “adequately blameworthy agent” to punish (Shklar, 1990, p. 62). Accountability processes can focus on hierarchical responsibility or individual responsibility (Bovens, 2007; Pesch, 2008).

Hierarchical responsibility argues that “whoever is at the top of the organization pyramid takes the blame” (Pesch, 2008, p. 338). Mulgan (2003, p. 206) argued that “if ministers can accept praise, they should therefore also be prepared to accept the public blame for collective faults and not unfairly pass the buck to their subordinates.” However, others have argued that civil servants thus are unfairly exempted from blame, despite their considerable influence on policy implementation (Thompson, 1980).

Individual responsibility argues that each individual should be judged “on the basis of his actual contribution” instead of their formal position (Bovens, 2007, p. 459). However, causal responsibility is rarely unambiguous. The search for a blameworthy actor can turn into a “scapegoating” contest, as all actors are generally motivated to avoid blame (Ellis, 1994; Weaver, 1986). As ministers have hierarchical powers over top administrators, ministers can turn their administrators into lightning rods (Ellis, 1994). Ministers can highlight the individual blameworthiness of top administrators by making them publicly answerable for failures or by sacking them publicly.

Accountability processes after policy failures are shaped by formal rules and informal norms among those in the “accountability chain” (Olsen, 2017). Informal norms can be defined as “shared meanings and practices that come to be taken as given over a long period of time” (March & Olsen, 1996, p. 249). Informal norms can follow a “logic of consequences” (how to avoid blame) or a “logic or appropriateness” (how to be a responsible minister) (March & Olsen, 1996). Formal rules and informal norms can shift ministerial-administrative accountability, both in an answerability and a blameworthiness sense, toward either individual or hierarchical accountability. A graphical representation is presented in Table 1.

TABLE 1 Focus: Political-Administrative Accountability

	Accountability as
	Answerability	Blameworthiness
Hierarchical	Ministers account to Parliament, top administrators to minister	Ministers shoulder the blame for departmental actions
Individual	Top administrators also account to parliament/media	Individual blameworthiness of ministers and administrators

Managerial Bargains and Scapegoating

The question remains whether changes in the “public service bargain” between ministers and top administrators have shifted the focus from hierarchical to individual responsibility. In the “traditional” public service bargain, “public servants provide loyalty and competent service to the government of the day in exchange for trust, anonymity, merit selection and permanent tenure” (Hondeghem & Dorpe, 2013, p. 21). However, ministerial responsibility as the clear hierarchical line of authority and accountability is under pressure (Holland & Fleming, 2003).

Since the 1980s, parliamentary democracies have experienced New Public Management (NPM) reforms. These reforms included as a shift from the ideal of direct ministerial control toward more managerial relations (Hood & Lodge, 2006). While top administrators in the old “bargain” were considered “serial loyalists” to the ministers whom they served, in managerial systems they lost their tenure, “enjoyed more latitude in operational decision making . . . and share liability for mishaps” (Elston, 2017, p. 87). Some have expressed concerns that these reforms have politicized the higher echelons of the public service into “promiscuous partisans” (Aucoin, 2012). The question is whether managerial bargains also change informal norms for ministers to scapegoat top administrators for policy failures (Ellis, 1994). Therefore, this article compares two parliamentary systems: one with a more “traditional” bargain and one with a more managerial bargain.

SYSTEM COMPARISON

This study selected two systems of parliamentary government with a similar “accountability chain” (Strøm, 2000). Both New South Wales (NSW) and the Netherlands are established parliamentary systems with similar rules for individual ministerial responsibility. In both systems, the lower house—the Legislative Assembly and the House of Representatives, respectively—can sanction ministers who have failed to execute their ministerial responsibilities by a vote of no confidence. However, both systems differ on three important points: (a) government dominance of the legislature; (b) a consensual versus adversarial style of politics; and (c) the “public service bargain.”

First, while NSW governments dominate the lower house of Parliament (Hesford, 2007, p. 137), Dutch governments have to cooperate with a strong working Parliament, which gives opposition parties ample influence (Andeweg, 1992). NSW ministers are also parliamentarians, while Dutch ministers are barred from serving as parliamentarians. A paradoxical expectation is thus that while Dutch ministers are barred from parliamentary service, Dutch Parliament has a stronger role in ministerial responsibility than correspondingly in the NSW Parliament.

Second, while Dutch political tradition is rooted in a consensual style of politics, where cooperation is highlighted, NSW is known for its robustly combative, “confrontational style of debate” and “winner takes all” approach (Hesford, 2007, p. 137). Actors in political consensus cultures could have different ideas of ministerial accountability than those in adversarial cultures. Therefore, it can be expected that NSW ministers are likely to repel blame accusations more successfully than in the Dutch case.

Third, the rights and duties of Dutch administrators are constitutionally protected, while NSW administrators’ rights and duties consist of pragmatic agency arrangements (Bourgault & Van Dorpe, 2013). NSW is a state government and therefore mainly responsible for policy implementation and service delivery. Service delivery could instill in NSW administrators more individual responsibility. Therefore, the expectation is that Dutch actors focus more on hierarchical accountability of ministers than do NSW actors.

The structured comparison of both systems encompassed three analytical levels. First, formal accountability rules were assessed using public administration and constitutional law sources. Second, informal norms were probed using 61 qualitative vignette interviews with former ministers; top administrators; ministerial advisors; parliamentarians; and journalists. Last, forced ministerial resignations and public administrators’ dismissals after policy failures were inventoried between 1995 and 2017. Forced ministerial resignations can be grouped in five categories: personal errors (minister is personally culpable, such as misleading parliament); departmental errors (no direct ministerial involvement); internal cabinet conflict; financial scandals; and other controversies (such as sex scandals) (Thompson & Tillotsen, 1999). Supplementary Appendix A provides further information on the design of this study.

THE NETHERLANDS

This part will explore the formal accountability rules, informal norms for political-administrative accountability and historical forced resignations and administrative dismissals in the Netherlands.

Hierarchical Answerability

Dutch ministers have a constitutional duty to inform Parliament about developments in their portfolio, even without explicit parliamentary requests (Visser, 2008, p. 109). Parliament can set her own agenda and hold an urgent debate with a minister after approval by one fifth of parliamentarians. Although ministers can claim that “the interest of the state” necessitates secrecy, ministers have only sparingly used this claim (Enthoven, 2011). Both houses of Parliament can establish parliamentary inquiries into policy failures. Dutch parliamentary inquiries are “powerful parliamentary weapons” to investigate government fiascos, as inquiry committees can compel both ministers and civil servants to testify (Andeweg, 1992, p. 192).

Ministers can assume parliamentary confidence unless a majority in one of the Houses supports a no-confidence vote. Parliamentary confidence is a political judgment and therefore not constitutionally codified (Commissie-Scheltema, 1993, pp. 12–15). Prime Ministers have no formal powers to appoint; dismiss; or reshuffle ministers (Bovens, Brandsma, & Thesingh, 2015).

Since 1980, there is an ongoing discussion whether civil servants can be called to testify in Parliament (Enthoven, 2011). A committee concluded that civil servants could inform Parliament on “factual” matters, while ministers remain responsible to explain and defend policies (Commissie-Scheltema, 1993). The committee acknowledged that this distinction is difficult to uphold in practice. After a couple of high-profile clashes between top administrators and their ministers in the 1990s, Prime Minister Wim Kok published “guidance” for civil servants on external contacts (https://wetten.overheid.nl/BWBR0031968/1998-06-10). The guidance was so strict that it is known as the “edict-Kok” (Enthoven, 2011). The edict gives ministers the ability to reject or severely restrict parliamentary questioning of their civil servants. While Parliament has multiple times called for a more “relaxed” approach to parliamentarian-administration contacts, successive adjustments have not significantly changed the arrangement (Enthoven, 2011).

Since 1995, administrators in the highest two tiers of a department are embedded in a top management group (TMG). They are statutory employees of the Ministry of the Interior, and assigned to another department for seven years. While these assignments aimed to improve interdepartmental mobility, informally they also intended to “curb the power of some top civil servants” (Steen & Van der Meer, 2011, p. 212). Departmental heads sign performance agreements with their ministers (Hondeghem & Dorpe, 2013). As administrators remain statutory employees, agreements are only loosely linked to dismissal (Steen & Van der Meer, 2011). Top administrators can be transferred if they are dysfunctional or experience ministerial-administrative conflict, but only after approval of the head of the Telegraaf Media Group and the Ministry of the Interior.

Norms Insulate Administrators

Dutch interviewees primarily focused on how ministers would account to Parliament in a way that respects Parliament’s institutional prerogatives. One top administrator admitted to scanning the newspapers every morning to signal possible incidents that could lead to parliamentary trouble (NLCS1). The departmental heads considered themselves responsible for ensuring that ministers had sufficient information to account to Parliament in a credible manner, but saw no public account-giving role for themselves (NLCS2; NLCS3; NLCS6). One Head explained the process:

It was a Sunday afternoon, and I was working in the garden when the junior minister called: “There is serious fraud with the central exams.” So, we came together in the Department that afternoon, and held a crisis meeting. The junior minister said: “I want you to lead this, because this is such a risk that I want the head of the department to manage this correctly . . . . In those moments, you are the highest advisor towards the minister on what is a good strategy towards Parliament. You are not responsible for the personal choices and performance of the minister, but you have to make such he is prepared properly. (NLCS2)

When giving account for bureaucratic failures, ministers who would “panic” and blame their civil servants in Parliament were considered “weak leaders” (NLCS1; NLM4; NLP3):

You surely see that weak ministers tend to blame the civil service in public, as a sort of excuse, hoping that they will get away with that. It is not very popular in the lower house, and it is not seen as a powerful performance. (NLCS3)

At most, ministers could claim their administrators did not properly inform them (NLP8) or assure Parliament that they had tough internal meetings (NLM7). “Strong” Dutch ministers “stand in front” of their civil service and shoulder the blame, even when parliamentarians explicitly criticize the civil service (NLJ2). Parliamentarians focused largely on holding the minister to account:

We can say: “Administrative mistakes have been made, and they are serious, but we want to give the minister another chance to take responsibility, in the sense of really taking actions to stop repeating these failures in the future. (NLP4)

In terms of blameworthiness and sanctions, interviewees also focused on hierarchical responsibility. Top administrators did not fear scapegoating tactics and most considered ministerial resignations a “political issue” (NLCS3). Some ministers supported the “whip” argument laid out by Dutch law scholars: a ministerial resignation strengthens hierarchical responsibility, and will “whip” the civil service back into shape (NLM1). Similarly, a journalist argued:

I: And if the minister would say: I will not resign but fire my top administrator, because he made the mistake?

J: That I would find even worse, because it detracts from the symbolic function. These processes are about political accountability. Of course, the new minister can, after the resignation of the former minister, dismiss the civil servant who personally failed, but that is an internal departmental matter. (NLJ2)

Dutch informal norms bolster the formal rules that emphasize hierarchical accountability, both in terms of answerability and blameworthiness. While the close involvement of top administrators in the parliamentary account-giving by ministers could signify politicization, top administrators have no public or parliamentary account-giving role.

Sacrificial Responsibility and “Whips”

Between 1990 and 2017, 25 Dutch (junior) ministers were forced to resign. Fifteen of those ministers resigned over personal or departmental errors related to policy failures. Personal errors can involve ministers’ inability to resolve policy failures or errors in account-giving to Parliament. Table 2 shows the ministerial resignations between 1990 and 2017.

TABLE 2 Dutch Ministerial Resignations (1990–2017)

Types	Ministerial resignations
Personal errors in portfolio	13
Departmental errors	2
Internal conflict	5
Financial scandal	3
Other controversy	2

Only two ministers resigned after a departmental error. However, these ministers resigned just three months before parliamentary elections. Therefore, even in cases of strong hierarchical accountability rules and norms, ministers only rarely take the fall for departmental failures.

Furthermore, an exploratory newspaper inventory showed that dismissal of Dutch top administrators is rare. In the 1990s, four top administrators were transferred after they publicly criticized their minister or government policy. These cases signify publicized political-administrative conflicts rather than administrative scapegoating. However, three dismissal cases show that top administrators are not always insulated from internal blame after ministerial resignations. This is in line with the whip argument propagated by some of the Dutch interviewees. Table 3 summarizes three cases wherein top administrators were transferred or dismissed after their ministers resigned over a policy failure.

TABLE 3 Dismissals of Dutch Top Administrators After Failures (1990–2017)

Ministerial resignations	Policy failure	Subsequent administrative dismissals
Justice and Interior ministers (1994)	Police and prosecutors used illegal methods to infiltrate organized crime	Multiple police commissioners and prosecutors
Justice minister and juniorminister (2015)	Misinforming Parliament over a deal the junior minister made with a criminal in his time as prosecutor	Departmental Head
Junior Transport minister (2015)	Parliamentary inquiry concludedjunior minister misled Parliament regarding problems with high-speed train	Head of the Transport Inspectorate (transferred)

While the Dutch system is focused on hierarchical responsibility after policy failures, top administrators are not always insulated from internal blame after ministerial resignations.

New South Wales

This part will explore the formal accountability rules, informal norms for political-administrative accountability, and historical cases of scapegoating in New South Wales.

Individual Answerability and “off-with-their-heads”

The NSW courts have ruled that both houses of Parliament can scrutinize individual ministers and hold them accountable. Specifically, ministers have to account to the house of which they are a member (Twomey, 2004). If an Assembly (lower house) MP wants to question a Council (upper house) minister, an Assembly colleague will substitute their colleague. A government majority usually controls the parliamentary agenda in the Assembly. Subsequently, Question Time is one of the few moments in the Assembly that the opposition can scrutinize government action (Hesford, 2007). A lack of government majority in the Council since 1988 has given that house an important account-holding role (Lovelock & Evans, 2008). Council MPs can demand government papers; amend government legislation; and hold parliamentary inquiries. However, ministers have often withheld documents under a “public interest immunity” clause and while they usually give evidence during parliamentary inquiries, they cannot be compelled to testify.

Most scholars argue that the Assembly has the prerogative to withdraw confidence in ministers by means of a majority vote (Twomey, 2004). In addition, the Premier decides on the appointment and withdrawal of all ministers. Therefore, cabinet reshuffles often end ministerial careers rather than parliamentary confidence.

NSW civil servants are often called to testify before Parliament (Lovelock & Evans, 2008, p. 503). NSW officials can offer “factual and technical background” information to Parliament but are not expected to advocate or defend “government policies and administration” (Lovelock & Evans, 2008, p. 503).

Until 1988, top administrators in the two highest tiers of the department were employed on a permanent basis, and their appointment and dismissal was controlled by the independent Public Service Board (Dempster, 2004). Since 1988, top administrators are on temporary contracts and ministers gained the power to remove unilaterally senior civil servants (Public Sector Management Act; https://www.legislation.nsw.gov.au/#/view/act/1988/33). Furthermore, administrators are accountable to the minister for “the effective, efficient and economical management” of the Department. The changes were supposed to enhance performance management and promote merit recruitment (Funnell, 1990). In 2002, the Carr Government arranged that top administrators can be dismissed “for any or no stated reason and without notice” (Public Service Commission, 2013, p. 7). This amendment has been described “blackly within the public service as the ‘Bob Carr’s off-with-their-heads clause’” (Dempster, 2004). In 2011, a Public Service Commission has been reestablished but top appointments have been excluded from the Commission’s scrutiny. Clearly, NSW civil servants are formally more publicly answerable and blameworthy than their Dutch counterparts.

Bureaucratic Answerability and Scapegoating

NSW interviewees focused on how ministers would account in the media and considered Parliament a secondary account-holder. After a policy failure, NSW ministers and civil servants expected that ministers would push the top administrator forward to “front the media.” According to ministerial staffers, this tactic implied that it was “just an operational problem, not a policy issue” and thus falls under administrative accountability (NSWMS5; NSWMS6). A department head explained why he had to testify for a parliamentary inquiry:

The tactical decision would have been that any photograph or television clip [of the minister] before this enquiry would suggest that there is something wrong. The mere fact that the minister is there is, to some extent, to concede that the minister has something to answer for. (NSWCS5)

Even if ministers would not push their administrator to testify, they could hedge that “their department” advised them the failures were manageable and if this was discredited, they could blame the department (NSWM5). However, this would not necessarily stop journalists and especially Opposition members from trying to get a “ministerial scalp” in parliamentary inquires, no-confidence motions, or Question Time (NSWP5).

NSW interviewees not only reflected on bureaucrats who had publicly to defend government practices. Some also expected ministers to ritually sack a bureaucrat if their position was in the firing line (NSWM1; NSWM5; NSWM6). Contrary to the Dutch case, scapegoating was considered a viable tactic:

The department stuffed it up in the first place. I mean, one response might be to drag someone out of the department and make them accountable. So: sack, sack them. In this way, the media gets the blood they want, and the people see that tough action has been taken. (NSWM1)

However, not all NSW interviewees agreed with either the effectiveness or the appropriateness of creating a bureaucratic “fall guy” (NSWM6). Both Opposition parliamentarians and journalists explained that they would push for a ministerial scalp if they considered the minister liable for that failure (NSWP5; NSWJ2). An administrator explained how scapegoating would erode trust and diminish the effectiveness of the civil service:

The ministers always try to set it up that it will be the public servant who pays the price. The question is: are they paying the political price, or are they paying the management price? If they have been invited to resign, because they are incompetent, that is a completely different kettle of fish, from what I would call being the fall guy for the political process. . . . [That] erodes trust between the politicians and the public service. (NSWCS4)

Clearly, NSW informal norms further individualize accountability of top administrators in terms of answerability and blameworthiness. The line between administrative answerability and “promiscuous partisanship” is thin (Aucoin, 2012). NSW norms focused more on effective blame management of ministers than on ideas of appropriate ministerial leadership.

Insolated Ministers and Scapegoated Administrators

Between 1990 and 2017, 22 NSW ministers were forced to resign. Only three of those ministers resigned over personal errors. Two ministers resigned after they had misled Parliament and one Premier resigned after he had misinformed the NSW anticorruption body. Table 4 shows the ministerial resignations.

TABLE 4 NSW Ministerial Resignations (1990–2017)

Types	Ministerial resignations
Personal errors in portfolio	3
Departmental errors	0
Internal conflict	1
Financial scandal	9
Other controversy	9

New South Wales clearly has a more political approach to sacrificial responsibility. However, Thompson and Tillotsen (1999) show that ministerial resignations for personal errors was also a rare occurrence when the traditional bargain with top administrators was still in place (1945–1990).

Furthermore, an exploratory newspaper inventory showed that NSW top administrators often front the media. Many examples abound during the Carr governments (1995–2005) and later governments. Recently, Health minister Jillian Skinner forced her chief health officer to front the media alone during a press release of a critical report into a deadly accident in a public hospital (2016). The minister was reshuffled from her portfolio within a year.

Incoming governments have also replaced top administrators of departments after elections (in 2003, 2005, 2011, 2014). These replacements underline a politicization of top administrators. Furthermore, ministers have sacked at least eight top administrators after policy incidents. This number does not include the sackings of other administrators, such as heads of hospital boards and administrative commissions. Table 5 summarizes the dismissals.

TABLE 5 Dismissals of NSW Top Administrators After Failures (1990–2017)

Department and minister	Policy failure	Ministerial changes	Administrative dismissals
Community Services(1997)	Child abuse allegations in foster care	Reshuffle year after dismissal	Department Head (“always in media headlines”)
Transport (1997)	Failing train schedule	Reshuffle year after dismissal	Head of State Rail
Community Services(1998)	“Systemic failure” of the Department	No changes	Department Head
Community Services(2003)	Multiple, incl. child protection	Reshuffle along with administrator	Department Head (transferred in “shake-up”)
Transport (2003)	Deadly train incident	Reshuffle before dismissal	Department Head of Transport
			Head of State Rail
Utilities (2003)	Failed IT-project	–	Head of Sydney Water
Education (2004)	Botched restructure of Department	–	Head of Education Department
Roads (2005)	Cost-blowout new road tunnel	Reshuffle year after dismissal	CEO Roads and Traffic Authority

The Australian concluded after the sacking of two Transport administrators (2003) that it was a “smart strategy” of the minister “to sack a few heavyweights and flush out all the bad news.” However, half of the dismissals are followed-up by (or after) a reshuffle of the responsible minister and thus show that dismissals do not necessarily shift blame effectively. Moreover, the dismissals occurred only in certain departments and specific governments (Premier Carr, 1995–2005). Therefore, while administrative scapegoating might be considered a viable tactic for NSW ministers, it is not necessarily common or effective.

ACCOUNTABILITY SHAPES ETHICAL BEHAVIOR

Ministerial-administrative accountability after failures can have important “ritual and purifying functions” (Bovens, 2010, p. 954). Governments can show citizens that they are responsible and can be held accountable for policy failures. At the same time, failures pose an ethical dilemma in ministerial leadership, as they are often caused by “many hands.” Ministers thus have to decide, in terms of answerability and blameworthiness, to either take responsibility or shift the blame to others. This study shows that ministers who are bound by stronger formal hierarchical accountability and informal norms regarding appropriate account-giving are more answerable and likely to take the political fall. Table 6 answers the research questions.

TABLE 6 Answers to the Research Questions

RQs	The Netherlands	New South Wales
Formal accountability mechanisms	Formal hierarchical responsibility insulates top administrators from public accountability	Formal mechanisms give topadministrators a publicaccount-giving role
Political-administrativeaccountability perceptions	Hierarchical focus in account-holding; only “weak” ministers blame civil servants	“Culprit-finding” focus in account-holding; scapegoating is a viable tactic although not necessarily stopping blaming the minister
Resignations and scapegoating	Ministers take “sacrificial” responsibility; top administrators not insulated from internal blame	Ministers rarely resign over policy failures; multiple top administrators have been sacked, not necessarily related to managerial bargain

While limited public answerability of Dutch administrators might sometimes prevent Parliament from exercising her account-holding role, the cut-throat adversarial culture of NSW politics has exposed civil servants to “political hunts” and not necessarily enhanced account-holding.

Ministerial resignations can provide a symbolic demonstration of responsible government. However, ministerial accountability encompasses more than ministers or top administrators resigning. Ministers are also expected to explain how they ameliorated incidents and take policy responsibility to mitigate future failures. Formal and informal accountability mechanisms can constrain blame shifting opportunities for ministers and encourage ministerial account-giving.

Additional information

Funding

This work was supported by Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NL) [406-13-005].