Abstract
The perspective of Public Service Bargains has been used to analyze the relationships between politicians and senior civil servants based on the premise of stable expectations about the roles of both parties. Changing governance arrangements and leadership changes, however, may destabilize and increase ambiguity about these expectations. Hong Kong provides a case of changing governance arrangements that has destabilized the roles of senior civil servants, providing them with new opportunities to slide back and forth between administrative and political roles. We discuss the case of Hong Kong, analyzing the move from a colonial trustee-type bargain to an agency-type bargain. The case study provides new insights into the applicability of the PSB concept in conditions of changing governance arrangements and unstable political and administrative roles.
The roles of politicians and civil servants may be less stable than is sometimes assumed. Changing governance structures may also upset these relationships, leading to contests for power and instability.
Introduction
The concept of the ‘Public Service Bargain’ provides an important insight into the relationships between public servants and their nominal political masters. This approach demonstrates that there are indeed bargains, even if subtle and tacit, among these major participants in the governance process. More specifically, the two variations of the basic model – trustee bargains and principal–agent bargains, developed by Hood and Lodge (2006) – can describe the relationships between politicians and bureaucrats. In order for these models to function, the actors involved must have stable expectations about their roles and the roles of the other parties to the bargain. In the trustee model, for example, if public servants appear to have a good deal of autonomy, those public servants may be unlikely to exercise that autonomy if the political leadership has proven itself unreliable in the past and may therefore punish public servants for their decisions. There is, of course, a thin line between enforcing accountability and reneging on an understanding between partners in an agreement.
These bargains may not function in more fluid settings in which the relationships among these actors are unclear. Some of the ambiguity in these relationships may arise from instability in the actors involved, especially the political leaders. For example, a change in government may lead to questioning the trustee relationship, especially if there are marked ideological differences among parties. As the more established political parties lose their appeal to ‘flash parties’, e.g. the anti-immigrant parties in the Netherlands and Denmark (Abedi, 2002; Deschouwer, 2008), the members of governments may not have the same commitments to understandings with the civil service. 2 The ‘cartel parties’ (Katz and Mair, 2009) that have dominated European politics have had a commitment to longer term relations with the civil service, but newer and often transient parties have fewer incentives to cooperate with the bureaucracy.
As well as the influence coming from changes in political parties, career structures for participants in these bargains may not be unambiguous. In the United States, for example, many occupants of positions that in other political systems would be career public servants are political appointees (Light, 1995; Maranto, 2005). These ‘hybrid executives’ have some motivation to function as political officials, but they also have reasons to focus on their administrative roles. There are also hybrids identified in the literature that focuses on the attitudinal characteristics of public servants. For example, in the original Aberbach et al. (1981) study, the authors identified officials who were ‘pure hybrids’, having some values of the classical bureaucrats and some of the political bureaucrats. Their playing one role or the other was a function of situations and perhaps the nature of the political leaders involved (see Aberbach and Rockman, 1988). 3
Other analyses of politicians and bureaucrats have identified some of the same hybrid roles for civil servants. Building on Aberbach et al., Campbell and Peters (1988; see also Lee and Raadschelders, 2008) explored the nature of the hybrid civil servant and the various games that these officials could play both to maintain their own position and to enhance their policy capacity. Thus, the simple trustee or principal–agent models may actually contain a number of dynamic elements and permit hybrids that are also chameleons. That is, officials may want to fit into either of the two worlds – politics and administration – and adopt protective coloration in order to move back and forth with ease.
We investigate the utility of the Public Service Bargain framework in a non-European setting, Hong Kong, to better understand the extent to which the theory travels well outside Europe. Following Hood and Lodge (2006), we argue that in colonial and pre-2002 Hong Kong, stable political and bureaucratic roles existed (the Governor/Chief Executive was a political appointee and career civil servants played both political and administrative roles). The transformation of the political system since 2002 provides an opportunity to examine a PSB that transformed from a trustee into an agency bargain. In this article we focus on the contested nature of Public Service Bargains in changing governance arrangements that have encouraged bureaucrats in particular to slide easily into and out of political roles. We acknowledge that Hong Kong is a special case, particularly because the new system has only been recently introduced and continues to evolve. This rather fluid situation contrasts with the stability of bargains in many European contexts.
Methodology
We rely on archival material and on a series of elite interviews that we conducted in 2009–11 with 57 politicians and bureaucrats, using a questionnaire adapted from Aberbach et al. (1981) that focused mainly on the interviewee’s perception of his or her role in decision-making and the roles of others. The questionnaire was composed of 26 open-ended questions about the appeal of the job, the frequency of contact with other key players, interviewees’ perception of their role, their evaluation of the state of relations between the various players, and their perception of the amount of influence on policy that they and key others had. Each interviewee also completed a 36-item closed questionnaire in the presence of the interviewer that examined their attitudes on the Aberbach, Putnum and Rockman (APR) dimensions of elitism, tolerance of politics, program commitment, and constitutionality. Each interview lasted about an hour and was conducted in English. We were especially interested in the relationship between political appointees (‘Ministers’ in the Hong Kong context) and permanent secretaries, the most senior civil service position, in the policy formulation process. The total population of this group since 2002 includes 28 ‘Ministers’, of whom we interviewed 13 (46 percent), and 41 permanent secretaries, of whom we interviewed 14 (34 percent). That is, of our total of 57 interviewees, 27 or about 47 percent had been either a ‘Minister’ or a permanent secretary. We interviewed all those in this population who agreed to be interviewed. In addition, we interviewed another 30 individuals who held other positions (Executive Councillors) or slightly more junior positions (e.g. politically appointed Deputy Secretaries and Political Assistants, and civil servants who were or had been Heads of Department) or were members of the pre-2002 government to better understand the relationship between Ministers and permanent secretaries.
Public Service Bargains
Following Hood and Lodge, we define Public Service Bargains as ‘explicit or implicit agreements between public servants and those they serve’ (2006: 6), which builds on Hood’s earlier definition: ‘any explicit or implicit understanding between (senior) public servants and other actors in a political system over their duties and entitlements relating to responsibility, autonomy and political identity, and expressed in convention or formal law or a mixture of both’ (2001: 13). By the ‘other actors’ in the Hong Kong case we refer to ‘political appointees’ (the Governor/Chief Executive (CE) and, since 2002, 16 to 18 politically-appointed Principal Officials or ‘Ministers’) in each government and not, generally, to Hong Kong’s elected legislators who do not form the government.
Hood and Lodge identify two major types of bargain: trustee bargains and agency bargains. In a trustee-type bargain, public servants are expected to act as independent judges of the public good and possess autonomy. In such a bargain, the tenure and rewards of public servants are not under direct political control, their expected skills and competencies are not determined by the instrumental interests of elected politicians, and they are loyal to an entity broader than the current government (2006: 24–25). Hood and Lodge argue that trustee-type bargains were most prominent in pre-democratic and colonial settings, such as Hong Kong (Hong Kong was a British colony until 1997, and has never had meaningful democracy). Hood and Lodge explain the persistence of trustee bargains into the democratic age as a result of historical lag, a desire for checks and balances, as well as politicians seeking to avoid blame (2006: 29–33). In an agency-type bargain, public servants (agents) are expected to follow the lawful orders of politicians (principals), and politicians in turn are responsible for the actions of public servants. Politicians directly control the reward and tenure of the public servants, the skills and competency required of the public servants are those needed to do the politicians’ bidding, and public servants are only loyal to the wishes of politicians. The agency bargain is seen as necessary for bureaucracy in democratic government (2006: 43–47).
Based on grid-group cultural theory and admittedly ‘loose linkages’, Hood and Lodge (2006) discuss various kinds of bargain that vary along three dimensions: rewards, competency, and loyalty. They acknowledge that ‘once we penetrate beyond first-order stereotypes, no public service system seems to fit within any one of these four bundles of bargains’ (2006: 135). Although we conclude that Public Service Bargains in Hong Kong, like the UK, Belgium, and the Netherlands (Steen and van der Meer, 2011; Van Dorpe and Horton, 2011; de Visscher et al., 2011) belong to the hybrid variety, there has been considerable variation and, indeed, contest between politicians and public servants over the nature of the bargains. During the late colonial era, the PSB in Hong Kong was largely a mix of a moralistic and technical trustee-type bargain, with some agency elements. Regime change in 1997 brought to power a political executive determined to strengthen the agency bargain (albeit delegated), a change contested by the public service. Further governance reform resulted in compromise and the PSB landscape in Hong Kong now is largely a delegated agency-type bargain with some trustee elements.
The Hong Kong context
Colonial Hong Kong was a stable bureaucrat-dominated state (Scott, 1989) where political leadership was provided by an appointed Governor sent from London and where the most senior civil servants, chameleon-like, played both political and administrative roles. They were responsible for making policy, defending it, and selling it to the public as well as policy advice and implementation (Burns, 2004; Scott, 2010). This situation characterized Hong Kong’s political arrangements from the late 1960s until the 1990s.
Since the 1990s, this context has changed in three significant ways. First, Hong Kong adopted some ‘New Public Management’-type reforms from 1989 that affected the terms of the bargain for civil servants. In particular, the reforms reduced the attractiveness of civil service jobs through, for example, abolishing pensions for new hires and cutting pay (Burns, 2004; Sankey, 2001). Second, in 1997 Hong Kong became a special administrative region of China, a constitutional change that enabled the terms of the PSB in Hong Kong to shift from a mostly trusteeship-type toward an agency-type bargain. Third, in 2002, the Chief Executive, businessman C.H. Tung appointed by the Chinese government in 1997, introduced new governance arrangements, the Principal Official Accountability System (POAS), that created a new layer of politically appointed ‘Ministers’ above the civil servants, the most senior of whom became permanent secretaries. This change introduced a high degree of uncertainty for civil servants, whose ‘bosses’ might no longer come from their ranks. The permanent secretaries retained financial and personnel control over their programs (a ‘delegated agency type bargain’). The reform was introduced after years of tension between the Chief Executive and the civil service, whom the CE perceived to be obstructing his policy initiatives.
The POAS era may usefully be divided into two phases. In POAS I (2002–07), CE Tung recruited many non-civil servants as ‘Ministers’ (eight of 14 came from among ‘outsiders’ including three medical doctors, three from business, an environmental consultant, and a solicitor). The remaining six positions were filled by civil servants turned political appointees (they had first to resign from the civil service to take up these new jobs). For many senior civil servants the POAS was a shock. The Chief Secretary for Administration, the most senior civil servant, resigned in protest and many others took early retirement. Many civil servants perceived that the new system diminished the power of the civil service and undermined neutrality (Chan, 2005). Chief Executive Tung himself resigned half-way through his second term and was replaced in 2005 by Donald Tsang, a retired civil servant. Tsang was then appointed CE in his own right in 2007. POAS II, which dates from 2007, saw most positions filled by retired civil servants and a kind of equilibrium re-instated between ‘Ministers’ and civil servants.
Components of the PSB in Hong Kong
Following Hood and Lodge, we discuss the changing nature of Public Service Bargains in terms of rewards, competency, and loyalty. The process of moving Hong Kong’s PSB from a mostly trusteeship-type to a mostly delegated agency-type has been uneven and contested as both political appointees and civil servants have cheated and sought to avoid blame.
Rewards
Politicians provide rewards (including access, anonymity, relatively permanent employment, and reasonable salaries and pensions) to public servants in exchange for loyal and competent service.
Career patterns
Following UK traditional practice, Hong Kong’s civil service is divided into two broad classes of occupation, generalist grades, the most senior of which is the Administrative Officer (AO) grade (about 600), and specialist or departmental grades. AOs staff virtually all the top positions (permanent secretary, deputy secretary, and principal assistant secretary) in the now 12 policy-making bureaus, and head many government departments. 4 Accordingly, our focus will be on the top AO grades. In 2011 there were 17 permanent secretaries (Staff Grade A1), the most senior civil service position. Hong Kong has neither a Senior Executive Service (such as in Australia, New Zealand, and the US) nor a Senior Civil Service (such as in the UK) that might bring together lateral entrants and civil servants from the generalist and departmental streams.
Selection, appointment, mobility
Senior civil servants themselves select the permanent secretaries and other senior generalists, led by the Permanent Secretary of the Civil Service Bureau who chairs Administrative Officer grade promotion and postings boards, consisting of other permanent secretaries. The Secretary for the Civil Service is herself a political appointee and formally not part of the selection process. 5 All of our interviewees agreed that politically appointed ‘Ministers’ had no say in who would become their permanent secretaries, nor did they participate directly in the promotion and posting decisions of top civil servants, which some ‘Ministers’ felt weakened their position.
Hong Kong’s permanent secretaries have become younger on appointment since the POAS was introduced in 2002. Then, most permanent secretaries on appointment were from 50 to 54 years of age, while by 2011 most were 45–49 years of age with three in the 40–44 year age group. On average, permanent secretaries serve slightly over four years, probably because the government is reluctant to rock the boat during a Chief Executive’s five-year term.
The risk of dismissal for a permanent secretary is relatively remote. In our data from 41 permanent secretaries from 2002 to 2011, we are aware of only two permanent secretaries that might be said to have been ‘replaced’, either because of clashes of work style or conflicts over their roles. Our interviewees confirmed that for the most part the Minister and the permanent secretary simply attempt to make the relationship work. Indeed, from 2002 to 2011 personal relationships among the dyads in our data ranged from open warfare to cozy partnerships (‘we were the “dream team”’, one former permanent secretary told us, an evaluation agreed to by the Minister), and various relationships in between, such as ‘speaking truth to power’ and ‘Yes, Minister’-types of arrangements. From 2007 to 2012, when the new CE drew most politically appointed Ministers from among retired civil servants, the relationships became smoother (‘village life’-like; Peters, 1987). Given the rare cases of incompatibility or conflict between Ministers and permanent secretaries, the government has seen no need to establish a conflict resolution mechanism as was set up in the Netherlands (Steen and van der Meer, 2011).
Pay levels and components
Like the rest of the civil service, permanent secretaries are paid for their position (there is no performance-based pay in the Hong Kong civil service) and their monthly (base) pay is less than that for politically appointed ‘Ministers’. The Ministers, however, receive a lump sum, no housing allowance and no pension. If these are included, then the pay package for the current permanent secretaries is clearly more generous than for ‘Ministers’.
In the absence of an elected government, Hong Kong’s approach to rewards for politically appointed ‘Ministers’ and civil servants has been entirely bureaucratic. From at least the 1970s, the government had established complex salary scales and institutionalized processes for reviewing and adjusting them (Lee, 2003). This approach has resulted in transparent but relatively high civil service salaries (see Figure 1), second only to Singapore in Asia (Hood and Peters, 2003). The high salaries for top civil servants, however, are considerably lower than salaries for comparable positions in the private sector, reflecting a kind of noblesse oblige-type bargain. In the absence of democratic politics which tends to keep public sector salaries down, civil servants in Hong Kong have been able to set their own salaries, constrained only by the occasional budget deficit which led to cuts in 2002 and 2009 (Burns, 2004).
Rewards for High Public Office in Hong Kong, 1997–2009
In 1999, the political executive introduced reforms of the civil service compensation system that resulted in several changes, including the abolition of pensions for all new entrants. This change, contested by civil service unions at the time, fundamentally altered the Public Service Bargain. The 1999 reforms also tied civil service salaries more directly to the private sector and made it more difficult to obtain a permanent job in the civil service, although these reforms have since been relaxed.
The introduction of the POAS in 2002 negatively affected the promotion prospects of senior civil servants. As part of the process, the government deleted over 1000 directorate-level posts (through merging the Housing, Education, and Environmental Protection bureaus with their policy implementation departments), and by appointing many outsiders as ‘Ministers’. In 2007, the new CE Donald Tsang shifted the balance in the appointment of ‘Ministers’, most of whom were drawn from among retired civil servants. The Tsang administration also expanded the directorate, improving promotion prospects for senior civil servants (Scott, 2010: 82). By addressing the concerns of the civil servants, these changes helped stabilize the PSB. But since the Chief Executive is not elected by universal suffrage, is not allowed any political party affiliation, and is appointed by the central government, there is a considerable lottery element for current civil servants who have ambitions to become ‘Ministers’.
An important part of the rewards for Hong Kong civil servants, where the civil service retirement age is 55 to 60 (civil servants may take ‘early retirement’ at 55 and, if eligible, receive a pension), has been very generous post-retirement employment opportunities. Since the 1990s there has been an explosion of new regulatory and hybrid quasi-governmental agencies, leadership positions of which mostly pay high salaries. The government has placed numerous retired civil servants in leadership positions of these agencies (SCMP, 4 January 2002). This has also resulted in some retired civil servants being paid substantially more than their politically appointed superiors.
Retired civil servants, who have experienced a relatively loose regime of control after leaving their positions, may also accept jobs in the private sector. A series of scandals in which high-profile retirees took jobs in businesses they had had dealings with while in government forced the government to conduct two reviews of the system and forced an apology from the Secretary for the Civil Service for mismanaging the approval process for the former Permanent Secretary for Housing. Hong Kong civil servants, then, are rewarded not only on the job but also after retirement.
Competency
In exchange for rewards, civil servants provide competent advice, which should be tailored to the needs of their political masters.
Education and training
Educational backround of permanent secretaries, 2002 and 2011
Source: Civil Service Bureau, Who’s Who in the Administrative Service (various years).
The most important training for permanent secretaries is ‘on the job’ through acting appointments, where their performance is evaluated by superiors. Although a few AOs have advanced degrees (MPA degrees, for example), most training is informal. A common career progression is for a deputy secretary to spend time as a department head, more focused on implementation, before being promoted to permanent secretary.
Role and function
Until 2002, civil servants in Hong Kong performed both political and administrative roles (Scott, 1989). With the introduction of the POAS in 2002, politically appointed ‘Ministers’ formally assumed responsibility for the ‘strategic direction, policy formulation, policy decisions and public advocacy of policy proposals’ (Civil Service Code, 5.1), that is, formulation, selling and defending policy, and more controversially were required ‘to accept total responsibility for policy outcomes and the delivery of services by the relevant executive departments’ (Framework of Accountability System for Principal Officials, 17 April 2002, emphasis added). The ‘Code for Principal Officials under the Political Appointment System’, subsequently effected in 2008, softened responsibility for policy implementation somewhat, stating simply that ‘Ministers’ were responsible to the CE for the success or failure of their policies, with no mention of their responsibility to the public.
The Code for Officials did not give ‘Ministers’ exclusive responsibility in any domain, however. Indeed, according to a separate Civil Service Code issued in 2009, permanent secretaries, who remained Controlling Officers and responsible for civil service personnel management, were responsible for developing policy options or proposals and assessing the implications of the advice, They were to assist ‘Ministers’ in selling and defending the policy, and therefore continued to play a residual political role (Civil Service Code, 5.6). The government instructed civil servants to be sure that their involvement in or contribution to any public debate or discussion in public matters agreed with the policies of government, and was appropriate for their position. They were ‘not to seek to obstruct or frustrate a policy or decision taken by politically appointed officials, or delay its implementation’ (Civil Service Code, 5.6). The government demanded that civil service advice to ‘Ministers’ was to remain confidential (anonymity). In the official view, senior civil servants and ‘Ministers’ were expected to establish a good working relationship based on mutual trust ‘in the spirit of partnership’ (Civil Service Code 5.9 and 6.1). The new arrangements then sought to move the relationship from primarily a trustee-type to an agency-type bargain, adding ‘partnership’ to the competency requirements of the position.
Although politically appointed ‘Ministers’ were supposed to accept political roles, some did, but others did not. During POAS I, the mostly outsider ‘Ministers’ depended heavily on senior civil servants for policy advice and to perform ‘political work’. ‘Ministers’ had no staff of their own, and neither could they rely on political parties or think tanks, both of which were weakly developed (Ma, 2007). Lines of responsibility were not clearly differentiated, reflected in the fact that a code of conduct for civil servants was only formally promulgated in 2009, well after the system started.
During POAS I (2002–07), political appointees and civil servants contested the very meaning of ‘competency’. Unlike the politicians that feature in typical PSBs (Hood and Lodge, 2006), Hong Kong’s outsider political appointees were experts in their own fields – including finance, medicine, education, and the environment. Indeed, in many cases they had more expertise than the generalist permanent secretaries who served them (Politicians/Bureaucrats Project, 2009–2011). Each side tended to see itself as the repository of expertise, sometimes resulting in conflict (Interviews PO6; PS3, PS11). As one ‘Minister’ pointed out: ‘Permanent secretaries didn’t like our existence … Somebody over them, who in their eyes are quite dumb because we don’t know the system and have never been in the civil service’ (PO9). From the other side, a permanent secretary told us: ‘He (the ‘Minister’) had his views on issues, actually he derailed a number of very major initiatives, much to the horror of everybody involved …. He had a very bad reputation and didn’t have a clue what he was doing … It was a power game more than anything else’ (PS4). Senior civil servants, even if they acknowledged the expertise of their ‘Minister’, sometimes perceived the ‘Minister’ to be mainly self-seeking and usually administratively (bureaucratically) incompetent. Said the same permanent secretary: ‘I was very disillusioned as a result after working closely together [with the ‘Minister’] for a couple of months. I realized the Minister’s top priority was his own personal survival …. In terms of mission, [it was] far stronger among civil servants than politicians. Politicians unfortunately are self-seeking more than anything else’ (PS4).
The introduction of POAS II in 2007 solved this problem by replacing most outsider ‘Ministers’ with retired civil servants. The expansion of the number of political appointees in 2008 (with appointment of Under Secretaries and Political Assistants, mostly outsiders, to assist ‘Ministers’) strengthened the policy-selling role of political appointees vis-à-vis the civil service (Politicians/Bureaucrats Project, 2009–2011: Interviews, PS11, PS8). Over time, the ‘public face’ of government became more represented by political appointees. For example, political appointees began attending more Legislative Council (LegCo) meetings than previously and spoke more often (Synergy Net, 2010). Among our interviewees, many (politicians and bureaucrats) expressed frustration with politics in Hong Kong and a general dislike of political work, especially attending LegCo, which some said they considered a waste of time. Elected politicians, said one former Minister, are ‘uninformed to say the least and intellectually basically morons’ (PO9).
Politicians and bureaucrats perceived each other’s competency differently. Political appointees, for example, saw themselves as having good leadership qualities, while about half of the senior civil servants we interviewed either disagreed with this evaluation or were neutral. More than 80 percent of senior civil servants saw themselves as having good leadership qualities, but around a third of political appointees were neutral on the competence of civil servants Also, civil servants did not regard the competency of the ‘Ministers’ as highly as did the ministers themselves (Politicians/Bureaucrats Project, 2009–2011).
We detected little difference, however, between the two groups in their perception of the extent to which they brought relevant experience to the job. Still, more political appointees thought of themselves as having relevant experience, an evaluation challenged by some senior civil servants. There was a consensus between political appointees and senior civil servants, however, that civil servants brought valuable experience to their jobs. Said one ‘Minister’: ‘If you have problems, they will tell you what article [of the regulations is relevant], I had to rely on them’ (PO2).
Several professional outsiders among POs expressed their disatisfaction with the generalist administrative officers: ‘In today’s age you do need people with special training and special professional training to advise you. It’s not just common sense any more’ (PO9). Still, some senior civil servants contested the view that AOs lacked appropriate professional expertise, one calling it a ‘public myth’ (PS4). And some political appointees were sympathetic to the generalist orientation of senior civil servants. One political appointee, who held more sympathetic views toward the senior civil service, disagreed that appropriate expertise in the public sector could be compared with the private sector: ‘Expertise, you can’t compare with [the] private sector. [The] private sector is very, very different. Our public role is so broad that even the biggest private corporation can’t be compared with [it]. So it’s still early times, I won’t be so complacent to say we [politically appointed ‘Ministers’] have [the] expertise’ (PO3). An administrative officer turned PO defended the generalists: ‘From my point of view, it is analytical power, not so much your expertise, although you’ve got [to have] a basic understanding of the policies’ (PO6).
We conclude then that during POAS I (2002–07), outsider ‘Ministers’ who were professionals in their fields were more likely to find senior civil servants lacking in competency. In the terms of Hood and Lodge (2006), the outsider ‘Ministers’ wanted a ‘wonk’ and ‘delivery’ type bargain, that is, civil servants providing technical and subject knowledge and creatively getting things done. But Hong Kong’s AOs were best at ‘sage’ and ‘go-between’ bargains, that is, asessing risk, providing policy options, and knowing how to move things around the bureaucracy. Without control of civil service personnel policy, the civil service promotion system or even the appointment of their subordinates, the outsider ‘Ministers’ could do very little to change the bargain in their favor. They could at best bypass the generalists and attempt to direct the professionals at departmental level (PO9). Introduction of the outsider ‘Ministers’ destablized the bargain in POAS I. Both sides feared that the other side would cheat. During POAS II, the CE relied on retired civil servants who were more likely to hold sympathetic views towards their civil servant subordinates, thus helping to stablize the bargain.
The reform of the POAS did little to change the competency bargain. The competency requirements for civil servants in Hong Kong have evolved similarly to those in Australia, New Zealand, and the UK, requiring more managerialist orientations and training. Unlike these countries, however, Hong Kong’s colonial heritage thrust civil servants into political roles from the beginning. The introduction of a delegated agency-type bargain in 2002 initially put outsider politicians – experts – in charge. They challenged the civil service, which then eventually retreated into focusing more on its expertise of managing process. Thus, these changes associated with the POAS provide the means to understand changes in the PSBs.
Loyalty and accountability
According to the Basic Law, the civil service is accountable to the government of the HKSAR (that is, the Chief Executive and the principal officials) (Basic Law, Art. 99). We argue that CE Tung interpreted this accountability in agency terms. This meant that civil servants were seen as servants of political masters, that civil servants were ‘directable at will’ and were expected to transfer their loyalty to the government of the day (Hood and Lodge, 2006: 21). In this sort of bargain loyalty is highly prized. In exchange for loyalty, demonstrated by enthusiastically accepting the direction of the CE and providing their expert judgment, civil servants were handsomely rewarded (Hood and Lodge, 2006). As Hood and Lodge observe, citing the example of Singapore, ‘post-colonial governments have also often sought to develop more of an agency style of PSB in their efforts to bring formerly autonomous colonial bureaucracies under the heel of their new political masters’ (2006: 45).
Loyalty and responsibility to whom?
Since the introduction of the POAS in 2002, the political executive has formally instructed civil servants through the Civil Service Code to serve (and, thus, be loyal to) the government of the day. Civil servants shall provide ‘full, honest, and impartial advice, without fear or favor, and whether or not the advice accords with the views of politically appointed officials’ (Civil Service Code, 6.1). Although permanent secretaries are supposed to serve and support political appointees, they are accountable only to the ‘Minister’, and are not in a hierarchical relationship to under-secretaries or political assistants (Civil Service Code, 6.4). Permanent secretaries as Controlling Officers are accountable to the Financial Secretary (a political appointee since 2002) 6 and must appear before LegCo’s Finance Committee and before LegCo and its Panels to answer policy questions. Unlike the UK and Dutch cases with individual performance agreements for permanent secretaries (Steen and van der Meer, 2011; Van Dorpe and Horton, 2011), there are no such agreements at this level in Hong Kong. 7 That is, formally at least, ‘Ministers’ and not permanent secretaries are held accountable (to the Chief Executive) for the success or failure of government policy.
Civil servants saw political appointees lacking commitment and a sense of mission, perceiving that they focused primarily on political survival, making policy merely for political expediency. We asked interviewees to evaluate senior civil servants and political appointees in terms of their sense of mission. We found that each group, political appointees and civil servants, overwhelmingly perceived that their own group had a ‘sense of mission’, but was much less sure that the other group was similarly motivated (Politicians/Bureaucrats Project, 2009–2011). 8
On the issue of to whom they should be responsible, we also found significant divergences of opinion. When asked whether ‘the primary duty of the senior civil servant was to serve the interests of his or her Principal Official as faithfully and as competently as possible’, over three-quarters of ‘Ministers’ interviewed agreed while less than half of senior civil servants interviewed agreed, and fully one-third of senior civil servants disagreed (Politicians/Bureaucrats Project, 2009–2011). Perceptions on the degree to which senior civil servants have a duty to act in the general public interest, even if the wishes of their Principal Official may contravene this interest, also demonstrated a gap. Eighty percent of senior civil servants among our interviewees agreed with this statement, compared to about half of political appointees interviewed (Politicians/Bureaucrats Project, 2009–2011).
There is some evidence, then, of the divided loyalty and the contested nature of the government’s attempt to impose an agency bargain in 2002. A significant proportion of senior civil servants continued to see themselves as the arbiter of public interest, trustee-style. In this environment, with contested perceptions and few written rules, cheating was inevitable. In one case, several ‘Ministers’ approved a plan for a private company to organize an event to re-launch Hong Kong after SARS. Yet, no ‘Minister’ was made responsible for the event, and when it was not executed as planned, they collectively blamed a civil servant who had been the Controlling Officer. He was disciplined and fined, an action that was overturned only through judicial review. In this case, the civil servant became a scapegoat for political failures (Li, 2011; Rowse, 2009). Political appointees perceived that they should only be held accountable for policy decisions rather than policy implementation. According to one retired ‘Minister’: ‘As ministers, we are not supposed to be on day to day execution. Otherwise, you can’t have time to sleep. So in my view, the responsibility lies on the civil servants to execute the policy.… We should distinguish between policy level and operational level’ (PO4).
Political affiliation
The civil service in Hong Kong has long held ‘political neutrality’ as a core value, both before and after 1997 (Burns, 2004). Critics of the POAS claimed that its introduction would undermine civil service neutrality, 9 in part because it would encourage senior civil servants to slant their advice to improve their chances of promotion and political preferment later on. Under the POAS the terms of a ‘partnership’ bargain that included notions of anonymity, political neutrality, ‘the right to be heard’ and confidentiality, were articulated, and political appointees were specifically admonished to uphold the political neutrality of the civil service (see Civil Service Code, 2009; Code for Officials under the Political Appointment System, 2008). The 2002 Circular and the 2009 Code differed somewhat in their interpretation of political neutrality, however. The 2002 Circular stressed neutrality as ‘loyalty to the government of the day’, which added a new dimension to the traditional concept of political neutrality within the Hong Kong civil service, which meant ‘speaking truth to power’ (Scott, 2010: 81). That is, once a decision was made, civil servants should faithfully implement it irrespective of their personal views. Civil servants were also banned from speaking against government policies in public (Civil Service Circular, 2002). In 2009 the Civil Service Code demanded that civil servants serve the government of the day with ‘total loyalty and to the best of their ability, no matter what are their own political beliefs’, a kind of serial loyalist bargain that further demarcated the realm of politics and administration. At the same time, ‘Ministers’ were admonished to ‘give fair consideration and due weight to honest, informed and impartial advice from civil servants’ (Code for Principal Officials under the Accountability System, 2002; Code for Officials under the Political Appointment System, 2008).
In practice, the terms of a ‘partnership bargain’ were contested by civil servants and political appointees. Among our interviewees, political neutrality was mostly valued by senior civil servants, while some outsider political appointees considered it a myth and an obstacle to civil servants exercising their vision. According to one political appointee: ‘My opinion on neutrality of civil servants, particularly senior ones is, nobody could be really neutral, and particularly for AOs. Up to a certain level, like the directorate level, they need to engage political parties and the public. And they need to have political sensitivity. So you can’t actually say totally neutral’ (PO3). Among our interviewees, civil servants overwhelmingly agreed with the statement that ‘government works better if civil servants are politically neutral’ (87 percent), while political appointees were less certain. Among our ‘Ministers’, seven (or 58 percent) agreed (Politicians/Bureaucrats Project, 2009–2011).
Conclusion
Our discussion of the evolving PSBs in Hong Kong indicates that they are of the pragmatic hybrid variety. We have traced the evolution of what began as primarily a trustee-type bargain with some agency elements during the high colonial era (see Hood and Lodge, 2006: 154–155) into what has become a delegated agency bargain with some trustee characteristics, especially as most ‘Ministers’ are retired civil servants. Returning to our initial imagery of chameleon-like actors, the Hong Kong PSB has been primarily one of civil servants adapting to their surroundings and donning political garb as necessary.
The Hong Kong case is interesting because it challenges the notion that PSBs are based on stable role expectations. Colonial civil servants at the top in Hong Kong played both political and administrative roles. The POAS introduced some role differentiation, but after a brief experiment with outsider political appointees, retired civil servants have taken most political roles. High mobility between political and administrative offices is not unique to Hong Kong. In the US and France, for example, civil servants and political appointees slide into and out of various roles at the top of the system. In all these settings the village at the top of government provides for less differentiation of the roles between the two sets of actors.
The Hong Kong case is also important because it demonstrates that even when many of the contextual factors change, civil servants and politicians will find means of working together to provide governance. Although the relationships between political appointees and their career civil servants was not always as clearly defined or as amicable as it might have been, these actors were capable of redefining their bargains and moving ahead with the tasks at hand. We would expect this to be true for other transitional regimes, although the exact nature of the redefinitions and the bargains may well be different from those observed here.