Leaving the Senior Civil Service – public service bargain and beyond: The case of Estonia

Abstract

The article explores why apolitical public sector managers decide to or are forced to leave the civil service in the example of the Estonian Senior Civil Service. The article shows that the concept of public service bargain can help to understand and systematise the causes of leaving the Senior Civil Service. It is particularly useful in distinguishing between voluntary and involuntary turnover and in linking the turnover with political–administrative relations and loyalty issues. Since public service bargain does not specifically focus on voluntary exit, other approaches known from management literature are relevant for the operationalisation of voluntary turnover and complementing the public service bargain-based model for researching turnover of top executives. The empirical study maps the people who left the Estonian Senior Civil Service in 2009–2013 and analyses and systematises various causes of their departure on the basis of semi-structured interviews (70% response rate). The empirical study shows that the turnover of top executives can be considerable even without much direct political influence. It is found that job insecurity combined with the domination of individual unwritten public service bargains tends to lead to ambiguity in the perception of the roles of top executives, which in turn causes conflicts and dissatisfaction, materialising in high voluntary turnover.

Keywords

Civil service turnover Estonia public service bargain Senior Civil Service top executives

Introduction

This article explores why apolitical public sector managers decide to or are forced to leave the civil service using the example of the Estonian Senior Civil Service (SCS). Senior civil servants (SCSs) usually include (administrative) heads of ministries, departments, bureaus and agencies within the core civil service and other senior officials as designated within the central government of each country (Halligan, 2012: 116). Previous studies have demonstrated that SCSs’ turnover can bring along some positive developments, such as bringing in “fresh blood” with novel ideas into the civil service and providing promotion opportunities to the lower-ranked civil servants (Lee and Whitford, 2008; McElroy et al., 2001). However, there can also be negative effects linked to SCSs’ turnover, such as loss of institutional memory and competence, loss of organisational investments related to the recruitment, selection and training of civil servants, and costs related to finding and appointing new public sector managers (Cho and Lewis, 2012; Lee and Whitford, 2008; Moynihan and Landuyt, 2008). High turnover of SCSs may also affect collaboration within the civil service as inter-organisational relations are often materialised within key SCSs and their networks as well as posing a negative effect on common identity, ethos and values within the SCS which are particularly important in counterbalancing recent individual- and organisational-centred tendencies in public administration. Moreover, in academic literature turnover of SCSs has often been related to politicisation and political loyalty of SCSs (see, e.g. Christensen et al., 2014). This implies that both public administration academics and practitioners may benefit from knowing the reasons of SCSs’ turnover in order to minimise its potential negative effects.

Although the practices of formally recognising SCSs as a separate group have been mostly addressed in Anglo-American countries (see, e.g. Halligan, 2012; Kim, 2007), the Senior Civil Service (SCS) as a specific category of civil servants has been largely neglected in Central and Eastern Europe. In addition to addressing turnover of SCSs in a less researched region, this study also explores turnover of SCSs from novel angles that may interest broader readership. First, while the literature on politicisation reflects upon involuntary departure from the civil service, this paper sheds light on both involuntary and voluntary exit. For many cultural contexts, voluntary departure from the SCS is an extremely rare and unexpected occasion but in other settings – as the Estonian case demonstrates – in-and-out mobility in the SCS is a common practice. While the involuntary exit is certainly a more attractive topic to both academics and journalists, the voluntary exit should not be left unnoticed. Second, while most existing empirical studies on public sector turnover address the sub-national level (e.g. Boyne et al., 2010a, 2010b; Meier and Hicklin, 2008; Moynihan and Landuyt, 2008; Moynihan and Pandey, 2008; Selden, 2006; Tekniepe and Stream, 2012), this article covers the entire core of the central government – all SCSs leading the ministries and executive agencies in Estonia. And third, previous public sector research has mostly focused on the impact of political change on SCSs’ turnover, turnover intention, interlinkages between performance and turnover, turnover behaviour, the effect of salaries and dissatisfaction on turnover, person-organisation value fit and the role of social networks (see, e.g. Boyne et al., 2010a, 2010b; Cho and Lewis, 2012; Christensen et al., 2014; Lee and Whitford, 2008; Llorens and Stazyk, 2011; Meier and Hicklin, 2008; Moynihan and Landuyt, 2008; Selden, 2006; Tekniepe and Stream, 2012; Wheeler et al., 2007). However, there are hardly any studies which address the main topic of this article – a retrospective view of reasons why top executives have actually left the civil service. The actual fact of leaving the SCS differs from more hypothetical views on turnover intention of existing civil servants (see, e.g. Hwang and Kuo, 2006; Moynihan and Landuyt, 2008; Moynihan and Pandey, 2008). The shortage of relevant studies is related to the fact that firstly, the turnover of SCSs is not very high (especially in several European countries, which offer life-long tenure for civil servants) and secondly, and perhaps even more importantly, because former SCSs are very hard to access, and/or they are not willing to share their reasons of departure.

This study builds on the concept of public service bargain (hereinafter PSB; Hood and Lodge, 2006), which will serve as the basis of the theoretical framework. PSB has been defined as

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. (Hood and Lodge, 2006: 6)

Thus, the PSB focuses mostly on the relationships between SCSs and representatives of political system (e.g. politicians). According to the PSB, this relationship is based on a bargain between the two parties whereby, to put it simply, the political masters expect competence and loyalty from the senior civil servants and the latter expect some mixture of rewards (both tangible and intangible) and autonomy in a particular responsibility area in return (Burns et al., 2013; Hood and Lodge, 2006). Although the concept of PSB has attracted a lot of interest by public administration scholars (e.g. Bourgault, 2011; Halligan, 2013; Hondeghem, 2011; Hood and Lodge, 2006; Salomonsen and Knudsen, 2011), it has not been previously applied in the research of SCS turnover and reasons for leaving the SCS. While PSB is taken as a basis for building an analytical framework for studying SCS turnover, it is further complemented by private sector research on employee turnover, particularly by studies on “pull” and “push” factors (Ali Shah et al., 2010; Hwang and Kuo, 2006).

The empirical study illustrates the analytical framework of SCS turnover by mapping the people who have left the SCS in Estonia during the five-year period of 2009–2013, and analysing and systematising various causes of why they left the SCS on the basis of semi-structured interviews. The turnover and particularly reasons of departure from the SCS have not been studied in Estonia before despite the high importance given to the development of the SCS since 2004 (see Randma-Liiv et al., 2015).

Analytical framework

Public service bargain

Senior civil servants, as the highest-ranking administrators, are closely linked with a variety of stakeholders in the political system (politicians, political parties, citizens). SCSs’ relationships with representatives of political system are influenced by certain formal or informal agreements on mutual role relationships which can be understood and operationalised through the concept of PSB (Hood and Lodge, 2006). As SCSs perform on the borderline of politics and public administration, their relationships with executive politicians (especially ministers) deserve particular attention.

The meaning and substance of PSB may vary across time and space, and even across particular jobs. Definitions of various components of PSB – loyalty, competence and rewards – may differ, as well as their relative weight in PSB (Hondeghem, 2011; Hood and Lodge, 2006). This has formed a basis for the development of typologies of PSB (Hood and Lodge, 2006). While previous studies on PSB have mostly addressed the description and comparison of a variety of types of PSBs in different countries and their dynamics over time (e.g. Bourgault, 2011; Halligan, 2013; Salomonsen and Knudsen, 2011), the use of a concept of PSB in analysing public service turnover and reasons for leaving the SCS, in particular, have not been employed so far.

Deriving from PSB, if one or both parties do not meet the expectations of another party, or are not satisfied (anymore) with the conditions of a once-agreed bargain, a situation may occur where the bargain is broken by at least one party of the PSB (Hood and Lodge, 2006: 153). This, in turn, may cause the exit of a civil servant. Based on the PSB, it is possible to distinguish two types of turnover. First, if an SCS fails to deliver on the expectations of his/her political masters (in terms of competence and loyalty – which may or may not be related to politicisation) then that can lead to involuntary turnover initiated by a political master. Second, if the political master does not fulfil the expectations of an SCS (regarding rewards and/or autonomy) then that can lead to civil servants’ voluntary turnover. Figure 1 illustrates the two types of turnover.

Figure 1.

Reasons for leaving the civil service based on PSB (composed by the authors on the basis of Hood and Lodge, 2006).

The non-implementation and breaking off of the PSB can be a consequence of a number of processes. For example, expectations of the two parties of PSB may shift as a result of environmental changes. Such changes can derive from certain reforms or events (e.g. regime change, general elections, structural civil service reforms). Alternatively, they can be gradual, taking place within a longer period of time without linkage to a certain reform or change event (e.g. long-term developments in a certain policy field). The non-compliance of PSB may also refer to the situation where one or both parties of the PSB have (consciously) cheated the expectations of another party, and not provided it with what has been either formally or informally agreed upon (Hood and Lodge, 2006: 154–160). As different sides of a bargain can perceive the conditions of a bargain and, consequently, also cheating differently (cheating in the eyes of one party may not be perceived the same by another party), especially in the cases of unwritten PSBs, it is not so unusual to make mistakes in the implementation of PSB (Hood and Lodge, 2006: 11).

Thus, PSB provides a good foundation for analysing the reasons for leaving from the SCS, especially by distinguishing between voluntary and involuntary turnover. Several authors (e.g. Lee and Whitford, 2008; Llorens and Stazyk, 2011) have highlighted the fundamental difference between voluntary and involuntary exit caused by substantially different motives. In addition to a framework provided by PSB, there are yet other factors which may contribute to either type of turnover. Therefore, both voluntary and involuntary turnover are addressed more comprehensively below by complementing the PSB-based model with other approaches proposed by the literature on employee turnover.

Involuntary turnover

According to PSB, competence and loyalty are the two key factors which may lead to involuntary dismissal of SCSs. This can be caused either by a shift of the expectations of politicians or when an SCS does not meet the existing expectations of political masters. Competence refers to a variety of skills, knowledge, abilities and experience of an SCS leading to a successful implementation of job-related tasks (Hood and Lodge, 2006: 86). Although competence can be seen as one of the most important criteria in modern meritocratic civil service systems, the substance and relative weight of different competency requirements may vary across states, organisations, jobs or preferences of single managers (especially in cases of individual PSBs) (Hood and Lodge, 2006: 91–92). Different weight could be given, for example, to intellectual abilities, expert knowledge of a policy field, political craftmanship or managerial skills (Bach and Veit, 2018; Kuperus and Rode, 2008: 12–13). NPM-related reforms in the public sector have led to a situation where managerial skills have gained importance vis-a-vis other competency requirements (Burns et al., 2013: 139).

Despite the importance of SCSs’ competence, another element of PSB – loyalty – can turn out even more crucial among the parties of PSB (Hood and Lodge, 2006: 110; Rouban, 2012: 386). On the one hand, SCSs are expected to be loyal to the state, public interest and government as an institution (i.e. not to a specific political party or politician). On the other hand, politicians often expect public managers to be committed to their political goals (political loyalty) or to a particular minister (personal loyalty). These two potentially contradicting expectations may lead to a variety of loyalty conflicts in the relationships between politicians and SCSs (Bourgault, 2011: 254; Peters and Pierre, 2004: 2).

Political loyalty of SCSs becomes particularly essential after general elections, when political parties in power change. Incoming politicians may have some doubts whether SCSs who used to be loyal to outgoing political forces will continue to be equally loyal to new political parties in order to successfully carry out their political goals (Boyne et al., 2010a). One option for ensuring the political loyalty of SCSs is to replace old civil servants with new ones, whose loyalty to political masters cannot be questioned (Boyne et al., 2010a). Most modern democratic and value-based civil service systems foresee that (a majority of) SCSs are protected against (predominantly) politically reasoned layoffs, which can also be seen as one form of politicisation (Boyne et al., 2010a; Peters and Pierre, 2004; Rouban, 2012).

Personal loyalty to individual political leaders pertains that senior official’s loyalty and commitment is targeted to a single immediate superior (e.g. minister), his/her decisions and reputation as well as their interpersonal relations (Hood and Lodge, 2006). A paradox of democratic civil service is that SCSs and their political masters must be able to efficiently collaborate despite a variety of loyalty issues (Bourgault, 2011: 254). Politically neutral SCSs are expected to be able to survive political changes and to serve different political masters equally well (Boyne et al., 2010a: 138).

In addition to PSB-based factors related to competence and loyalty, involuntary dismissal can also result from SCSs not delivering on the expected (and agreed upon) results (Boeker, 1992; Boyne et al., 2010a). Achieving results is closely related to competence and loyalty: competence contributes to achieving better results, and loyalty to politicians may ensure greater commitment to achieving these results (Boyne et al., 2010a; Hood and Lodge, 2006; Selden, 2006). Failure in meeting the results does not necessarily lead to firing, especially in civil service systems which provide job security (Boyne et al., 2010b). As one of the most dire consequences, this may, however, exist (Boeker, 1992; Selden, 2006). This has attracted a concern of scholars, for example, on how to assess who is responsible for not achieving the results – a political master or an SCS – which is likely to lead to disagreements or even mutual accusation (Boyne et al., 2010b).

Involuntary dismissal provides a difficult research object for scholars because of its “hidden” character. For example, if the dismissal of an SCS proves complicated (e.g. because of legal regulations or tenure), it is possible for the political leaders to create uncomfortable conditions for further work or barriers for the successful implementation of tasks: for example, by diminishing the decision-making power of an SCS, increasing control or limiting information channels necessary for the successful implementation of tasks. This may lead to de jure voluntary departure of an SCS which is de facto an involuntary dismissal (Boyne et al., 2010a; Hood and Lodge, 2006).

Voluntary turnover

According to PSB, voluntary departure from the civil service is likely to occur in cases where the expectations of SCSs related to rewards and decision-making autonomy are not met. A usual reward bargain includes the reputation of public service as an employer, job security, public service pension and pay – which may not be equal to that of private sector managers but still solid and stable (Hood and Lodge, 2006: 65).

The decision-making autonomy of SCSs in a certain policy field and/or organisation can be regarded as a reward as well. At the same time, decision-making autonomy and accompanying responsibility can be valued differently: some SCSs may see them as part of the reward and a source of motivation, while others may perceive them as a source of dissatisfaction and stress (Hood and Lodge, 2006: 110). The autonomy of SCSs is closely linked to politico-administrative role relationships. For example, whether decision-making is subordinated to close political steering and control, on the one hand, or professional independence of SCSs (ability of SCSs to influence policy-making), on the other hand. The role of SCSs differs depending on whether an SCS is seen as a political advisor or as in charge of policy implementation and neutral policy advice (Hansen and Salomonsen, 2011: 199).

If the SCSs’ expectations related to reward and autonomy are not met, it may lead to job dissatisfaction. Previous research (e.g. Hwang and Kuo, 2006; Llorens and Stazyk, 2011) has demonstrated that job (dis)satisfaction is one of the major factors of voluntary turnover both in the public and private sectors. The higher the job satisfaction of an SCS, the less likely s/he is willing to leave the job (Moynihan and Pandey, 2008: 208). Thus, if an SCS wants to voluntarily leave the job, there are likely to be certain organisation- or job-specific or other factors that cause dissatisfaction and consequently motivate the decision to leave (Ali Shah et al., 2010; Tekniepe and Stream, 2012). Such factors are often called “push factors” in the literature.

Push factors may include rewards and decision-making autonomy, which are not in line with SCSs’ expectations. In addition, a number of other factors can lead to SCSs’ voluntary turnover: unclear goals, routine work, insufficient in-service training opportunities, limited promotion opportunities, an unfair performance-assessment system, job insecurity and overload (Kuperus and Rode, 2008; Lee and Whitford, 2008; Llorens and Stazyk, 2011; Moynihan and Landuyt, 2008). Push factors also involve dissatisfaction with relationships with colleagues and superiors (e.g. minister or higher-ranked SCS) (Moynihan and Pandey, 2008). Dissatisfaction with superiors may derive from poor management quality (Llorens and Stazyk, 2011), from political misunderstandings or conflicts (Tekniepe and Stream, 2012).

In addition to organisation- and job-based push factors, additional push factors may be based on the broader external environment. For example, the (dis)satisfaction of SCSs may also depend on the reputation of civil service as an employer in society – if being a civil servant is not valued in society, and negative attitudes towards state institutions and civil servants prevail, the motivation of SCSs may decrease (Demmke, 2005).

Next to PSB-based and other push factors, there can be yet other reasons for voluntary departure from the civil service. The decision to leave the job may depend on alternative employment opportunities, the so-called “pull factors” which “pull” the SCS towards joining a more attractive job elsewhere (Hwang and Kuo, 2006; Wheeler et al., 2007). Pull factors may include better pay, more generous training and promotion opportunities, more challenging tasks, better fitting workload or working conditions (Ali Shah et al., 2010; Cho and Lewis, 2012). However, the decision to leave may not necessarily depend on the availability of a specific alternative. Leaving an SCS is more likely if a senior official perceives the general situation in the labour market to be positive, his/her employment opportunities outside the current job to be sufficiently good and easily available (Hwang and Kuo, 2006; Llorens and Stazyk, 2011; Moynihan and Pandey, 2008; Wheeler et al., 2007).

In addition to push and pull factors, reasons for deciding to leave the job can be individual factors, such as the age of an SCS, education level, family issues and the number of years with the same employer (Cho and Lewis, 2012; Lee and Whitford, 2008; Llorens and Stazyk, 2011).

Reasons for voluntary turnover of SCSs are summarised in Figure 2. The decision to leave can be motivated by more than one factor at the same time.

Figure 2.

Reasons for voluntary turnover (based on Ali Shah et al., 2010; Hwang and Kuo, 2006).

Leaving the sacred land: The case of the Estonian Senior Civil Service

The Estonian Senior Civil Service

Although the Estonian state celebrated its 100th anniversary in 2018, it can be characterised as a young state which regained its independence in 1991. Despite the smallness of the country (population of 1.3 million), the Estonian central government operates a highly segmented administrative system in which the responsibility for public policies and programmes lies within individual ministries. The modern meritocratic civil service system was established by a 1995 Public Service Act (PSA). The public service covers only a small part of the public sector workforce (4–5% of the total workforce of the country) – people working for the ministries, government agencies, local government institutions and a few other state institutions, i.e. at the core of the public sector. This study focuses solely on the civil service (central government). The open, position-based civil service system provides no institutionalised tenure and offers limited opportunities for SCSs to move within the (senior) civil service. It is possible to enter the civil service at any rank, including SCS positions. In general, civil service is by and large perceived as “any other employer”, leading to mobility between the public and private sectors. This also explains the relatively short duration of civil service careers and the presence of rather young people in senior ranks (see Table 1). Annual turnover in the general civil service has been 7–10% in 2009–2016 (Ministry of Finance, 2017). The politicisation of the civil service is, however, one of the lowest in Central and Eastern Europe (Meyer-Sahling, 2011).

Table 1.

Summary of SCS turnover in Estonia (2009–2013).

SCS categories	Total no. of SCSs (n = 135)	No. and percent of SCSs who left the SCS (n = 53)
SCS position
Secretary General	16	6 (38%)
Deputy Secretary General	73	30 (41%)
Director of agency	46	17 (37%)
Gender
Female	31	14 (45%)
Male	104	39 (38%)
Age
Under 30	1	1 (100%)
30–39	56	19 (34%)
40–49	45	14 (31%)
50–59	20	12 (60%)
Over 60	12	6 (50%)
Unknown	1	1 (100%)
SCS experience
Under 5 years	70	26 (37%)
5–10 years	35	14 (40%)
11–15 years	24	12 (50%)
Over 15 years	5	0 (0%)
Unknown	1	1 (100%)

SCS: Senior Civil Service.

The Estonian civil service (for a PSB-based overview, see Pesti and Randma-Liiv, 2018) represents an agency-type PSB with characteristics of pragmatic, hybrid-type PSB (for types of PSB, see Hood and Lodge, 2006). Public service reform of 2012 that established a shift towards the managerial PSB is evidenced by the growing emphasis on flexibility in employment relations, the use of fixed-term contracts, increased private sector-style practices at all levels of the civil service, an emphasis on performance management, and the reduction of formal statutory employment protections. The managerial bargain implies more decision-making autonomy for SCSs in exchange for increased responsibility. The Estonian civil service system is very much in line with the private sector practice, as most employment conditions are the same for public and private sectors, and the service relationship for civil servants is organisation- and person-specific (Pesti and Randma-Liiv, 2018). As for the reward component of PSB, the Estonian system reflects the most the competition- and individual-based “turkey race” type of reward (for the components of PSB, see Hood and Lodge, 2006). The Estonian competency bargain resembles a “deliverer” type of PSB because managerial skills of SCSs are regarded as more important than technical skills, and in line with the managerial approach, result-orientedness is highly valued. The prevailing loyalty bargain is a “partnership”-type bargain according to which SCSs work together with elected politicians but have no separate identity either in directing public organisations or in determining policy. This implies that the SCSs’ loyalty to politicians tends to prevail over semi-autonomous performance based on the constitution and public interest. Personal (rather than political) loyalty to supervisors dominates as typical of the hybrid-type bargain (Pesti and Randma-Liiv, 2018).

The 1995 PSA did not stipulate the SCS as a distinct category although the recruitment and selection processes of SCSs differed from the rest of the civil service. The Selection and Evaluation Committee of Senior Civil Servants at the Government Office had an advisory role in screening and short-listing candidates for certain SCS positions. However, final decision-making power over the selection of the corresponding SCSs was left to the relevant minister. Senior civil servants have come to be treated as a special group for training and development purposes since 2004, and an SCS system evolved incrementally under the leadership of the Government Office (Randma-Liiv et al., 2015).

The new 2012 PSA (enforced in 2013) formalised the Senior Civil Service. It distinguished SCSs as a separate civil service category with special regulations for fixed-term service, recruitment, selection, assessment, and development, thus changing the previous PSB in Estonia. The nature of this reform, however, is still to be determined, and broader consequences for SCS turnover will yet to be seen since the fixed-term contracts with all senior civil servants were signed for a five-year period. This is the main reason why the current study focuses on the pre-reform years 2009–2013.

The positions covered by the Estonian SCS include the Secretary of State (1 person), Secretary Generals (11) and Deputy Secretary Generals (50) of the Ministries, and Director Generals of the executive agencies (boards and inspectorates) (28). The entire SCS is considered apolitical. In 2013, 96 positions belonged into this group and senior executives made up 0.4% of the civil service. This falls into an internationally recognised share of SCSs, ranging between 0.13 and 2.1% (Halligan, 2012: 116). Salaries are rather homogenised for Secretary Generals (about 3.5 times the Estonian average) and more individualised for Deputy Secretary Generals and Director Generals of executive agencies (in the latter case ranging from two to five times the Estonian average). Collective agreements cover only a marginal number of civil servants.

Until 2013, a great majority of SCSs had employment contracts for an undetermined period of time (with the exception of 11 Secretary Generals and a few other SCSs, who had five-year renewable contracts). As a rule, the person who has the right to end the contract with an SCS is the respective minister. In the case of Secretary Generals (and a few other SCSs, mostly related to the fields of security and safety), it is the Cabinet of Ministers which appoints and ends contracts with them. According to departmental hierarchy, we refer to a minister as a political superior (for all SCSs), and to a Secretary General as an administrative superior for Deputy Secretary Generals. A senior official may leave the civil service on his/her own will, due to the end of service term, as a consequence of structural changes leading to the abolishment of a respective position, because of incompatibility with job requirements (including shortage of necessary knowledge and skills), and as a result of disciplinary charges. In addition to this, Secretary Generals can be forced to leave in cases when the minister finds their collaboration difficult after one year of working together.

Method

The study addresses the following research question: What are the reasons for leaving the SCS in Estonia? The research question is addressed by the explorative, holistic and in-depth analysis of the Estonian SCS. The Estonian civil service system offers an interesting example because it is one of the most decentralised systems in Europe (Meyer and Hammerschmid, 2010) with much discretion delegated to single organisations and thereby to individual SCSs, as each ministry and executive agency is responsible for the recruitment, training, performance appraisal and pay of its officials. It is an extremely open system, where it is possible to enter the civil service without any restrictions on the senior level.

The analytical model of SCSs’ turnover will be further explored through an empirical study of the Estonian SCS during the five-year period of 2009–2013. The adoption of the new PSA determines the end year of the study; in addition, in choosing the period for the empirical study, the aim was to involve general elections (2011).

The empirical study was conducted in two stages. The first stage involved mapping the people who had left the SCS based on the data of the Government Office of Estonia and by complementing the existing official sources. The information provided by the Government Office turned out to be incomplete and unsystematised. Thus, the goal was to reliably map the entire turnover in the SCS through the use of bibliographical research covering all people who served in the Estonian SCS in 2009–2013. This was carried out by desk research mostly on the basis of Internet sources (e.g. websites of individual government organisations; publicly accessible pay information of SCSs). Data were cross-checked from different sources. In total, 135 SCSs were identified, out of whom 53 top executives had left the SCS in 2009–2013. The small size of the sample sets limits for using quantitative analysis, which is why only descriptive statistics are used.

The second stage involved analysing and systematising the reasons why the SCSs had left the service, based on semi-structured interviews. Out of 53 SCSs who left in the selected period, 37 agreed to be interviewed (70% response rate). Such a remarkable response rate adds specific value to the study. We are not aware of any study which could present interview data on turnover of such a high proportion of former SCSs. The sample included at least half from each sub-group of SCSs: four Secretary Generals, 23 Deputy Secretary Generals and 10 Directors of agencies. Ten respondents were female (71% out of female SCSs who left at that period) and 27 were male (69%). The age of respondents varied from 29 to 63, and the work experience in the SCS from half a year to 13 years.

In most cases, the interviews lasted an hour, and in some cases up to two hours. In a few cases, responses were later specified by phone or email. Semi-structured interviews made it possible to get a close insight into reasons for leaving the SCS. The limitation of the empirical study is that only the perspective of SCSs was addressed. PSB has two parties, but in this study, the viewpoints of respective political and/or administrative superiors were not considered. Also, leaving top executive positions is often sensitive and may involve factors that respondents would rather not reveal. Although the interviews were anonymous and the goal was to exclude a possible bias from the actual situation, we cannot fully eliminate the tendency that respondents may be inclined to portray involuntary leaving as voluntary, since they may be reluctant to reveal issues that can shed a negative light on themselves (see also Morrell et al., 2001: 221).

Overview of SCS turnover

Altogether, 135 people belonged to the group of SCSs in 2009–2013. Out of 135, 53 top executives left the SCS, 14 people changed a job within the SCS and 2 were temporarily away from the SCS (e.g. on maternity leave). Out of 53 departures, six left because their fixed-term contract was not renewed; the remaining SCSs had an agreement for undetermined time or left before the expiration of their fixed-term contracts. Table 1 provides an overview of those who left the SCS.

The Estonian SCS is in general characterised by a high turnover (39% during a five-year period). Different categories of civil servants are proportionally represented in the turnover, as seen from Table 1. The relatively high exit numbers among SCSs over 50 years old and those with more than 10 years’ experience within the SCS are somewhat surprising.

Changes in the numbers of SCSs are only weakly related to the changes of Ministers, general elections in 2011 and to change of a government coalition in 2009 (see Table 2), indicating that attention should be paid to not only to political but also to non-political reasons behind turnover.

Table 2.

Changes in the numbers of SCSs and ministers in 2009–2013.

Year	Changes in the number of Ministers (n = 12)	Changes in the number of SCSs (n = 96)	Remarks
2009	3	15	Change of a government coalition in May 2009
2010	0	2
2011	6	9	General elections in March 2011
2012	3	14
2013	0	13

SCSs: senior civil servants.

Semi-structured interviews shed light on the reasons of departure from the SCS. The collected data show that SCSs mostly left their job voluntarily (78% of respondents), but involuntary turnover also existed (22% of respondents). Two types of voluntary turnover were identified: leaving mainly due to “push” factors (62% of respondents) or due to “pull” factors (16% of respondents).

Involuntary turnover

Involuntary turnover has manifested itself in a “hidden” form as opportunities enabled by the institutional framework (e.g. not renewing fixed-term contracts, restructuring organisations) have been used to make SCSs leave. All respondents from this group admitted that they were willing to continue in the SCS, but they did not have an opportunity to do so. For example, one SCS was forced by a minister to voluntarily submit a request to leave, and two were told that they would not have a chance at getting a job in case they applied for their “own” renewable SCS positions after the end of their fixed-term contracts. According to the Estonian institutional framework, it is possible to announce an open competition or to renew the fixed-term contract with an incumbent without announcing an open competition. If the new competition was announced, the incumbents perceived this as a statement of distrust, and the majority did not participate in such open competitions. Only one did but did not succeed.

Two reasons were identified for involuntary turnover. First, most respondents in this group (six people) highlighted cooperation problems with a minister referring to political and/or personal (dis)loyalty. Such problems involved both conflicting views on specific policy areas and the role and expectations from an SCS. Respondents claimed that political leaders saw them as “silent” policy implementers rather than policy advisors. Consequently, the attempts of SCSs to “speak truth to power” often led to cooperation problems with political leaders. Different viewpoints of SCSs were sometimes perceived as personal disloyalty by political leaders. And, second, the (relatively) high age and long-term SCS experience were perceived as the main reasons for involuntary turnover in two cases. Political leaders had told these SCSs openly about their desire to bring in “fresh blood” into senior positions. Almost all SCSs accepted involuntary dismissal as they realised the depth of conflicts with political leaders.

Voluntary turnover led by push factors

Voluntary exit was considerably more frequent among the SCSs compared to involuntary dismissal. In the case of voluntary turnover, the decision to leave was based mostly on multiple reasons. Sixty-two percent of respondents left voluntarily, motivated by “push factors”. The main push factors as well as secondary motives for departure are summarised in Figure 3.

Figure 3.

Push factors of voluntary turnover.

Push factors can be divided into three groups: organisational push factors, push factors based on external environment and individual factors. Organisational push factors constitute the most frequent reason for turnover. The two most important organisation-related reasons were: (a) working too long in a certain position, organisation or policy area, which has caused exhaustion and tiredness and (b) cooperation problems with political and/or administrative superiors. Seventeen senior executives admitted that they had worked for too long in one job, organisation or policy area. This had led to a lack of perceived challenges in the current job, which, in turn, motivated them to seek for such challenges elsewhere. Several SCSs confessed that their departure was also necessary for organisational development as “subordinates stay in their ‘comfort zone’ with a long-serving superior” (interview with a Director of an executive agency). One SCS found that he was losing his chances in the labour market and decreasing his future employability when staying with the same organisation for too long.

When I joined the Ministry, I knew from the outset that I would like to work there for about five years in order to stay most productive. When the sixth year was running, I started asking questions what additional I could offer to the organisation. In parallel, I was afraid that I wouldn’t find another employment in the future if I stayed for too long in one position. (Interview with a Deputy Secretary General)

Another most frequent organisation-based push factor was poor cooperation with either political or administrative superiors (14 respondents). The main source of interpersonal conflicts with a political leader was related to disagreements on substantive policy matters.

As one civil servant put it:

I got tired by a new Minister. He wasted everybody’s time at the Ministry. He had several points in the (political) agenda which didn’t make any sense in the context of the legal framework of the Estonian state and which didn’t fit together with other policies. Ministers are temporary, and I could have waited until he leaves but I didn’t have the patience and I left myself. (Interview with a Secretary General)

SCSs got demotivated by goals which were derived from minister’s politically determined populist or personal agenda and which were perceived as unnecessary by SCSs. In some cases, this led to personal conflicts, distrust and loyalty issues, which were accompanied by reducing SCSs’ discretionary power and decision-making autonomy. Even when the political leader did not force the departure, several SCSs decided to voluntarily leave under such circumstances. Cooperation problems with administrative leaders were mostly based on their unfitting management styles – on the one hand, too little autonomy and decision-making discretion on the lower administrative level, and on the other hand, too little steering by Secretary Generals. The other organisational-level push factors were only occasionally mentioned by respondents (see Figure 3).

Around a third of respondents highlighted push factors based on the external environment. This included the low reputation of civil service as a whole, and the general publicness of SCS reflected by negative media coverage. In addition to this, some SCSs brought up individual reasons leading to the departure from the civil service. These included relocation and the motivation to do more specialist-type work instead of administrative responsibilities.

Voluntary turnover led by pull factors

The smallest share of interviewed SCSs (16%) opted for voluntary exit based on pull factors related to alternative employment opportunities. All people in this group remained in the public sector. The pull factor mentioned most often was related to the challenge that the new job offered. Half of the respondents in this group moved to an important job within the same policy area and linked their move to a commitment to the policy area. A few respondents perceived their new jobs within the public sector as more attractive (“step ahead”) compared to their existing SCS positions and some because the new jobs offered more managerial autonomy. Other rarely mentioned pull factors included the opportunity to do a more specialist-type job, a reduced workload, a better team, more generous monetary rewards, more opportunities to openly express his/her opinion, less (or more) responsibility.

Half of the respondents in this group decided to leave despite the fact that they were satisfied with the SCS job. The other half also mentioned some push factors that contributed to the decision to leave (shortage of new challenges in their SCS positions, dissatisfaction with political choices either in substantive policy matters or in operational management, routine and tiredness related to too lengthy experience in one position, intangible results, negative reputation and media coverage).

Discussion

The concept of PSB proves useful for understanding and systematising the causes of turnover, distinguishing between voluntary and involuntary turnover and explaining reasons of exit which are related to expectations of representatives of the political system and SCSs themselves. However, PSB does not pay sufficient attention to voluntary exit, intra-organisational, interpersonal and individual factors, as well as external contextual variables such as reputation of civil service as an employer in a given society and alternative employment opportunities which are of key importance in studying turnover (and PSBs in general). This study of turnover thus shows that while PSB proves a valuable approach for general conceptualising of SCS turnover, the PSB-based approach turns out to be complex and multi-faceted by containing so many different dimensions that it is somewhat problematic to apply it as a basis of an empirical study. This is the reason why for its further operationalisation, one has to seek for help from other models such as “pull” and “push” factors in this study.

The empirical study at hand demonstrates that one, albeit not the most important, reason to exit the Estonian SCS has been problematic cooperation with immediate political superiors, which has led to either voluntary or involuntary departure from the SCS. In academic literature, turnover of SCSs has predominantly been related to political change and political loyalty of SCSs (Bach and Veit, 2018; Christensen et al., 2014; Rouban, 2012). At the same time, a relatively stable political situation in Estonia in a selected period (new ministers belonged predominantly to the same political party) and low politicisation of the Estonian civil service in general (Meyer-Sahling, 2011) indicate that personal loyalty – including personal fit and trust among political leaders and SCSs – is as important as political loyalty in analysing SCSs’ turnover. It is possible to conclude on the basis of interviews that there can be a reciprocal relationship between personal and political loyalty: conflicts based on political choices (political loyalty) can lead to personal conflicts (personal loyalty), and the other way round.

The current study shows that distinguishing between political and personal loyalty can be cumbersome. It would require further operationalisation, measuring and capturing each type of loyalty, where PSB, however, runs short in offering empirically relevant tools. One may, for example, look at the timing of turnover by exploring the correlation between SCS turnover and changes in government (as provided in Table 2) to detect politically reasoned turnover. However, this is clearly insufficient since problems with political loyalty can also occur and lead to SCS turnover without a change in government. Institutional framework could shed further light on loyalty issues by contributing to operationalisation of relevant principal–agent relations and determining who is in charge of recruiting/dismissing SCSs, and when and on what grounds dismissal can take place, by that showing the administrative power of individual political masters over their subordinates. In addition, not all SCSs are directly subordinated to political masters, but they may as well be subordinated to higher-tier (apolitical) SCSs which may reduce the likelihood of political loyalty and rather refer to interpersonal issues and management styles (personal loyalty).

Loyalty-related cooperation problems in Estonia are closely linked to the broader understanding of roles that SCSs are expected to perform. For example, the case study shows that the reason for both voluntary and involuntary departures from the SCS has been the occasion when an SCS has disagreed with a political superior, i.e. did not meet the expectations of a political leader who assumed that the SCS should silently implement his/her decisions and requests. Such an expectation is in conflict with meritocratic principles of modern civil service, where SCSs not only have the right but also the obligation to provide political leaders with professional advice (Eichbaum and Shaw, 2008: 339). Limiting the role of SCSs to simple policy implementation and punishing them for “speaking truth to power” can endanger the independence and neutrality as core values of the civil service.

Conflicts deriving from different understandings of loyalty and the role of SCSs may refer to the fact that neither Estonian politicians nor SCSs themselves may fully realise the role of civil servants and, particularly, of SCS in democratic governance. There is no coherent understanding of politico-administrative roles in Estonian society (see also Tammekänd, 2004). This can be explained by problems of a (still) newly democratic state where civil service development has been carried by legalistic-technocratic motives (see also Randma-Liiv et al., 2015), and a broader discourse on politico-administrative roles has been non-existent. It is expected that SCSs base their performance and behaviour on the constitution and public interest on one hand and on the will of democratically elected political representatives on the other hand. Conflicts can easily arise if these happen to differ from each other, or if different parties assign different weight to them. It occurred from the study that the perceptions of roles of SCSs can vary considerably, and even ministers belonging to the same political party can have very different expectations from SCSs (for similar findings, see Althaus and Vakil, 2013: 482; Steen and van der Meer, 2011: 221). This shows that in terms of role understanding, the Estonian civil service system gives predominant importance to individual (unwritten) PSBs among SCSs and representatives of the political system. If a new SCS or minister does not adjust to expectations of another party, it may lead to conflicts and exit from the SCS.

Although the Estonian case study shows that the concept of PSB offers generally a good basis for studying involuntary turnover, it is somewhat limited with regard to explaining voluntary exit from the SCS, which was found to clearly dominate in Estonia (78% of all departures). For understanding voluntary turnover, push and pull factors as well as individual factors complement the PSB approach. For example, the main reason for SCS turnover in Estonia – tiredness of working too long at the same institution/position – is beyond the scope of PSB. The Estonian case study suggests that for the operationalisation of voluntary turnover, the distinction between push and pull factors proves useful. It is, however, difficult to draw a clear line between push and pull factors, as most often the exit decision is based on the combination of the two.

The Estonian SCS operates in the context of an open position-based civil service system with no institutionalised tenure and limited opportunities for top executives to move within the SCS, which is, among other reasons, explained by the very small size of the SCS. Significant SCS turnover of 39% within five years provides a number of challenges to the government, including loss of institutional memory and competence, loss of organisational investments related to the recruitment and selection of civil servants, and expenses related to finding and appointing new public sector managers (see also Cho and Lewis, 2012; Lee and Whitford, 2008; Moynihan and Landuyt, 2008). In addition, high turnover represents a serious problem for a highly segmented system which is dependent on collaboration among civil servants. A frequent change of partners in the top ranks creates an unfavourable context for collaborative action and poses considerable obstacles to developing common SCS culture. High turnover in the SCS may also undermine considerable efforts put into the training and development of Estonian SCSs (see Randma-Liiv et al., 2015) when executives use the competencies obtained for career-building in the private sector or in international organisations.

This study thus underlines the importance of motivating and retaining top executives. It appears that during the period studied, turnover of SCSs was not sufficiently addressed by the Estonian government (to the extent that the government even did not collect or monitor relevant data). It is somewhat alarming that a great majority of studied SCSs (three quarters) did not receive any proposal for staying in the SCS or applying to other SCS vacancies in the future. Moreover, the study found beyond its main focus that many former SCSs had specifically decided not to continue in the SCS (and in the public sector in general). This indicates that the Estonian PSB may not be properly balanced, as the reward and autonomy offered to SCSs do not optimally compensate for the expectations of loyalty and competence.

From the perspective of PSB, the absence of tenure refers to job insecurity, which can be considered part of SCSs’ rewards package. The Estonian civil service system, which does not offer (mandatory) rotation within the SCS and where there is no guarantee about the future job within the SCS or in the larger public sector, leads to job insecurity, which, in turn, contributes to clinging to a particular SCS position for a long time (and the accompanying tiredness and boredom). Efficient rotation opportunities within the SCS and outplacement services within the larger public sector may alleviate these problems, which was also strongly proposed by distinguished respondents of this study. Altogether, the smaller the job security, the more it should be compensated by other rewards (such as pay) and autonomy in order to retain an optimal balance of PSB. While civil service systems which are based on life-long career and tenure address such problems per se (Kuperus and Rode, 2008: 13), in open position-based systems, the motivation and retaining of SCSs deserves particular attention. Insufficient rewards and autonomy of SCSs may contribute to a vicious cycle of temporary SCSs including problems with recruiting new SCSs and a further increase in their turnover, as well as a loss of political craftsmanship among SCSs (Bach and Veit, 2018), ultimately leading to a “temporary state”.

Conclusions

This article shows that the concept of PSB can help to understand and systematise the causes of leaving from the SCS. PSB proves useful in distinguishing between voluntary and involuntary turnover and in linking the turnover with political–administrative relations. The Estonian case study indicates that PSB provides a robust framework for explaining involuntary turnover, disagreements in the perception of the roles of SCSs and individual vs. institutional PSBs. Since PSB does comprise voluntary exit of SCSs based on pull (and some push) factors, other approaches known from management literature, especially organisational, environmental and individual factors, are relevant for the operationalisation of voluntary turnover and complementing the PSB-based model for researching SCS turnover.

While in previous studies turnover of SCSs has predominantly been related to politicisation and political loyalty of SCSs, the Estonian case shows that personal loyalty – including personal fit and trust among political leaders and SCSs – is as important as political loyalty in analysing SCSs’ turnover. Consequently, the turnover of SCSs can be considerable even without much direct political influence, as shown in the case of Estonia. Job insecurity, combined with the domination of individual unwritten PSB, tends to lead to ambiguity in the perception of the roles of top executives, which, in turn, causes conflicts and dissatisfaction, materialising in high voluntary turnover.

This study opens several avenues for future research. First, while this article focused on a five-year period, further analysis of turnover could benefit from a more longitudinal approach in order to better test linkages between SCS turnover and political change. Second, the distinction between political and personal loyalty deserves further attention, and the concept of loyalty in politico-administrative relations should be further operationalised by developing appropriate tools for the measurement of political and personal loyalty. Third, in order to receive a more comprehensive overview of reasons of turnover, further research could also cover the other side of the bargain, i.e. representatives of the political system, especially when studying involuntary exit from the SCS. Finally, in the case of Estonia, the recent adoption of the new PSA with invariably fixed-term contracts for SCSs provides an opportunity for comparing findings of this study with the new institutional arrangement.

Footnotes

Declaration of conflicting interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research leading to these results has received funding from the European Commission’s grant 587498-EPP-1-2017-1-IS-EPPJMO-NETWORK and from the Estonian Research Council’s institutional grant IUT19-13.

References

Ali Shah

Fakhr

Ahmad

, et al.(2010) Measuring push, pull and personal factors affecting turnover intention: A case of university teachers in Pakistan. Review of Economic and Business Studies 3(1): 167–192.

Althaus

Vakil

(2013) Political transitions: Opportunities to renegotiate the public service bargain. Canadian Public Administration/Administration publique du Canada 56(3): 478–490.

Bach

Veit

(2018) The determinants of promotion to high public office in Germany: Partisan loyalty, political craft, or managerial competencies? Journal of Public Administration Research and Theory 28(2): 254–269.

Boeker

(1992) Power and managerial dismissal: Scapegoating at the top. Administrative Science Quarterly 37(3): 400–421.

Bourgault

(2011) Canada’s senior public service and the typology of bargains: From the hierarchy of senior civil servants to a community of “controlled” entrepreneurs. Public Policy and Administration 26(2): 253–275.

Boyne

James

John

, et al.(2010a) Does political change affect senior management turnover? An empirical analysis of top-tier local authorities in England. Public Administration 88(1): 136–153.

Boyne

James

John

, et al.(2010b) Does public service performance affect top management turnover? Journal of Public Administration Research and Theory 20(2): i261–i279.

Burns

Wei

Peters

(2013) Changing governance structures and the evolution of public service bargains in Hong Kong. International Review of Administrative Sciences 79(1): 131–148.

Cho

Lewis

(2012) Turnover intention and turnover behavior: Implications for retaining federal employees. Review of Public Personnel Administration 32(1): 4–23.

10.

Christensen

Klemmensen

Opstrup

(2014) Politicization and the replacement of top civil servants in Denmark. Governance: An International Journal of Policy, Administration, and Institutions 27(2): 215–241.

11.

Demmke

(2005) Are Civil Servants Different Because They Are Civil Servants?, Maastricht: European Institute of Public Administration.

12.

Eichbaum

Shaw

(2008) Revisiting politicization: Political advisers and public servants in Westminster systems. Governance: An International Journal of Policy, Administration, and Institutions 21(3): 337–363.

13.

Halligan

(2012) Leadership and the senior service from a comparative perspective. In: Peters

Pierre

(eds) Handbook of Public Administration, 2nd ed. London: Sage, pp. 115–129.

14.

Halligan

(2013) The evolution of public service bargains of Australian senior public servants. International Review of Administrative Sciences 79(1): 111–129.

15.

Hansen

Salomonsen

(2011) The public service bargains of Danish permanent secretaries. Public Policy and Administration 26(2): 189–208.

16.

Hondeghem

(2011) Changing public service bargains for top officials. Public Policy and Administration 26(2): 159–165.

17.

Hood

Lodge

(2006) The Politics of Public Service Bargains. Reward, Competency, Loyalty – and Blame, New York: Oxford University Press.

18.

Hwang

S-I

Kuo

J-H

(2006) Effects of job satisfaction and perceived alternative employment opportunities on turnover intention – An examination of public sector organizations. The Journal of American Academy of Business, Cambridge 8(2): 254–259.

19.

Kim

(2007) Transforming higher-level civil service in a new age: A case study of a new senior civil service in Korea. Public Personnel Management 36(2): 127–142.

20.

Kuperus

Rode

(2008) Top Public Managers in Europe. Management and Working Conditions of the Senior Civil Servants in European Union Member States, Maastricht: European Institute of Public Administration.

21.

Lee

S-Y

Whitford

(2008) Exit, voice, loyalty, and pay: Evidence from the public workforce. Journal of Public Administration Research and Theory 18(4): 647–671.

22.

Llorens

Stazyk

(2011) How important are competitive wages? Exploring the impact of relative wage rates on employee turnover in state government. Review of Public Personnel Administration 31(2): 111–127.

23.

McElroy

Morrow

Rude

(2001) Turnover and organizational performance: A comparative analysis of the effects of voluntary, involuntary, and reduction-in-force turnover. Journal of Applied Psychology 86(6): 1294–1299.

24.

Meier

Hicklin

(2008) Employee turnover and organizational performance: Testing a hypothesis from classical public administration. Journal of Public Administration Research and Theory 18(4): 573–590.

25.

Meyer

Hammerschmid

(2010) The degree of decentralization and individual decision making in central government human resource management: A European comparative perspective. Public Administration 88(2): 455–478.

26.

Meyer-Sahling

J-H

(2011) The durability of EU civil service policy in Central and Eastern Europe after accession. Governance 24(2): 231–260.

27.

Ministry of Finance (2017) Public Service Yearbook 2016. Available at: www.rahandusministeerium.ee/et/riigi-personalipoliitika/personali-ja-palgastatistika (accessed 3 October 2018).

28.

Morrell

Loan-Clarke

Wilkinson

(2001) Unweaving leaving: The use of models in the management of employee turnover. International Journal of Management Reviews 3(3): 219–244.

29.

Moynihan

Landuyt

(2008) Explaining turnover intention in state government: Examining the roles of gender, life cycle, and loyalty. Review of Public Personnel Administration 28(2): 120–143.

30.

Moynihan

Pandey

(2008) The ties that bind: Social networks, person-organization value fit, and turnover intention. Journal of Public Administration Research and Theory 18(2): 205–227.

31.

Pesti

Randma-Liiv

(2018) Towards a managerial public service bargain: The Estonian civil service reform. NISPAcee Journal of Public Administration and Policy 11(1): 135–154.

32.

Peters

Pierre

(2004) Politicization of the civil service: Concepts, causes, consequences. In: Peters

Pierre

(eds) Politicization of the Civil Service in Comparative Perspective: The Quest for Control, New York: Routledge, pp. 1–13.

33.

Randma-Liiv

Uudelepp

Sarapuu

(2015) From network to hierarchy: The evolution of the Estonian senior civil service development system. International Review of Administrative Sciences 81(2): 373–391.

34.

Rouban

(2012) Politicization of the civil service. In: Peters

Pierre

(eds) Handbook of Public Administration, 2nd ed. London: Sage Publications, pp. 380–391.

35.

Salomonsen

Knudsen

(2011) Changes in public service bargains: Ministers and civil servants in Denmark. Public Administration 89(3): 1015–1035.

36.

Selden

(2006) The impact of discipline on the use and rapidity of dismissal in state governments. Review of Public Personnel Administration 26(4): 335–355.

37.

Steen

van der Meer

(2011) Public service bargains in Dutch top civil service. Public Policy and Administration 26(2): 209–232.

38.

Tammekänd

(2004) Tippametnike politiseerumine – kas ohu märk või loomulik areng? (Politicization of senior civil servants – A sign of danger or natural development?). Riigikogu Toimetised 9: 117–122.

39.

Tekniepe

Stream

(2012) You’re fired! county manager turnover in large American counties. The American Review of Public Administration 42(6): 715–729.

40.

Wheeler

Gallagher

Brouer

, et al.(2007) When person-organization (mis)fit and (dis)satisfaction lead to turnover. The moderating role of perceived job mobility. Journal of Managerial Psychology 22(2): 203–219.