Western Australia’s Local Government Act 25 years on and under review: A qualitative study of local government Chief Executive Officers

The Local Government Act 1995 (WA) , which prescribes local governments’ (LGs) powers and responsibilities in Western Australia (WA), is undergoing a major review. This is likely to have significant impact on LGs and the communities they serve. Qualitative data from 43 (31%) WA LG Chief Executive Officers (CEOs) were obtained and analysed to inform the review. Structural issues emerged across three points of conjunction: the interfaces between (i) state–local government, (ii) councillors–CEO and (iii) councillors–community. The state–local government conjuncture related to the perceived need for state government to address harass-ment towards LGs and staff. The councillors–CEO interface revealed interconnected issues that predis-posed these political–administrative branches to conflict. The principle conflict was the statutory requirement for CEOs to report councillors for legislative breaches (whistleblowing) while depending on councillors abilities of councillors regarding human resource management. The councillors–community interface related to the lack of role clarity regarding councillors’ responsibilities. Overall, the key concern was the tension at the councillors–CEO interface. Practical recommendations include structural and legislative changes to reconcile identified issues. This research provides useful insights that may improve the functioning of LGs.


INTRODUCTION
Australian local governments (LGs) advance state and national priorities while being empowered to act in the best interest and promote the overall well-being of their communities (Grant & Drew, 2017;Wensing, 1997). As such, LGs play an important role in the daily lives of their citizens, routinely influencing social, environmental, economic, and cultural matters (Dollery, Wallis, & Allan, 2006;Hutchinson, Walker, & McKenzie, 2016). Contemporary LGs are complex organisations with multi-million dollar budgets, delivering a broad range of services (Dollery et al., 2006;Grant & Drew, 2017). In Western Australia (WA), LGs have a collective annual income of $4.27 billion and public assets valued at over $45 billion (2018-2019) (DLGSC, n.d.). Yet despite their important role, LGs are not formally recognised in the Commonwealth Constitution, but rather are recognised in their respective state constitutions, to varying degrees (Grant & Drew, 2017). Accordingly, the powers and functions of LG are prescribed by legislation at the state level.
LG in WA is in the midst of a major review that entails the creation of a new statute that will supersede the Local Government Act 1995 (WA) (the Act). This review is part of a national trend of LG reassessment given that all states and the Northern Territory have recently conducted, or are in the process of conducting, their own major reviews (DLGSC, 2019). As LG legislation is usually the largest and most complex of state legislation, reviews that invite a full rewrite are a rare and notable occurrence (Wensing, 1997). The last time this happened in WA was 25 years ago when the current Act was introduced.
The purpose of the present study is to explore LG Chief Executive Officers' (CEOs) perceptions of the Act and suggestions for revisions. As such, this research makes several contributions. First, there is a notable dearth of scholarship exploring LG CEOs' views of the legislative framework under which they operate (Svara, 2006a). This gap in the literature is concerning because LG CEOs are in a unique position to comment on LG legislation given their (i) deep technical knowledge of the statutes, (ii) role as leaders of LG administrative teams, and (iii) role as the lynchpin between the elected members and administrative body of LG. Second, there is scant literature on LG CEOs in Australia (Hutchinson et al., 2016). This gap in the literature is notable given the critical role LG CEOs play in the operation of LGs. Third, the present study is the first to examine the potential impact of the Act on LG CEO well-being. The results reveal structural issues related to the Act that have fostered perceptions of job insecurity. Job security is the subjective psychological experience regarding the surety of employment (Emberland & Rundmo, 2010;D'Souza et al., 2005) and is an important source of job stress (Cheng, Chen, Chen, & Chaing, 2005) that may have important implications for CEOs' overall health (D'Souza et al., 2005;D'Souza, Strazdins, Lim, Broom, & Rodgers, 2003). As such, the present study also contributes to the nascent literature on LG leader well-being (Kaluza, Boer, Buengeler, & van Dick, 2020).

A brief overview of the Local Government 1995 Act (WA)
This Act was a significant departure from previous LG legislation as it embraced 'New Public Management' (NPM; see Hood, 1991;Lohrey, Horner, Williams, & Wilmshurst, 2019), reflecting a trend in major reforms that occurred across all Australian LG jurisdictions between 1989 and 1995 (Aulich, 1999). NPM advocates the public sector to emulate private industry's keen focus on efficiency, effectiveness, and accountability, as well as role separation and delineation (Grant & Drew, 2017). Under NPM, responsibility for operational matters is devolved through the public service, with oversight and accountability through formal reporting and performance review processes (Aulich, 1997). These changes impacted LG relationships at two key levels. First, between state and LG as prior to the Act, the state mandated the activities to be performed by LG. However, WA's 1995 Act and the equivalent LG statutes enacted in the other states from 1989 devolved a degree of responsibility to LGs under the legal doctrine of 'general competence' (Wensing, 1997). For the first time, LGs had a level of discretion to decide what services they provided to their community in addition to those mandated by the state. Accordingly, strategic deliberation and planning -the strategic, reiterative process of identifying community needs, proactively planning services and infrastructure for current and future residents, ratepayers, and businesses within resource constraints (Kemp, 2017) -became highly salient to LG. The partial independence granted to LGs to determine what services they provided to their communities led to the second relationship change, as the Act articulated that the strategic domain was the role of LG councillors. It was a fundamental shift for councillors as their managing role became a governing role or, as is more colloquially phrased, steering rather than rowing (Grant & Drew, 2017;Newnham & Winston, 1997). Further, the Act specifies that the administration of the LG is the domain of the CEO. In other words, the Act created a structural separation of roles between the political and administrative branches of LG; this separation of powers is directly linked to the NPM philosophy (Grant & Drew, 2017). Svara (2006b) has observed that relations between the political-administrative branches of government vary along two axes. Namely, (i) how much control elected officials have over the administration, and (ii) the degree of differentiation between the roles of elected officials and administrators. He claims that extremely high control or extremely low control by elected officials over the administration is problematic. For example, extremely low control by elected officials in conjunction with very high differentiation between the roles of elected officials and administration can result in an autonomous, bureaucratic regime. Yet, a very high level of control over administration combined with high differentiation of roles can result in isolated administrators, or if low role differentiation, with manipulated administrators. However, Svara (2006b) states that NPM aligns with a mid-to-high level of control of administrators by elected officials that leverages the different lens that each embodies, which he terms the 'separate role' model. For example in policy-making, administrators contribute technical expertise, whereas elected officials (councillors) apply values to the policy-making (Svara, 2006b). The elected officials apply a mid-to-high level of control over administration through setting broad goals, policy-making, allocating resources, and general oversight and administrators have the discretion to select appropriate means to achieve these broad goals (Svara, 2006b). However, this 'separate role' model assumes that elected officials trust the expertise of the LG administration and are 'eager to incorporate administrative knowledge and skills into the policy-making process' (Demir, 2014, p. 507) and thus the separate roles are complementary rather than competitive (Svara, 1990).
Thus, in Australia generally as well as in WA in particular, grounded in the NPM philosophy, the roles for LG councillors and administration have evolved, at least in theory. Yet the current legislation prescribes the councillor role at the macro level; it does not provide granular detail regarding what the role includes or excludes. This may have contributed to the lack of clarity and consensus on the role that councillors have in LG and a resulting tendency for role tension between councillors and CEOs (MLGRP, 2012;Newnham & Winston, 1997;Tan, Morris, & Grant, 2016).
Although revisions to the Act are underway, the state has already introduced some changes to ensure LGs meet current community expectations for standards and performance. For example, the WA government has proposed compulsory training for councillors on governance and finance. With this said, research examining the training topics most useful to the councillor role suggests there is much left to be desired (Davy & Brands, 2012). Their research found that the two topics Australian councillors perceived to be most relevant for their role were 'strategy and vision' and 'community engagement', both of which are currently absent from the state's compulsory training program, despite strategy being a key role for councillors (Tan et al., 2016).
Another mechanism that the state government has proposed to meet societal expectations is instituting standards and guidelines for the recruitment, selection, performance reviews, and employment termination of LG CEOs (DLGSC, 2019). Although LG CEOs employ, and are responsible for, all staff in the organisation, the Council (the collective group of the elected councillors) are the employers of just one LG staff member -the CEO. Councillor involvement in the management of CEOs has been problematic. One such pivotal activity is CEO recruitment and selection, in which some councillors have exhibited bias towards a stereotyped leadership in relation to age and gender (Fox & Broussine, 2001;Hutchinson et al., 2016;Smith-Ruig, Grant, & Sheridan, 2016), predetermined personality, and political stance (Diamond, 2007;Hutchinson, Walker, & McKenzie, 2014;Suleiman, 2003). Therefore, LGs are potentially not employing the most competent candidate for the highly important CEO role.
Although the Act has undergone revision in the past and the state government has attempted to address perceived deficiencies, there remain a number of identified issues with the Act and its subsidiary legislation (DLGSC, 2019;MLGRP, 2012). With this said, the majority of the existing evidence for the issues with the Act are from the community members, organisations, LGs (as organisations), and state bodies (DLGSC, 2019), meaning the personal perspective of LG CEOs is lacking. The present study seeks to address this gap by giving voice to LG CEOs.

METHOD
The Local Government Professionals Australia WA commissioned this independent and exploratory study of the factors impacting LG CEO well-being. 1 This study therefore sought insight from the experiences of LG CEOs and followed a qualitative approach (Creswell, Hanson, Plano, & Morales, 2007). As the research question was contextually sensitive and exploratory, researchers initially conducted focus groups with 37 LG CEOs to formulate interview questions. Results of the focus groups strongly suggested the CEOs perceived issues with the Act and believed it contributed to tensions between the CEOs and Councils as well as lacked reach in terms of addressing antisocial behaviour from members of the general public. This observation guided the development of the semi-structured interview questions. The study was approved by the University of Western Australia Human Research and Ethics Office (RA/4/20/4807).

Participants
The study used a purposive sample of current and recently departed WA LG CEOs as they had in-depth knowledge and first-hand experience operating under the Act. At LG conferences and by personal email invitation, CEOs were invited to participate in the study by the independent research team. Forty-three WA LG CEOs were interviewed: 39 males and four females. All participants identified as Caucasian. Ten participants were CEOs of metropolitan governments and 33 were from regional LGs. This represents 33% and 31% of metropolitan and regional WA LGs, respectively, and overall, 31% of the 134 WA LGs (this excludes the two Federal external territories by the WA coast; Shires of Cocos (Keeling) Islands and Christmas Island (DLGSC, n.d.). Most participants held a post-graduate certification (e.g. graduate certificate, Masters, PhD; 58%), followed by Bachelor degrees (14%), Diplomas (7%), and High School (5%; missing 16%). The LGs that the participants managed in their current/most recent employment represented the smallest to largest WA LGs, as assessed by the numbers of full-time equivalent staff. Table 1 reports CEO demographics. Forty-three CEOs subsequently participated in semi-structured interviews (Barriball & While, 1994). Ten (23%) of those interviewed had participated in the earlier focus groups. The interview questions (Appendix A) asked CEOs for their recommendations for changes to the Act. Interviews were professionally transcribed verbatim.

Analytical strategy
ATLAS.ti was used to aid data management and analysis (Friese, 2014). Analysis occurred according to framework analysis procedures. Framework analysis provides a systematic and rigorous means of analysing qualitative data, particularly data relevant to policy questions, while providing transparency through an audit trail (Dixon-Woods, 2011;Ward, Furber, Tierney, & Swallow, 2013). Initial coding identified utterances (quotations) from the interviews relevant to potential changes to the Act. These became codes, the basic unit of analysis. In total, there were 244 codes generated from the interviews. A theoretical framework was developed by identifying codes of similar meaning that were grouped into tentative categories. The codes and categories were subjected to constant comparison to refine the characteristics of categories and merge/split/create new categories and re-assign codes to categories as necessary to ensure the consistency of codes within each category and thus ensure category integrity. Axial coding was employed to identify the relationships between categories and sub-categories. The WA state government has developed a series of discussion papers outlining potential revisions of the legislation. These include issues of cost-shifting (referred to as 'unfunded mandates' in the United States), 'regulation' of councillor conduct, and creating different levels of LG governance requirements to better recognise the different levels of capacities across LGs (DLGSC, 2019). Although these themes emerged in our own data, we focused on the unique insights derived from the LG CEOs; these were around three points of interaction. The findings are contextualised within the existing literature on LGs to offer pragmatic recommendations for change that have not been discussed elsewhere.

RESULTS AND DISCUSSION
Eight themes, not otherwise covered in the state's publicly distributed discussion papers for the review of the Act (DLGSC, 2019), emerged from analysis of the data. These themes related to three points of interaction between different LG stakeholders ( Table 2). The points of interaction were the interfaces between: (i) state and local government, (ii) councillors and the CEO, and (iii) councillor interaction with the community.

Process to deal with 'persistent' complainants
The theme that emerged within this point of interface related to the management of complaints to the LG. CEOs reported that a small but persistent group of residents/groups placed LGs under 'unwarranted' pressure due to their high volume of complaints related to issues the CEOs believed had already been addressed or were without foundation. CEOs reported their responses to these complainants to provide the facts did not resolve the issue, and often escalated to accusations and even threats against the CEO and staff. This often resulted in CEOs perceiving their workplace to be unsafe for their staff, themselves, and their own family. Although regional CEOs perceived that personal attacks on their family and at their residence were unique to regional LGs, the data indicated it was endemic to both regional and metropolitan LGs.
Had to lock the place down when someone was getting quite aggressive with other staff. Had to call the police. Had the same happen several times (Participant 14).
It's not just the CEO, but it's the families that are impacted and that is certainly going well beyond the line. Being harassed outside your home (Participant 44). Although the current LG legislation has provisions to address undesirable behaviour of LG councillors and staff, the legislation does not include the actions of people not employed in LG. CEOs expressed a desire for an external mechanism to review such cases. This would allow a LG, where warranted, to legitimately refuse to further correspond with an aggrieved resident/group regarding the issue without further accusations from these parties, or media, of being non-responsive or failing to listen to community members. Such accusations were perceived to be a tactic employed to harm LGs and divert public resources into dealing with non-productive issues.
We don't want to have a point where people are too scared to complain, because there are some legitimate reasons . . . But we need to be able to deal equally severely with those that are out there simply to disrupt . . . legislation needs to be tighter to deal with vexatious complainants and the ramifications to someone who is proven to put a vexatious complaint in (Participant 11).
The ability to declare some residents vexatious and to prevent them from being able to continually bully and harass staff (Participant 20).
Due to the face-to-face contact with the public inherent in many LG roles, LG has a higher likelihood of being afflicted with aggression (Chappell & Di Martino, 2006) from those who are dissatisfied or irritated with the service or a decision of the LG (Mullen, 1997). Some scholars have argued that NPM's recasting of citizens as consumers may have engendered in ratepayers a consumer perspective focused on self-interest over a community orientation (Clarke, Newman, Smith, Vidler, & Westmarland, 2007;Halligan, 2012). Inherent in a community perspective is the understanding that decision-making for the community involves considering the whole community, balancing sometimes conflicting goals such as the needs or preferences of different demographics within the community, which often results in a concession by some in the society in order to facilitate a 'necessary' gain for others (Kemp, 2017). One CEO in this study also referenced the shift from communal considerations to the individualist perspective.
I think what [has been lost] is the whole sense of a society, a village, we're all in this together . . . what happens is that people pay rates and it's like buying a product. And like putting 20 cents into, or a dollar into a vending machine to get a can of Coke . . . If you don't get the product or service you want, you kick the machine, you shake the machine (Participant 1).
Research undertaken in Australia compared 'persistent complainants' with a control group of complainants (Lester, Wilson, Griffin, & Mullen, 2004). Persistent complainants often sought outcomes that could not be provided by the organisation. Over 50% of the persistent complainants made veiled or direct threats to staff, compared to no threats from the control group. The average time of interaction from persistent complainants to the organisation was also significantly lengthier (35 months versus 8 months).
However, other than a couple of bodies that cover specified domains, such as the Information Commissioner and the WA Ombudsman, there is a lack of an external arbiter that can address the complaints of residents who are dissatisfied with an LG's actions. Dissatisfied complainants take their issues to multiple authorities at state level which begins the complaint cycle anew (Lester et al, 2004;WA Ombudsman, 2009) and thus consumes state government resources as well. According to the Ombudsman, although persistent complainants are less than 1% of their complainants, they consumed between 15% and 30% of their resources (Lester et al, 2004). One WA LG reported that over a 14-month period it had spent $178,000 in staff time responding to information requests from 10 residents/groups (CoM, 2017). Furthermore, the drawn-out process can also be detrimental to the complainants' own mental health and social lives (Mullen & Lester, 2006), suggesting a double-edged sword effect that may harm the LGs and the complainants themselves.

3.1.2
State-local government interface: Practical implications -Recommendations for change Although the CEOs did not directly make a suggestion to address this issue, one of the CEOs alluded to a potential solution. The CEO noted that although members of the public complaining about LG had multiple avenues for action (e.g. state politicians, ombudsman, and the public sector commission), 'I'm not aware of anywhere where the local government can go for assistance [for resolution with persistent complainants], apart from the courts' (Participant 8). Therefore, a recommendation may be for the state government to delegate a body to which LG referrals can be made about persistent complaints to evaluate the issue and identify a course of action (e.g. allow the LG to refuse further correspondence). Such a system would aim to provide closure to both parties, prevent issues from escalating to aggressive behaviour (Mullen & Lester, 2006), and limit the consumption of public resources at local and state government levels.

CEO and councillor interface
Research in Australia indicates that the legislative structure allows mayors and councillors to be arbitrarily subjective in the selection of the CEO (Diamond, 2007;Hutchinson et al., 2014;Smith-Ruig et al., 2016). The present research finds that the precarious nature of employment is continued once the CEO is appointed and is inadvertently exacerbated by current legislation. This study found that the most salient issue to CEOs was job insecurity created from the requirement of CEOs to report councillors for legislative breaches (whistleblowing) while also being dependent on these Council members for their own employment. This perceived vulnerability has arisen from six interconnected issues. First, CEOs are legislatively required to report councillors, their employer, for suspected breaches of legislation. The second issue relates to the length of CEOs' contracts and severance payments. The third issue pertains to the level of councillors' skills, experience, and knowledge in human resource management and supervision. The fourth issue is a perceived lack of role clarity for councillors in relation to responsibility for the health and safety of their CEO at the workplace. The fifth issue arises as Council discussions and processes regarding their employees are confidential, as this can obscure 'poor' human resource practices. The sixth issue relates to councillor training and competence in operational matters, such as finance and strategic planning. Although this latter issue does not directly contribute to CEO job insecurity, it may contribute to misunderstandings and tension between the CEO and councillors, thus indirectly contributing to increased perceptions of job insecurity.
3.2.1 Issue 1: CEOs legislatively required to report councillor behaviour suspected to breach legislation A key concern of participants was the legislative requirement for CEOs to report when councillors may have breached legislation and the perceived retaliation from councillors. The weight of this issue was evident from the quantity of codes and also as a point of stress felt by CEOs, given the professed potential to lead to termination/non-renewal of the CEOs' employment. Depending upon the severity of the suspected legislative breach, there are different external authorities that councillor behaviour can be reported to. The least severe breaches are reported to the Local Also supporting the findings of the current study is a 2017 review into the Queensland LG councillor complaint system. At that time, Queensland CEOs were required to be the first assessor of complaints made against councillors, passing only 'legitimate' complaints to an external authority. The report stated that CEO involvement in reports of councillor misconduct 'can lead to . . . severely prejudic[ing] their relationships with the mayor and councillors -sometimes leading to their dismissal' (Solomon, Playford, & Kellar, 2017, p. 29).
The actions of the LG CEOs in reporting councillor behaviour to external authorities meets the whistleblowing definition; 'the disclosure by organization members (former or current) of illegal, immoral or illegitimate practices under the control of their employers, to persons or organizations that may be able to effect action' (Near & Miceli, 1985, p. 4). Legislation across Australia protects public interest disclosures (whistleblowing) by employees in the public sector. However, Lewis, Devine, and Harpur (2014) report there had yet to be a 'successful' outcome under these statutes. The authors attribute this to a lack of coordination between civil remedies and employment law. As one of the present study's participants commented, 'you're vulnerable. You're exposed. And the system doesn't protect you' (Participant 13).
Due to their legislative requirement to report breaches by councillors, which is then exacerbated by the other five issues in this councillor-CEO conjuncture, CEOs expressed a high level of job insecurity; a 'perceived powerlessness to maintain desired continuity in a threatened job situation' (Greenhalgh & Rosenblatt, 1984, p. 438). Job insecurity is an acknowledged stressor and a predictor of decreased work engagement, effort, job satisfaction and performance, lowered life satisfaction, and increased physical health issues (De Cuyper et al., 2008;Reisel, Chia, Maloles, & Slocum, 2016;Roskies & Louis-Guerin, 1990). An extended period of job insecurity has a cumulative negative effect on job satisfaction and physical health (Heaney, Israel, & House, 1994). In other words, when CEOs feel that their employment may be in jeopardy, the productivity of the organisation and the health of the CEO are likely diminished.

3.2.2
Issue 2: Length of CEO employment contracts and severance payments Similar to the New Zealand experience (Asquith, 2008), this research found the LG CEO employment contract creates 'enduring tension' (p. 631) in relation to its fixed term. Legislation previously stipulated that the maximum severance payout when a CEO's employment contract was terminated was the remainder of their contract, but this has been reduced to a maximum of 12 months. This change appears to have intensified the level of job insecurity experienced by CEOs as they are keenly aware that their employment contract could be terminated as retaliation from elected members. The regional CEOs felt that the loss of their job, whether through early termination or nonrenewal of an expired contract, had a greater social and financial impact as it could additionally mean relocating their family.
Our home, [it is] our workplace. We lose our job, we may need a base to go back to. Our kids need schools. Partners need to find jobs. The councillors want CEOs with families to boost numbers in towns. They don't want drive in, drive out CEOs or senior staff. But . . . when you're only on contract and if they don't renew contracts. . . what's the impact on your family if that happens and everyone has to move? (Participant 25) The reduced contract length and the insecurity of tenure were also perceived to impact how a CEO performed in terms of giving honest advice.
[I]f you're expecting CEOs to act without fear or favour and to provide strong advice to their Council, there needs to be a level of protection as well. . . provide CEOs with a level of protection around tenure (Participant 9). However, the state government has proposed requiring all LGs to re-advertise CEO positions where an individual has been in the role for 10 consecutive years (DLGSC, 2019). This change may bolster accountability of the CEO role (Svara, 1990), although mandatory re-advertising of the CEO position has the potential to further exacerbate perceived CEO job insecurity in the context of the other five issues within this councillor-CEO conjunction.

3.2.3
Issue 3: Councillors may lack the skills, experience, knowledge, and time for human resource management Performance management and appraisal of CEOs is important because they play a critical role in creating a shared understanding between councillors and the CEO. Specifically, performance management and appraisal should demonstrate (i) how the CEO's activities link with strategic goals, (ii) how goals are to be achieved, (iii) where a CEO may need support such as task-based resources or professional development, and (iv) provide feedback on the CEO and administration's overall performance (Aguinis, 2014;Cascio, 2006). Human resource management, including performance management, thus impacts organisational performance (Ferguson & Reio, 2010). However, in general, the CEOs perceived that the performance feedback they received from Councils/councillors was of poor quality. They attributed this to councillors' lack of skill and experience in conducting performance reviews.
I was actually quite keen for good feedback to know how I could improve. But Council weren't, or didn't provide that . . . one of my reviews where I just walked in and they just said, 'Nup, good'. And that was the extent of the feedback (Participant 16). Hutchinson et al.'s (2016) interviews with WA Mayors/Shire Presidents revealed that they, too, held general concerns regarding the skills, knowledge, and experience of people who nominated themselves to be elected as a councillor, and affirmed that councillors needed to upskill to meet the job requirements. Australian councillors report being surprised and unprepared for the demands of the role, yet uptake of councillor training is low (Davy & Brands, 2012). However, as the only councillor roles eligible to receive equivalent to a full-time salary are the Mayor/Shire President and the Deputy Mayor/Deputy President, councillors are often engaged in other paid work, which makes it challenging to find the time to commit to training for upskilling (Davy & Brands, 2012;Public Accounts Committee, 2006).

3.2.4
Issue 4: Lack of role clarity for councillors regarding responsibility for the health and safety of the CEO at work The CEOs had an inconsistent understanding of who was responsible for their health and safety at work. This appeared to be a result of the CEOs being unaware of any formal processes or Councilled actions to support their health and safety. One participant also highlighted that if the CEO were to advise councillors about CEO workplace safety, it might be perceived as a conflict of interest.
Leaders of employees are critical to the achievement and maintenance of a 'safety climate' (Fan et al., 2020) where fear and intimidation perpetrated by uncivil/bullying individuals/groups cannot flourish, and physical hazards are mitigated (Dunlap, 2011;Mullen, 1997;Zacharatos, Barling, & Iverson, 2005). This is supported by research comparing employee-driven safety initiatives across multiple organisations that found the most important predictor of safety success was the quality of leadership in relation to safety (Krause & Weekley, 2005). This suggests that without intentional leadership support by Councils in ensuring CEOs can work in a safe environment, a 'bottom-up' approach to workplace safety by CEOs may be unsuccessful. The perception of interviewed CEOs was that those Mayors or councillors who had supported their health/well-being did so as a result of their personal relationship rather than because of a prescribed role. CEOs were also reticent about revealing a health/well-being issue to councillors as it tapped into their perceived job insecurity. Mayhew (2012) affirms that fear of potential job loss inhibits reporting workplace bullying and aggression.
In addition, several CEOs held reservations as to whether councillors had the requisite skills/experience to be effective in addressing workplace issues that CEOs perceived to impact their health and well-being. I do have one or two of the elected members, who are more just amicable, affable type people who, you know; 'Is everything going, are you alright'? . . . technically they're my employer. Do I really want to show that [my] heart's out here, my guts are over there? And I need help. They're my employers . . . And you can't put that responsibility onto the elected members because they're untrained, unskilled people (Participant 32).
Although CEOs recognised their role in looking after their own health and well-being, they also questioned the ability of Council to provide a safe workplace if councillors were harming the CEO's health/well-being -as reported by several participants. The literature also suggests that general conflict between councillors and the CEO may undermine attempts to address safety hazards. Tjosvold (1990) found that cooperative, rather than competitive, goals and the ability to engage in open and constructive discussions to resolve opposing views are features of a safe workplace.

3.2.5
Issue 5: Lack of oversight/transparency over council in their role as an employer Council discussions regarding the CEO are confidential and held behind closed doors. This is understandable as there are privacy concerns. However, CEOs perceived that procedural justice in performance reviews -in other words, performance assessment that is without bias (Erdogan, Kraimer, & Liden, 2001) -did not always occur and the lack of transparency could hide the failure to ensure procedural justice. To address the concern about procedural justice, some CEOs advocated for an independent review following termination of employment. However, there was also recognition that by then, a functioning and productive relationship between CEO and councillors was likely to be irretrievable.
[It would be good to] make sure that the . . . 'I'm going to get you' is taken out of the system. But ultimately, if they want to get you, then effectively they probably will. (Participant 15) The data also indicated that it was unlikely that engaging an independent human resources specialist would ensure procedural justice as councillors were believed to have overruled expert knowledge on other occasions.
[Councillors] fought with [the consultant] during the performance review. She was trying to hold them to task and make them objective. Good KPI's, good performance evaluations, and councillors railed. (Participant 29) 3.2.6 Issue 6: Areas of mandatory training (minimum levels of competence) for councillors From their understanding of the councillor role, the CEOs submitted that there were six specific training areas to support councillors. In order of most mentions to the least, these were as follows: governance, finance (specific to local government), urban planning, CEO performance review, teamwork, leadership, and how to become a conduit between the community and the LG administration. In addition, the CEOs of the present study indicated that strategic thinking was lacking in councillors. Overall, the findings from surveying Australian LG councillors (Davy & Brands, 2012) strongly overlap with the results of the current study suggesting training is needed in these areas, given a minimum level of competence is necessary to fulfil the councillor role.

CEO and councillor interface: Practical implications -Recommendations for change
Providing organisational support and protection reduces the anticipated personal costs of reporting wrongdoing by work colleagues (Cho & Song, 2015) as it creates an environment of psychological safety (Detert & Burris, 2007). However, Council support and protection of their CEO may have an inherent conflict of interest when a councillor is both the object of the CEO complaint and the 'protector' of the CEO. This suggests that if CEO reporting councillor wrongdoing is a priority to support LG transparency and integrity, CEOs should be removed from the direct employment of Council to mitigate potential retribution for 'speaking up'. Such a separation of powers is fundamental for responsible government (Althaus, Bridgman, & Davis, 2018).
One way to ensure the separation of powers is through revisions to CEO employment conditions (MLGRP, 2012). Existing recommendations include state government representation across CEO recruitment, performance management, and contract negotiation (MLGRP, 2012). Although a step in the right direction, results of the present study indicate that this level of involvement from state government may be insufficient to address the deep level of perceived employment insecurity. Such an approach is also unlikely to mitigate the perceived lack of councillors' skills, knowledge and time for human resource management, and perceived lack of responsibility for CEO workplace health and well-being.
Several participants raised the possibility of shifting responsibility for human resource management of LG CEOs from Councils to state government. Indeed, such an approach is already in use for the 46 CEOs of WA government departments and agencies, as the Public Sector Commission is the employing authority for these roles (PSC, 2018). One of the primary goals of this arrangement is to shield these employees from political pressure that could be asserted through human resource management powers of recruitment, appointment, remuneration, and termination (Public Sector Management Act 1994 (WA), Section 8) as well as to ensure they are 'provided with safe and healthy working conditions' (Section 8(1)(e)). Although the separation of human resource functions from the LG to state levels may ease some issues, such an arrangement raises concerns as to whether this will 'diminish the authority of the Council as the employer' (PSC, 2018, p.13) or undermine the independence of LGs (Grant & Drew, 2017). For example if the state government were to become responsible for human resource management of LG CEOs, there could be new issues that emerge if the Minister or Department has influence in these matters (Aucoin, 2008).
In relation to the mandatory training proposed by state government for councillors, this study, supported by Davy and Brands (2012), has identified a gap. Accordingly, it is recommended that training for community engagement and strategy development (which includes strategic thinking/decision-making) is added to the governance and finance topics that councillors need to develop and demonstrate competence.

3.3
The councillor-community interface

Lack of clarity (insufficient detail) in legislation about the councillor role
The interviewed CEOs conveyed their understanding that councillors are meant to operate at the strategic, rather than operational, level. However, they perceived that passing on requests for service from individuals in the community to the organisation was at the operational level and therefore was interfering in LG administration. It was evident that the CEOs did not have issue with councillors passing on information about broader, more strategic issues that affected the community that would be likely to trigger formal community engagement or consultation, such as creating a dog exercise park, but took issue with requests directly related to a privately owned property, such as noisy neighbour or verge tree issues.
We've had 250 [councillor requests] in the last couple of months. This is just from councillors saying do this or do that and do that. . . I think they shouldn't be doing it at all. (Participant 8) CEOs indicated that councillor requests for service on behalf of residents were an intensive use of organisational resources. Anecdotally, there are two reasons for this. The first is the level of executive scrutiny these potentially routine tasks receive. As the request is submitted by a councillor, it is lodged at the executive office and transferred down the hierarchy to the actioning officer. After the request is complete, any actions taken are written up and sent to the executive office who forwards this information to the councillor. In comparison, if a community member lodged the request in the first instance, the request is simply tasked directly to the actioning officer to complete according to the LGs policies and procedures.
The second issue is that residents unhappy with a decision of their LG have been known to subsequently approach their elected councillor to get the desired results. Hence, the LG may be dealing with issues twice. Scrutiny of decisions made by public organisations is not an issue; rather, it is the concern for democratic decision-making and transparency. If councillors put pressure on the administration to achieve a particular outcome, the LG may operate idiosyncratically and in a manner that is counter to its public policies.
This nuance to LG operation appears to be a large concern for LG CEOs; however, the legislation and scholars are silent on the issue of councillors lodging requests for service on behalf of individual customers/community groups. There are also conflicting views on this matter from evaluations of LG. The investigator into the operation of the City of Canning (Kendall, 2014) affirms that passing on individual requests for service relating to a property or household is a part of the councillor role. He also describes a potential solution while circumscribing councillor's involvement once the request has been passed on to the administration, as is discussed in the following recommendations for change. However, the Metropolitan Local Government Review Panel's view was that councillors need to act more like a board, shifting from a representative role to a leadership role, particularly acting for the 'advancement of the community as a whole' (MLGRP, 2012, p. 153) rather than narrow interests. In this scenario, Council leadership is complemented by representative democracy enacted through mechanisms of public deliberation such as participatory budgeting, citizen juries, and deliberative opinion polls (Blacksher, Diebel, Forest, Goold, & Abeslon, 2012;Jones & Stewart, 2011

3.3.2
Councillor-community interface: Practical implications -Recommendations for change To enhance role clarity of councillors within the LG, the legislation should make it explicit whether the councillor role includes lodging requests for service for a privately owned property/household on behalf of community members. If it is decided that this is a legitimate aspect of the councillor role, in order to prevent perceived administrative direction by councillors, the legislation could enshrine a practice that occurs in New South Wales LG and was also outlined in the City of Canning Inquiry Report (Kendall, 2014). Namely, if a councillor wishes to support a different result than the one the administration delivered, this can only be done by bringing the request to a Council meeting. Bringing the request for a particular action to a public forum facilitates a transparent decision-making process and potentially alerts the Council to where a policy may be inadequate, raising the issue from an administrative to a policy level.

Limitations and future research
The present study makes several notable contributions to our understanding of the concerns and proposed changes to the Act from the perspective of LG CEOs. That said, this study has several limitations and presents opportunities for future research. First, although nearly a third of WA LG CEOs participated in this study, participants were self-selecting and may not be fully representative. With this said, the sample size and representation obtained is similar to, and in excess of, other similarly designed studies in the LG context (Hutchinson et al., 2016;Smith-Ruig et al., 2016;Tan et al., 2016). Second, changes to the Act are likely to affect CEOs as well as councillors and community members. The present study only captures the views of CEOs, who are one of several significant participants in relation to the LG legislation. It is therefore a suggestion that future academic research should elicit the views of other stakeholders, such as LG councillors, to further identify unique and common viewpoints on these issues to comprehensively evaluate the potential impact of recommendations on the system. Third, given each Australian state legislates the framework for their LGs, the generalisability of these findings may be limited to states with similar legislation. However, research conducted in other regions such as New Zealand (Asquith, 2008) and Australia (Lohrey et al., 2019) have found similar tension points, and thus insights of the present study may generalise. Future research may seek to replicate the present study in other legislative contexts.

CONCLUSION
The present study sought the opinions of WA LG CEOs regarding potential revisions to the Local Government Act 1995 (WA). Findings relate to three focal interfaces: (1) state-local government, (2) councillors-CEO, and (3) councillors-community. The resulting recommendations include the establishment of a mechanism to review persistent complaints, widening the breadth of councillor training to include community engagement and strategy development, and investigating the feasibility and risks in shifting responsibility for human resource management of LG CEOs from Councils to state government. The recommendations outlined herein may assist in reconciling the chronic tension between politics and administration (Demir, 2014;MLGRP, 2012;Newnham & Winston, 1997;Tan et al., 2016), address major job-related stressors for CEOs. The recommendations also align with the public sector shift from government to governance with the accompanying greater focus on the outcomes to be achieved for the community (Grant & Drew, 2017;Stoker, 1998).

A C K N O W L E D G E M E N T S
The authors gratefully acknowledge that this work was partially funded by the Local Government Professionals Australia WA (LGPWA). Interpretations, conclusions, and recommendations are those of the authors and are not necessarily endorsed by LGPWA. The authors also extend their thanks to the CEOs who partook in this research as well as their research assistants Leah Zoszak, Lin Yong, and Andrew Connell.

E T H I C S A P P R O VA L S TAT E M E N T
The study was approved by the University Human Research and Ethics Office (RA/4/20/4807).