Public Sector Contracting: An Australian Study of Changing Work Conditions

Throughout the 1990s, government reports such as the Competitive Tendering and Contracting Report (Industry Commission1996) and the Report of the Independent Commission to Review Public Sector Finances (McCarrey 1993), suggested that the contracting-out of government services would improve the delivery of public goods in Australia. This article reports on the findings of a case study of contracting-out of building maintenance services in the state of Western Australia in Australia. The study focuses on evaluating the impact of the contracting-out of these services on the wages and conditions of employees affected by moving from public sector to private sector employment. The study found differences in the employment conditions of white collar and blue collar workers. In general, the conditions of white collar workers improved but there was a marginal decline in the conditions of blue collar workers. The article concludes that while contracting-out created budgetary savings for government, these resulted partly from a reduction in wages and benefits of former blue collar staff, despite the best efforts of the Western Australian government to protect staff wages and benefits. The research makes an important theoretical contribution by arguing that the impact of contracting-out on staff wages, benefits and conditions is conditional upon the level of competition in the labor market for worker’s expertise. In “thick markets”, wages, benefits and conditions are likely to reduce, due to strong competition for expertise. However in “thin markets”, wages, benefits and conditions are likely to increase, due to weak competition for expertise (Borland, 1994).


Introduction
Contracting-out of government services has fundamentally changed the nature of service delivery for government agencies (Verspaandonk 2001). Proponents have argued that the contracting-out of government services will result in considerable cost savings to government, due to the market efficiencies of a competitive tendering system (Domberger 1994). Concerns have been raised, however, with the effectiveness and efficiency of contracting-out government services, as well as concerns about competition and quality (Prager 1994). Quiggin (1999) has also raised concerns about the redistribution of risk from government to private sectors, and the source of any efficiency gained, arguing that these cost savings are most likely to ensue from reductions in staff wages and benefits. This article will extend the research into the contracting-out of government services by examining the effect of the policy of contracting-out of building maintenance services in Western Australia. Evaluation of both the intended and actual outcomes of the policy outcomes are a critical part of this process (Department of Finance 1994:8). When undertaking evaluating a government policy or regulation, good practice includes an assessment of the impacts of the policy, in terms of its direct and indirect costs on government, business and the community, as well as the economic, social and environmental impacts of the policy (Productivity Commission 2004;Regulatory Impact Statement Guidelines 1995).
This research project follows this approach when evaluating the effect of contractingout policy in Western Australia, by firstly establishing the intended outcomes of contracting-out by reviewing main elements of the academic literature and government reports that guided the implementation of the policy. A specific research question will be developed from the academic literature concerning contracting-out, together with a methodology for undertaking the research. This research question will then be examined based on the actual outcomes of the contracting-out process of maintenance workers in Western Australia. While efficiency gains will be noted, the focus will be on the impact of contracting-out policy on employee wages and benefits -the economic and social costs of the policy. The specific case of contracting-out in Western Australia will then be explored in order to determine the intended and actual industrial relations outcomes, for government employees of the contracting-out process. The article finds that staff wages and benefits were negatively affected following the contracting-out of maintenance services, however, the position of white collar workers improved through obtaining better jobs in transitioning to the contracted environment.
The conclusion drawn from the research is that white collar workers in "thin" labor market sectors are better able to position themselves for new types of jobs in the contracting-out regime.

An Overview of Contracting-Out in Australia
In examining the effects of contracting-out of government services in Australia, the Industry Commission (1996) predicted benefits of an expansion of contracting-out throughout Australia would be an increase in GDP of between 0.3% and 1.7%. The Industry Commission (1996) made broad recommendations in relation to the imple mentation of competitive tendering and contracting-out policies. These recommendations covered issues such as accountability, quality, industry development issues, the evaluation and tendering for contracts, local government provision of service and employment effects. Contracting-out is determined to occur when: a government or public agency ceases to provide a specific activity or service through directly employing its own staff and instead purchases that service or activity from a private corporation or non-government agency (Paddon/Thanki 1995).
The use of contracting-out and competitive tendering was advocated as a means by which government agencies can deliver more client-focused services while achieving savings and maintaining accountability (Australian Public Service Commission [APSC] 2003). Figure 1 depicts the increase in the value of services that have been contracted out in Australia since 1999. Notifications 1999Notifications -2004 Figure 2 depicts the number of services contracted-out since 1999. It can be seen from this data that the number of contracts has increased progressively over the past five years.

Graphics derived from Statistics on Commonwealth Contract
Government statistics demonstrate a steady increase in both the value and the number of government services contracted-out since 1999. Borland (1994) is concerned that this push to implement contracting-out may be as a result of it being the latest fad or fashion, rather than as a result of careful planning and research. Nevertheless, contractingout of government services has grown steadily in Australia, and remains a key policy of government.
The potential benefits from contracting-out are argued to include lower costs, improved service delivery and quality, and greater flexibility (King 1994). Competition between potential bidders for government contracts creates incentives to minimize costs, or to develop innovative products and services, which provide a commercial edge in the marketplace (Rimmer 1994:1). Steane and Walker (2000) note that contractingout has generally delivered cost savings to government, although results vary depending on what is contracted-out, and how the contracting-out process is managed.
The Industry Commission (1996) likewise found that savings could be achieved by government in the contracting-out of services, in most circumstances. Domberger (1994) considered the trend towards the contracting-out of government services as a key aspect of public sector reform. Based on case studies from a wide range of countries and services, Domberger et al. (1995a) calculated that the average savings obtainable from contracting-out would be around 20%. Domberger et al. (1995a) also suggested that efficiency gains could be obtained by removing control from protected in-house producers to encourage a competitive environment with both public and private sector providers bidding on public contracts. These results, however, have been criticized for potentially not considering all costs involved (Quiggin 1996(Quiggin , 2002, and for using measures and statistical tools unsuited for the task for which they were implemented (Paddon/Thanki 1995: 28-29). While the net fiscal result of contracting-out is generally argued as being positive for governments, the basis of those savings has been questioned by some researchers.
While in theory contracting-out appears to offer a Pareto improvement -a way to make some people better off without harming anybody else -in reality contractingout can leave some parties worse off (King 1994;Rimmer 1994). Quiggin (1994) argued that the cost savings deriving from contracting-out are due to a reduction in wages and benefits for workers, or an increase in work intensity. Lafferty and Roan (2000) also argue that the contracting-out of government services affects the skills and training of staff by reducing access to training, consequently resulting in a net loss of industry specific skills.
The benefits and problems associated with the contracting-out of government services extends into areas such as the cost, service quality, contract management, regional development and risk, together with a reduction in wages and benefits for workers. As noted above, this article will focus on the issue of salary and employment conditions of staff to specifically test Quiggin's (1994) contention that budgetary savings from contracting-out are likely to be the result of a reduction in staff wages and benefits, and an increase in work intensity. While Quiggin (1994) also contends that tax avoidance is a possible outcome of contracting-out policy, the researchers considered it unlikely that sufficient data would be forthcoming from interviews to be able to properly validate this aspect of Quiggin's theory. Academic literature would there fore suggest that the policy of contracting-out could be expected to have a negative economic impact for former employees, but would be positive economic outcome for government.

The research on contracting and employment conditions
Employee benefits and salary are contentious issues in contracted arrangements. Domberger (1994) contends that gains from contracting-out are mostly from improvements in efficiency and only minor cost-benefits are achieved through job losses, lower wages and reduced benefits. However, Quiggin (1996) predicts that reductions in wages and benefits would be a significant source of cost saving. This aspect is a key debate in the academic literature. This divergence of opinion by academics has sparked a lively interest in determining the actual source of savings in contractingout, and is a key driver for this research.
While widespread redundancy of public employees is sometimes used in conjunction with contracting-out, this is not always simply a case of net job losses. Domberger (1994) notes that public sector redundancies are matched, in part, by increases in private sector employment, especially in small to medium enterprises. The Industry Commission (1996) likewise found in a range of contracting-out cases that considerable numbers of public employees had successfully been reassigned or found work with contractors. Domberger et al. (1995a) maintain that public sector employees may not be as efficient as their private sector counterparts and suggest this could be a major driver behind contracting-out by the public sector. Quiggin (1996) criticizes the use of contractingout as a means to increase work intensity, as it is argued that the increased productivity gains from the increased effort of workers is a free good. Increasing the working hours or working intensity of workers is not a costless method of increasing output, as it is the workers themselves who are paying the cost of the increased efficiency.
In their study of electrical generation employees and union actions, Teicher et al. (2000) found that contracting-out is sometimes used to decrease union control, and can provide the mechanism for maneuvering around basic award benefits, working conditions and unfair dismissal laws. The issue of the relative efficiencies of a unionized workforce versus a non-unionized workforce has not been a strong focus in the literature to date, although Domberger et al. (1995a) argue that public service employees are less efficient than their private counterparts. Teicher et al. (2000) also notes a trend away from providing training to employees in contracts. Lafferty and Roan (2000) considered the effects of less training and the lack of skills in the workforce, finding that there was a loss of industry-specific skills in administrative staff and that training was rarely provided by either the contractor of the agency hiring the staff. Lafferty and Roan (2000) also suggest that women receive less training than men, based on the type of work commonly being outsourced. Similar to those issues mentioned by Lafferty and Roan (2000), training and gender effects are also explored in detail by the Industry Commission (1996) who found that there might be a range of equal employment opportunity [EEO] provisions that need to be addressed with the contracting-out of government services. The Industry Commission (1996) received submissions from many sources suggesting that women, people from non-English speaking backgrounds and older employees found it harder to find jobs in the private sector compared to more advantaged groups.
While contracting-out is argued to result in cost-efficiencies, some authors contend that these efficiencies are gained through the reduction in pay, benefits, and training for employees, or an increase in work intensity. Changing work practices that can occur as a result of the contracting-out of government services has not been strongly featured in the literature to date, apart from general notions of efficiency. More research is needed before strong conclusions can be made about the impact of contracting-out on costs in Australia (Rimmer 1994).

Research Question
Contracting-out has been widely implemented in Australia and is argued to achieve significant efficiencies due to the efficiencies wrought by competition (Domberger et al. 1995a). Some authors have claimed that the source of these efficiencies could be at least partly attributed to reduced wages and benefits of staff, and increased work intensity (Quiggin 1994). This article seeks to explore this difference by examining the impact of contracting-out in a specific case study: maintenance workers in Western Australia. A suitable research question would therefore be: What effect did the contracting-out of maintenance workers in Western Australia have upon employee wages, benefits, and conditions?

Research methods
This article examines the contracting-out of building maintenance services by the Western Australia Building Maintenance Authority [WABMA], in Australia. Case studies involve the systematic and in-depth analysis of information on a particular "case"individual, group, organization, or community and utilizes a variety of data gathering methods (Cavana, Delahaye & Sekaran 2000: 112). Case studies are useful for testing theories in a real life setting in order to confirm, challenge or expand a theory (Yin 2003). Yin (2003: 40) argues particularly that the "case can then be utilized to determine whether a theory's propositions are correct or whether some alternative set of explanations might be more relevant". Since the goal of this research is to specifically explore the theoretical assertion of Quiggin (1994) noted above, that contracting-out is likely to result in the reduction of staff wages and benefits, and increased work intensity, then a case study method would be an appropriate methodology.
The unit of analysis of this particular case study will be at the individual level, and seeks to examine the impact of contracting-out of building maintenance services by the Western Australian government on the wages of individual workers. This research forms part of a larger research project examining the contracting-out of government services in Queensland and Western Australia. The process and outcomes of the contractingout of the building maintenance services in Western Australia focuses particularly on the maintenance section of the Western Australian Building Management Authority, which became part of the Department of Contract and Management Services, and then part of Western Properties, which is a section of the Department of Housing and Works. The contracting-out of government services in Western Australia was en couraged by Domberger et al. (1995b), who acted as a consultant to the Western Australian government. Additionally, data concerning the actual efficiency gains and the impact on staff wages is available from either primary or secondary sources. The contractingout of Western Australian building maintenance services provides a very useful study for examining the contention between Domberger at al. (1995a) and Quiggin (1994) on efficiency gains and its sources.
In this article, individual informants and companies have been de-identified and any commercial in-confidence information has not been divulged. The names of government departments, government reports, and most government policies have not been obscured as most of this information has already been made freely available by the Western Australian government, either on the Internet or in public libraries. Individuals responsible for the management of the contracting-out of the maintenance services were initially identified and interviewed. "Snowball sampling" was used to identify and interview additional individuals who had participated in the contractingout process. People interviewed had participated in multiple levels of the process before and after the contracting-out of government services in Western Australia.
Primary qualitative data was gathered by way of semi-structured interviews, which were conducted with 15 former government employees who had been employed in those areas contracted-out to the private sector, public servants in charge of the contracting-out process, and managers of companies that gained contracts from government.
Secondary documentation includes both public and private company documents, including "Expression of Interest", "Request for Tender", tender submissions, contracts, as well as published and unpublished government policy documents and reports. Employee wage rates were obtained from the WA Industrial Relations Commission, and the WA Department of Employment and Consumer Protection.
Data was analyzed through three interrelated phases: coding of primary and secondary data in order to determine outcomes for staff; analysis of data and development of themes and theoretical development; and the writing up process itself (Funnell 1996). Analysis followed an iterative process between these three phases. Triangulation was used between primary and secondary sources to identify and validate information. Where primary sources cannot be validated by triangulation with secondary sources, these areas have been noted in the text.

The contracting-out of maintenance services in Western Australia
In 1993, a Commission established by the West Australian Government, the McCarrey Review, argued strongly for increased contracting of government services: Most public sector managers do not believe the outmoded public sector management systems and the lack of any real authority vested in them by past governments provide any encouragement or incentive to improve efficiency or effectiveness (McCarrey 1993a: xxxvii).
The objective of introducing widespread competitive tendering in Western Australia was argued to "improve the efficiency in government agencies, leading to the same level of service at lower costs or improved services at the same cost" (McCarrey 1993b: 29). In 1993, $AUS1.3 billion of Western Australian government services had already been contracted-out. However, the McCarrey Review suggested that much of this had occurred in an unstructured manner in the absence of a whole of government policy (1993b: 33). Building maintenance was one of the services already contracted out in Western Australia, although it was used primarily to address peak periods of demand for building maintenance services (Public Sector Management Office 1997).
The McCarrey Review (1993b) identified several issues relating to evaluating the performance of the Western Australian Building Management Authority (WAMBA), the agency at the time responsible for maintenance of the majority of government buildings (Interview 11 2005), including • The operations of WABMA were regarded as inefficient by government agencies and the private sector; • The day labor force was considered about 50% more expensive than private contractors; • Part of the workforce did not have sufficient work to keep them gainfully occupied; • If the construction and maintenance activities of the BMA were contracted out to the private sector, $20 million per annum was predicted to be saved.
The McCarrey Review (1993b: 34) argued that extending the contracting-out arrangements of building maintenance in Western Australia would produce an additional 23.9% of savings to government, and recommended to government that competitive tendering be used wherever possible. The potential savings from the competitive tendering and contracting of building maintenance services included the following: Importantly, it was recommended that all the design, construction and maintenance activities carried out by WABMA be contracted out to the private sector with WABMA taking on solely an inspection and maintenance scheduling role (McCarrey 1993b: xxxii). This recommendation amounted to the downsizing of the WABMA from 1,400 staff to 400 staff, and disbanding three sections of WABMA: construction, design and maintenance sections (Interview 11 2005). Contracting-out of government services was argued to produce outcomes of effectiveness and efficiency of service delivery (Domberger et al. 1995b: 15). Domberger et al. (1995b: 19) in his study of contractingout across multiple Western Australian governments, reported that there had been an average savings of 24.1% between 1993 and 1994.
While accepting that contracting-out tends to reduce budgetary costs Quiggin (1994), contends that the majority of budgetary gains from contracting-out of government services are as a result of the reduction in wages and increases in work intensity, and thus the cost savings are as a result of transfers from staff. The Industry Commission (1996: 556), in its major review of a number of cases of the contracting savings was from reductions in labor costs. This result was either from increased productivity on the part of the private sector, or from reductions in wages costs, with the relative importance varying from case to case.
Following the handing down of the McCarrey Review, the then Premier of Western Australia, Richard Court, wrote to each of his Ministers, urging them to extend the implementation of competitive tendering and contracting-out in their departments (Court 1993). A further letter (Court 1994) indicated that competitive tendering and contracting-out was a high priority for the government. The letter stated that departmental CEO performance appraisals and reporting to parliament were now directly tied to the implementation of competitive tendering in each department. Contractingout gained impetus in the Western Australian government from that time. WABMA responded to the calls for the expansion of contracting-out by Premier Court, by converting hundreds of small building maintenance contracts into a few facilities management contracts, which would manage the building contractors and day to day facilities management activities (Domberger et al.1995b: 23). This model of contracting-out went beyond the normal contracting-out of government services, as the facilities management arrangements involved the use of third parties who would undertake purchasing on behalf of government agencies (State Supply Commission 1996).
In 1995, the Western Australian Building Management Authority engaged consultants to develop both a process for managing the contracted maintenance services and to screen potential Facility Managers (those in charge of organizing contract maintenance) to make sure that they were able to meet the goals of the contractingout policies (Public Sector Management Office (WA) et al. 1997). In October 1995, a "Request for Proposal" for Metropolitan Buildings was called. This was the first phase in the contracting-out of government services and was intended to ensure that there was sufficient interest from potential contractors to ensure that market forces could be brought to bear on the actual tender process. In other words, the "Request for Proposal" was designed to test the marked and ensure that there was sufficient competition to result in budgetary savings.
In 1996, the Contract and Management Services (CAMS) was established. CAMS was a central contracting service for the public sector in Western Australia (Minson 1996). This entity took over the previous responsibilities of WABMA in the management of building maintenance contracts. CAMS accounted for 2% of total Western Australian government expenditure in 1998/1999 (State Supply Commission 2000).
CAMS was a significant shift for the Western Australian government in disbanding WABMA and moving staff to CAMS, ceased to provide building maintenance services directly. Maintenance services were purchased from private providers. This shift of the function of government department echoes the quote of Osborne and Gaebler (1987: 25): "The job of the government is to steer, not to row the boat. Delivering services is rowing, and the government is not very good at rowing". The McCarrey Review (1993a/1993b) determined that government was not very good at "rowing", or delivering services, and as a result CAMS was established to manage the service contracts.
The intended efficiency outcomes for this process were anticipated in the range of 20% (McCarrey Review 1993b: 34), and is in line with previous research findings and recommendations of Domberger (1994Domberger ( , 1995aDomberger ( , 1995b.

Industrial relations objectives
Between 1985 and 1990, approximately two-thirds of all building maintenance and breakdown repairs in the Perth metropolitan area were conducted by the "in house" labor force. By 1992, significant amounts of work had already been contracted out by WABMA. The following graph demonstrates that contracting-out resulted in a significant transfer of human capital and expertise from the public sector to the private sector (Interview 10 2005)2.

Figure 3: Comparison of Labour Force Composition*
*This chart is illustrative only, and is indicative of worker numbers. Actual numbers of workers could not be verified from secondary sources.
The Industry Commission (1996) indicated that a "clean break" approach was used by some government departments where the government agency made no attempt to transfer redundant workforce to the private sector. An alternative is the negotiated transfer approach, where the government agency negotiates the transfer of its staff as part of the tender process. This is assumed to minimize costs to government, potential service disruption due to the contracting-out process, and staff losses (Industry Commission 1996: 381).
It is evident from interviews with senior staff involved in the establishment of the new agency, CAMS, that there was considerable effort conferring with, and informing staff of the process involved in contracting-out maintenance services (Interview 10/ Interview 12 2005). The choice of whether to stay employed by government, accept redundancy, or move to the FM's was a voluntary decision by staff (Interview 10 2005). In effect the Western Australia followed the second approach suggested by the Industry Commission (1996) an approach that negotiated on behalf of its staff for their transfer to the facilities managers as part of the contracting-out process, although the choice of redundancy, redeployment or transfer laid with the individual employee (Interview 10 2005). Domberger (1994) argues that in cases similar to Western Australia, where guaranteed employment was given to existing staff for the duration of the contract, cost-savings could still be achieved.
The Industry Commission (1996) argued that requiring the contractor to maintain government wages and conditions should not be mandated, and should be subject to negotiation. The Western Australian government went against this recommendation and mandated the staff benefits as part of the contracting-out process, although these additional costs were built into the contract price. The Request for Proposal (RFP) considered the transfer of public sector employees and the protection of their accrued benefits (sick leave, holidays, pensions and leave) (Western Australian Building Maintenance Authority 1995). The transfer of public sector employees were a criterion in the evaluation process, mainly focused around the number of employees to be transferred and the protection of their existing benefits (Western Australian Building Maintenance Authority 1995).
Most facility managers employed both white collar and blue collar workers (Facility Manager 1 1995). Beyond the public sector employees, the proposal did not address the recruitment and selection process, and other human resource processes including training. On wage conditions, one facility manager indicated they would apply the "appropriate award" for each worker (Facility Manager 1 1995). While indicating a preparedness to cooperate with unions, there was some reluctance to participate in a union "closed-shop" (Facility Manager 1 1995).
Specific wage conditions for employees were not covered in the contracts between the government and the facility managers; however clauses concerning the transfer of some public sector employees were included (Facility Manager 2 1996). Industrial Relations were not significantly addressed in the RFP documents (Public Sector Management Office (WA) / State Supply Commission (WA) / Western Australian Municipal Association (1997); Western Australian Building Maintenance Authority 1995). It was evident that attention was given to ensuring existing sick leave benefits were preserved for staff. Moreover the employment of former government employees was guaranteed with the facility managers for the life of the initial contract (Interview 10 2005).

Efficiency Outcomes
In reports that cover the period being examined by this research, all report that costefficiencies were achieved by the Western Australian government following the contractingout of maintenance workers, although these were all at the lower levels of expectations (Public Sector Management Office 1997, Auditor General of Western Australia 2001). Initial savings from implementation of contracting-out were around 6.5% (McEachran/Ayres 1997), and the Auditor General of Western Australia (2001) found a savings of 6%. These are significantly less than the estimated savings from further contracting-out of building maintenance services which were estimated at $AUS 23.9 million, or 35% of total costs by the McCarrey Review (1993a). Significantly, these savings were attributed to reduced fees for service provision, and reduced expenditure on wages (Public Sector Management Office 1997). Having if any, on wages and conditions of employees.

Work condition outcomes
Following the establishment of CAMS and the contracting-out regime, there were several important outcomes for staff. First, after 1991, about a quarter of WABMA staff accepted redundancy packages as part of the early efficiency reforms. Within CAMS, over half the staff (54%) opted to take redundancy packages, 22% transferred to the new facility managers, and 2%, stayed in government. Many who took redundancy packages either established their own business and worked for the facility managers, or took the opportunity to move into new areas of employment. Most of the white collar staff transferred to the facility managers (Interview 14 2005). McEachran and Ayres (1997) found that 80% of staff felt that the transfer to the private sector was a positive experience.

Figure 4: Staff Transfers at Implementation of CAMS
This supports the contention of Domberger (1994) and the findings of the Industry Commission (1996) that numbers of public employees can be reassigned, or find work with contractors, following contracting-out of government services.

Staff experiences: Skilled Blue Collar Workers who transferred to Facilities Managers
Comparisons of the wages rates of Government and non-Government trade workers indicate that government workers received higher pay for the period studied. This pay variation is identified in the following table and graph. These diagrams indicate that skilled private sector workers in maintenance related trades (such as carpentry, plasterers, glaziers, drainers and general tradespersons) receive between 97% and 98% of the base salary of those covered by government awards.  Additionally, opportunities for advancement were reduced under the contracting-out period, as well as wages: "If I was doing the same job [in WABMA] … my salary would have been better and we would have had the opportunity to move to other positions … I was lagging behind in incremental wage increases as well" (Interview 4, 2003).

Award Pay Rates for Selected Trades 1996 -20043*
Direct wages were not the only area that was reduced. Indirect benefits such as retirement income through superannuation contributions were affected when transferred employees found they had a decreased ability to contribute to superannuation funds: "I would have been better off still being with government financially, from the point of view of superannuation and all those things" (Interview 3 2003).
The Western Australian government attempted to preserve the benefits of workers when they transferred from the public sector to the private sector (Interview 10 2005). The initial private sector arrangement paid employees higher rates than the government award (Western Australian Industrial Relations 1996). However, there was potential for facility managers to reduce the wages of those who were in their direct employment over time, particularly once the initial few years with the guaranteed protection of wages and conditions lapsed, with base wages in the private sector consistently lower than the public sector. A number of blue collar workers reported a reduction of total salary, superannuation, and training opportunities (Interview 3 / Interview 4 / Interview 6 2003).

Staff experiences: White Collar Workers who transferred to Facilities Managers
In contrast to blue collar workers, some of the white collar staff were financially rewarded, and experienced career progression as a result of the contracting-out process. Three went from junior positions in government into senior positions with the facilities managers (Interview 14 2005). These staff believed this improvement in conditions resulted from the contracting-out of maintenance services. Some blue collar workers were also appointed to administrative positions with the facility managers, before returning to traditional trades (Interview 12 2005).

Staff experiences: Unskilled Blue Collar Workers who transferred to Facilities Managers
Another section of the workforce unhappy with the transfer of employment to the private sector was trades assistants. These employees did not have trade qualifications (Interview 12 2005), however were paid wages close to that of skilled trades people by government due to their experience (Interview 10 2005). Following contracting-out, many were unable to work at the same level as prior to the contracting-out process (Interview 14 2005), as the private contractors or facilities managers were not required to pay above the award rates, or to recognize the experience that these workers had developed in the public service. These workers lost base wage rates as they were unskilled, and the market rates for their services were considerably lower than in government.

Increased work intensity
There was only an indirect indication of increased work intensity from the contractingout process. A senior staff member who remained in government indicated that he was unable to get his former staff to sit down for a cup of tea. "We always used to have time for a cup of tea and a chat, but now they all keep saying -we are just too busy for that now" (Interview 10 2005). Unfortunately there is no other primary or secondary data that can support an increase in work intensity beyond this single quote.

Discussion
Cost reductions and efficiency gains were well documented by the agencies that controlled contracting-out of maintenance services. The Public Sector Management Office (1997) analyzed the cost-efficiencies gained from the initial contracting-out process, reporting that for metropolitan building maintenance that had been contractedout, there was an average savings of around 6.5% for the first 6 months of activity on the contracts. The Auditor General of Western Australia (2001) found a savings of 6% over a 39 month period for metropolitan breakdown services. Areas such as building structure maintenance, grounds maintenance and other maintenance services showed significant decreases in cost, while there were only minor decreases in costs for filament replacement (Public Sector Management Office (WA) et al. 1997). While wage reductions were not considered as a source of potential savings in the submissions (Facilities Manager 1), the Public Sector Management Office (1997) equates the savings achieved to reduced fees charged by the facilities managers, and the reduced direct project-related costs charged for the maintenance workers themselves, or reduced wages costs. The contracting-out of government maintenance services in Western Australia resulted in budgetary savings to government, although these are at a lower level than predicted (Domberger 1994;Industry Commission 1996;Rimmer 1994). Government reports attribute part of these savings to reduced wage costs. Domberger, et al. (1988: 89) contends that when contracting-out results in wage reductions, this should be seen as socially desirable; however Quiggin (1994) argues that this is not necessarily a good thing. While government may well be achieving a reduction in budget expenditure, this is as a result of a transfer of wages and/or benefits from former employees. Employees are being required to pay for decisions of government, which means that the contracting-out policy had a negative economic impact on employees. While there is a net fiscal gain to government, there is a net wages and benefits loss to former employees.
There is no doubt that the contracting-out of maintenance services resulted in cost efficiencies of the Western Australian government (Interview 15 2005). However this study supports the contention that the contracting-out of government services can negatively impact employee wages and benefits (Quiggin 1996) and, training and advancement opportunities (Laferty/Roan 2000). The wages and benefits of former government employees, transferring to the private sector, had several work conditions outcomes.
One interviewee commented on maintaining at least the Government award standard pay and benefits. However, this employee conceded that some of those benefits were combined into a few, encompassing benefits. For example, travel allow ances were cut significantly (Interview 2 2003). A second interviewee rejected a position with a facility manager after being offered $2000 less than his/her public service salary package (Interview 6 2003). These poorer conditions were evident despite the government seeking to maximize the benefits of staff transferring to the facility managers.
Wage case data collected from the Western Australian Industrial Commission for blue collar workers indicates that facility managers were able to use lower award wages then what would have been used in a government controlled agency (Building Trades (Government) Award 1968, No. 31A 2004Building Trades Award 1968, No. 31 2004Western Australian Industrial Relations Commission 1996;Western Australian Industrial Relations Commission 1998;Western Australian Industrial Relations Commission 2001;Western Australian Industrial Relations Commission 2003). This result supports the contention of Quiggin (1994) that these savings could be at least partly attributed to a reduction in wages of employees. There is also some evidence to suggest that unions were being phased out of the enterprise bargaining process, but it is unclear if the facility managers controlled this process (Interview 2 2003). Existing staff, transferred at the commencement of the new government facility management agency, CAMS, were guaranteed existing wages and conditions. However, once these original set of contracts expired, there was no longer any guarantee that previous conditions would be maintained (Interview 14 2005). Facility managers would eventually have the opportunity to reduce wage expenses over time.
One of the major outcomes that the government was aiming to achieve was the protection of benefits gained by public employees in the private sector (Western Australian Building Maintenance Authority 1995). Where staff were transferred to facility managers, none reported a loss of benefits due to the transfer to the private sector (Interviews 2/3/4/5/7/9 2003). However, one interviewee reported that benefits would be lost if they were made redundant (Interview 4 2003) According to interviewees, the transition period was mixed, with some interviewees suggesting the transition was fair and amicable (Interviews 1/2/3/7 2003; Interviews 10/13/15 2005), and others indicating the system was unfair, difficult or inflexible (Interviews 5/6/8/9 2003). Blue collar workers, unskilled workers and poor performing workers were more likely to report a negative experience than white collar workers. Employee benefits were preserved for the duration of the contract. Senior public service managers maintained contact with their former staff, now working in the private sector, to ensure employment conditions were being maintained.
Training arrangements were also considered less then adequate by interviewees, with at least one reporting not having access to meaningful training in the 8 years of employment with a facility manager (Interview 4 2003). This result supports the findings of Lafferty and Roan (2000) who suggest that contracted-out workers receive less training than their in-house equivalents. There is only very limited support from the data (one interview) that increased work intensity was a result of the contracting-out process, and this is only by inference.
This study has found that in the case of the contracting-out of maintenance workers in Western Australia there has overall been a negative impact on employee wages, training and advancement opportunities, and superannuation. Employee leave entitlements appear to have been maintained. This research therefore supports the assertion of Quiggin (1994Quiggin ( , 1996 that a significant source of savings for contracting-out would be staff wages and benefits. This result was not universal, and the relative differences between white collar and blue collar staff needs further examination. Borland (1994) advanced the hypothesis that where there are a large number of people with competence in the delivery of certain services in a market (e.g. carpentry) then there is a "thick market" and significant competition for the delivery of services could be expected. The competition in the market for unskilled trade's assistants would be presumably even stronger than for skilled trade's people, which would help to explain the pay reductions for these workers.

Differences between white collar and blue collar
However, if there are relatively few people able to provide a specific service or skill, then there is a "thin market" and relatively little competition could be expected (Borland 1994). Under these conditions, there would be little cost savings to government, as employees could charge a premium for their services. The Industry Commission (1996) supported this notion, arguing that transfer of staff from the public to private sectors could be very beneficial for the employees, if there was specific expertise lacking in the general marketplace, that government employees happened to possess.
While there is a thick market of carpenters and other trades people in Western Australian, the government also contracted-out public servants who had the capability of managing state-wide emergency breakdown maintenance services and contracting services across large government departments. These people where relatively better off following contracting-out, as they had expertise that was lacking amongst facilities managers (Interview 10 2005). The Industry Commission (1996) suggested that in such situations workers would be able to charge a premium for their expertise as it was lacking in the market.
This finding from Western Australia suggests a modification to the theory of Quiggin (1994) that contracting-out automatically results in reduced wages and benefits, as for some staff in Western Australia the opposite was the case, with increased wages and opportunities. Consequently, the following hypothesis is suggested: Cost savings achieved by government from the contracting-out of services, can result in reduction of wages, benefits and conditions for workers when there is an abundant supply of these workers (i.e. competition is high) in the labor market. However, workers faced with little or no competition in the labor market due to the possession of specialized skills, will fare considerably better from the contracting-out process, as they are able to extract a premium for their skills.

Conclusions
While the findings indicate that there was some effort expended in preserving wages and conditions of workers who transferred from the public sector to the private sector, the translation to the private sector came with some cost for workers. The overall package of wages and conditions has been shown to be less than that received in the public sector for blue collar workers. Moreover, once installed in the private sector, these employees were more vulnerable to the exigencies of private sector employment conditions such as poorer redundancy arrangements. Longer term employment considerations, such as adequacy of retirement income, were also highlighted as problematic for workers who transferred to the private sector in their access to more limited superannuation entitlements. The ability of private sector trades and blue collar workers to secure adequate career trajectories and career development opportunities including training were also more limited in the private sector. However, the immediate promotion and advancement opportunities for white collar workers who were able to demonstrate special skills not readily available in the market, were more favorable in the private sector contracted out arrangements.