University of New South Wales Law Journal
SEAN COONEY,[*] JOHN HOWE[**] AND JILL MURRAY[***]
This article is an assessment of the changes to labour standard-setting in Australia brought about by the Workplace Relations Amendment (Work Choices) Act 2005 (hereafter WorkChoices). We examine whether WorkChoices is an example of sound regulatory practice and whether it is likely to address the key social problems associated with work in Australia in the early 21st century. We conclude that it is not, and indeed is largely directed at increasing the discretionary power of employers.
Part II of this article establishes a framework for analysing WorkChoices. We raise normative arguments in favour of maximising not mere work but decent work and contend that the private law of employment is, on its own, unable to deliver decent work. Public regulation of the employment relationship is also required. We then consider which modes of public regulation are most appropriate in the employment context, drawing on the now extensive literature on effective (and in particular responsive) regulation.
Part III applies this framework to WorkChoices, focusing on the regulation of time and pay. The Part begins with an explanation of the ways in which WorkChoices departs from the previous system of setting labour standards for decent work. We then find that WorkChoices favours ‘command and control’ rather than responsive regulation. We further observe that although this form of regulation purports to ‘guarantee’ basic labour standards, it often exacerbates the problems of private law by expanding the scope of employer action.
The article concludes by suggesting that a better approach to workplace reforms, while not returning to the past, would draw on, rather than marginalise, the successful elements of the previous system.
Work is a central activity of our society. It is regulated by law for a variety of purposes and through a variety of legal mechanisms. Working relationships take a very diverse range of forms and generate a large number of ethical, economic and social issues. The law of work needs therefore to be highly sophisticated if, faced with complex social interactions, it is to avoid the ‘trilemma’ of ineffectiveness, non-responsiveness and incoherence.
While the law of contracts and other private law obligations are in theory fundamental to the contemporary legal regulation of work, public regulation of working relationships through state agencies and institutions has played at the very least an equally important role. Both private and public modes of regulation have created norms specific to employment relationships, as opposed to other forms of work relationships (such as independent contractors). The boundary between employment and categories of working relationships is unstable in many contexts and there have been many legislative attempts at transcending it. Nonetheless, the legal category of the employment relationship remains highly relevant; it applies to eight and a half million Australian workers (85 per cent of the paid force ) in Australia. The employment relationship is also a central concern of the government’s WorkChoices reforms and, consequently, we take it as the focus of analysis.
Public regulation of employment is pursued in order to achieve certain socio-economic policy goals. One of the most prominent of these goals in recent years has been the generation of employment. However, there is significant disagreement as to how this goal should relate to other policy objectives.
Standing’s analysis of the different concepts of work, labour and employment provides an insight into the reason for this disagreement. In Standing’s view, work is a fulfilling and an essential human activity which should be supported in all of its guises – paid employment, volunteer community involvement, domestic care work, artistic endeavour and so on. Labour, on the other hand, is ‘toil’ work that is performed out of sheer necessity, often in conditions of alienation. Labour law traditionally conceptualises its topic as ‘labour’ in this sense, hence, for example, its focus on regulation to secure relief from toil in the form of protected ‘leisure’ time. The public policy commitment to generating ‘employment’ frequently obscures this distinction, as employment is usually defined as the antithesis of unemployment, itself a politically determined category with its roots in the need to manage the welfare state and its commitment to support those without a paid job through the public purse.
Standing argues that one may wish to see everyone in society engaged in meaningful, rewarding work without necessarily agreeing that maximising employment (regardless of the nature of that employment) is the primary social goal. Work is not just a necessary part of economic life, but integral to human well-being and healthy community development. Decent work is a necessary foundation for a healthy society; failure to institute decent work in the past has the potential to lead to great social disruption and exclusion of citizens, and is, inter alia, contrary to the economic interests of the country.
Adopting Standing’s analysis, we examine workplace law from the point of ‘decent work’ and with an interest in the quality of the jobs created. In this approach, the aim of creating jobs, even if they have extremely low rates of pay or are dangerous or dehumanising, is not a valid goal of social policy, and certainly not one that trumps the social need to create decent conditions through legal regulation. In other words, decent paid employment, but not employment per se, should be maximised.
The promotion of decent forms of work will call for a multiplicity of legal approaches attuned to the particular needs of the mode of work. There is no inherent conflict between such a goal and the attainment of an efficient economy. As Deakin and Wilkinson argue, failure to institute measures which protect and promote the capabilities of workers is a barrier to full labour market participation and the efficient operation of this market.
What precisely is meant by ‘decent work’ and why is public regulation necessary to achieve it? Decent work is a term widely used by the International Labour Organisation (‘ILO’) to ground its activities. According to the ILO:
decent work sums up the aspirations of people in their working lives. It involves opportunities for work that is productive and delivers a fair income, security in the workplace and social protection for families, better prospects for personal development and social integration, freedom for people to express their concerns, organise and participate in the decisions that affect their lives and equality of opportunity and treatment for all women and men.
As Hugh Collins points out, law has a role in enabling workers to engage in decent work, ensuring that they are not subject to ‘acts and omissions which fail to treat [them] with concern and respect’ and that ‘each person’s attempt to bring meaning to his or her life through work’ is furthered.
Decent work entails the construction of standards and the means to enforce them in areas including the following: the provision of adequate knowledge about the legal status of the worker and the employer’s legal obligations to them; the maintenance of a workplace free from harassment and risk of physical or mental injury and compensation where harm results from failure to do so; adequate remuneration for work undertaken which cannot be subject to arbitrary diminution; facilitating the worker’s capacity to adjust working time and work content to meet changing life circumstances; protection against arbitrary termination; freedom from unlawful discrimination; protection from breaches of civil liberties such as freedom of speech, the right to privacy and so on; freedom of association; and prohibitions on the exploitation of children and other vulnerable groups.
Certain standards should not be derogable (an obvious example is the prohibition on the exploitation of children). That is, persons should be entitled to insist on the standards notwithstanding any other arrangements to the contrary. However, it is also important that the catalogue of legally protected entitlements does not ossify and lose touch with the interests of workers, their communities, firms and the economy generally. New social problems may call for new coalitions of actors and discrimination may mutate to encompass new forms of wrong and so on. What is needed, then, is some kind of open process for the careful monitoring and revision (where necessary) of the corpus of rules that is to apply to all to ensure that labour standards are dynamic and evolving.
The achievement of decent work requires not only the capacity to engage in private ordering, facilitated legally by the law of obligations, but also standard-setting though appropriate modes of public regulation. In many settings, private law, and the law of contract in particular, is at least in theory a useful and adaptable device enabling workers and firms to mould their legal relationship in mutually beneficial ways. However, private law, unqualified by public regulation, has several serious shortcomings. These are most evident if we consider the law of contract of employment. The employment contract is not simply a regulatory device to give legal effect to the mutual commitments of individual and firm. The courts have constructed contractual default rules which, at least until recently, entailed the subordination of the employee to the employer. These rules include the employee’s duty to obey all lawful and reasonable commands of the employer and the employee’s duty to render loyal and faithful service. In the absence of express terms to the contrary, the default rules enable the employer to exercise what Collins calls ‘bureaucratic’ power, that is, the on-going control of job content, labour process, temporal arrangements and career opportunities (if any) within the firm. Thus contract by itself does not ensure that work is decent, and indeed supporting such employer discretionary power can facilitate degrading treatment.
This can be seen by developments in the common law, especially in overseas jurisdictions, which seem to indicate a shift from the default rules towards requiring greater obligations from the employer to the employee. The implied term of mutual trust and confidence is one such illustration. Furthermore, some judges have held that an employer’s contractual powers, even where expressly conferred by an employment contract are circumscribed by implied obligations of reasonableness. However, the scope of these doctrines have not been settled, especially in Australia where even the underlying doctrines themselves have not yet been accepted definitely by an appellate court and attempts to link the duty of trust and confidence to facilitation of ‘family friendly’ work arrangements have failed. It may be objected that whatever the implied terms may be, parties are free at the time of original contracting, or later through contract renegotiation, to determine express contractual rules that give effect to their own preferences, including the imposition of limitations on employer discretions which jeopardise the provision of decent working conditions, or the conferral of rights on employees to alter their working hours to accommodate their family obligations. However, many Australian judges have been reluctant to give contractual effect to arrangements that favour employees. This is particularly evident in attempts to enforce, through contract, workplace arrangements to accommodate family responsibilities made at the request of women workers. Judges have rejected attempts to establish a qualification on or variation of an employer’s express or implied entitlement to require adherence to originally agreed working hours.
In any case, as Joellen Riley has pointed out, the idea that employment contracts are the subject of extensive bilateral negotiation is often illusory, particularly in cases not involving senior professionals. While bargaining power obviously reflects conditions in the labour market, employers in a negotiation are frequently better placed because they have greater information, institutional support and wealth. This is especially true of large firms which usually have a specialist human resources unit, responsible for setting employment conditions. One of the purposes of these units may be to attract and retrain staff which make a positive contribution to the firm, but this purpose is subject to the overriding goal of prioritising firm, not employee, interests. The worker approaching such a firm for the first time as a job applicant may know very little about negotiating a contract which translates her or his key concerns into legally binding obligations, and thus finds it difficult to analyse or alter the terms proposed by the firm. Further, an employee will frequently be unaware of future developments in their life course (such as the birth of a chronically sick child or the incapacity of an elderly relative) at the time of initial contracting.
Even where workers have some labour market power (usually characterised as the capacity to easily find another equivalent job) they may still be harmed by asymmetry of information. For example, workers can have little a priori insight into the intensity of work obligations within the firm, a matter usually solely within the knowledge of the employer. Although they succeed in negotiating temporal restrictions on their work or arrangements to allow them to attend to their children or elderly relative, these may be overridden in practice by the requirement to complete one’s workload satisfactorily. And of course, workers with little market power may have to accept a contract on a ‘take-it-or-leave it’ basis.
Furthermore, the rules of contract are ill-suited to address the consequences of employment contracts for persons other than the immediate parties. Contract does not readily enable employees to coordinate claims for decent conditions. On the contrary, it enables, in those industries where employees have relatively low bargaining power, the conclusion of deals involving poor conditions which create downward pressure on future deals in analogous circumstances. In the so-called ‘race to the bottom’, employers are forced to compete by cutting labour costs. Workers are forced to compete – one against the other – on the basis of pay and basic conditions, including working time. The existence of such extremely poor quality jobs in the mid to late 19th century was predicated in part on contractual rules unmitigated by compensating public regulation. The return to such a position is undesirable for many reasons: workers’ living standards would be unacceptably low, employers would be discouraged from developing smarter ways of working, some potential workers are likely to be excluded from the labour market because of extreme poverty and disadvantage and overall national productivity would suffer.
Conversely, contractual rules do not always facilitate the diffusion of good employment practices – ‘a race to the top’. Although it is possible for innovative practices (such as processes for balancing work and family commitments) to spread rapidly throughout certain labour markets by means of private ordering, the absence of public institutions requiring transparency and facilitating public evaluation retards the transmission process.
Finally, contractual remedies are frequently expensive, time consuming and, given the limited availability of specific performance and injunctions in the employment context, a clumsy and ineffectual way of giving substance to non-pecuniary obligations. Workers other than those on very high incomes are thus most unlikely to pursue contractual remedies through the court system; indeed while the employment contract is fundamental to the judicial frame of reference when it considers work relations, it is doubtful that many ordinary workers are even aware of contractual terms to which they are subject.
Given these shortcomings of private law, the intervention of public law to achieve decent work is necessary. The need for public regulation in many areas is not the subject of major public controversy (although the precise nature of the relevant norms may well be). Thus, legislative enactments at the state and/or federal level establish regimes of anti-discrimination, occupational health and safety, compensation for workplace injury, freedom of association and the prohibition on certain forms of child labour.
Far more controversial, though, is the appropriate form of public regulation for dealing with issues such as work time and remuneration. Consider the difficulties confronting law as it seeks to establish the temporal boundaries of work. On the one hand, law is called on to give legal effect to individual preferences about working time and to enable mutually satisfactory agreements to be made between an individual and a business where preferences as to time coincide. Contract is sometimes able to perform this function. On the other hand, there are numerous instances where preferences do not coincide, change or have negative consequences for the individuals concerned or community life. These include requirements that a person undertake working hours injurious to health, work on days of public commemoration or national holidays, such as Anzac Day or Australia Day, or at times ordinarily spent with family and friends (such as Sundays), asleep or caring for others. In these instances, contract is quite inadequate to provide an appropriate policy response to McCann’s insight that working time (time spent in paid employment) should be conceptualised as ‘time out of life’ and treated accordingly. But how should public regulation deal with this issue?
Consider also the question of remuneration for work performed. Again, contract provides a vehicle for some individuals and firms to come to mutually satisfactory and legally binding arrangements. However, complex social and economic questions arise about the desirability of preventing poverty in our society. Questions involving consideration of minimum remuneration thresholds, the link between these and systems of social security and taxation, and the desire to avoid low-skilled poor quality jobs in an advanced economy. Contract is not well equipped to deal with these questions. Likewise, issues of the regularity of payment, permissible deductions and treatment of payment during leave are matters that are frequently not adequately dealt with by contract. But the stipulation of minimum wages through public regulation is highly contested.
Our view in the face of this controversy is that public regulation has a role in establishing dynamic labour standards. There are two important aspects of such standards. The first is to provide a floor to competition in the labour market, including in relation to wages and working time. We acknowledge that any ‘floor’ may involve complexity and the need for responsive standard-setting, as well as the capacity for change over time. We note that there is no reason to think that the parties to employment relationships have a monopoly on the development of novel working standards. Expert opinion, international developments and the views of customers, clients and others should be sought through rigorous, open processes.
Secondly, law should assist in the development and evolution of appropriate conditions in a wide range of areas above the floor. Issues such as job design, classification structures, pay rates, penalty rates and hours of work are complex regulatory questions. Nuanced approaches are called for which permit adaptation to very local circumstances at the workplace level, and even at the level of the individual worker.
There is a need then for public regulation to complement, and sometimes displace, private ordering. There are very many modes of public regulation. Which are appropriate in the employment context? As our starting point, we use two key insights of recent work in the field of regulation theory. First, we adopt the multi-faceted concept of ‘regulatory space’ as developed by Scott and others. Rather than focus on the hierarchical, vertical operation of law and state-made regulation, Scott argues that in fact law tends to operate in complex, decentred ways involving relationships which are both horizontal and vertical, public and private, hidden and visible. This analysis is particularly helpful for understanding labour legislation, the unfolding of which depends on the interaction between many different actors and social and economic forces, as well as on the interaction between private and public modes of regulation.
Secondly, we draw on regulation theory which has highlighted the deficiencies in a ‘command and control’ approach to legal regulation. Command and control regulation consists of governmental standards or rules (commands), backed by coercive penalties or sanctions, which require specified behaviour of persons external to government in order to prevent social harm. This state-centred, unilateral and legalistic regulatory mode is appropriate in some circumstances but is prey to unintended consequences and overall failure to achieve its regulatory goals. One response to this failure (associated with neo-liberalism) is often to resort to private ordering, which we have just argued is deficient in important respects, or voluntary self-regulation, which regulation theorists have shown lacks the capacity to deliver the public goals of intervention.
In contrast, we consider that regulation needs to be responsive. Complex and changing circumstances tell against exclusive reliance on simple command and control and/or self-regulation modes. Instead, there needs to be a way in which to engage the actors and to respond to the processes within the regulatory space. This approach recognises the fact that standards need to be set and monitored in ways which respond to local conditions and evolving circumstances. Regulation needs to involve those who are impacted by the issue being regulated, broadly defined. Thus, to the armoury of government rule-making should be added other relevant modes of engaging and shaping the behaviour of actors (such as permitting industry agreements, financial incentives and so on). Responsive regulation can thus include arrangements such as co-regulation (or tripartism – in which governments and social actors devise standards and implementation methods together) and enforced self-regulation (in which social actors devise their own standards and implementation methods subject to public oversight and, frequently, mandatory criteria). Of course, once it is acknowledged that the parliament is not the sole source of rules, important questions about transparency and accountability are raised. The nature of the social contestation over regulatory goals, standards and enforcement must also be assessed to ensure that voices are heard and socially useful outcomes pursued.
A good system of regulation is therefore one which emphasises the dynamic nature of standards and the need for participative, transparent and democratically sound processes for their identification, implementation, monitoring and enforcement. It is also one which leads to the intern ationalisation of standards in the operations of those subject to them.
These lessons from regulatory theory apply in the employment context. A ‘command and control’ to the problem of work time, for example, may lead to the stipulation of standards which rapidly become more complicated as they attempt to respond to problems of implementation. For instance, while it is a matter of public importance to designate one or more days a week as not ordinarily requiring work, a ‘brightline’ rule that prohibited work on Sundays would need to be qualified to permit the operation of hospitals, the police force and public transport and many other industries. It would be potentially discriminatory for employees of non-Christian faiths. If instead of prohibition, a disincentive or compensation scheme was established, it will be difficult to adapt the content to specific industries and workplaces.
Regulatory theory suggests that public regulation of such issues should be constructed quite differently from ‘command and control’ in order to address such problems. It is in relation to the dynamic aspect of labour standards that the lessons of regulation theory are most relevant. It has long been recognised that the broad concept that workers are entitled to decent work must in fact be elaborated in every historical period according to the wishes of workers, employers and society as a whole. As examined above, the field is not well-suited to centralised, legalistic standard setting by the legislature. Rather broad principles – social justice, fairness, the public interest – or regulatory frameworks, within which nuanced rules can be elaborated by non-state parties, need to be established. Both the floor of standards itself and the other dynamic labour standards ‘above’ it must be established and maintained in ways which are consistent with fundamental principles: creativity and innovation in production must be encouraged, the interconnectedness of work and life must be acknowledged, the non-derogable standards must be maintained and so on.
In other words, the institutional arrangements that facilitate the production, modification and enforcement of standards at work need to be responsive. They need to enable the translation of broad public goals into dynamic locally applicable norms. They need to ensure that those goals and norms are developed with input from those affected by them, particularly those immediately affected. They need to be transparent, so that they may be publicly monitored, evaluated and compared. They need to be accountable through democratic processes. They also need to create compliance mechanisms that operate within the regulated firms, not merely externally.
In order for responsive regulation to operate effectively, there must exist channels for persons affected by those standards to formulate in an informed and considered way their views about the content and implementation of those standards, and to communicate those views. Since it is obviously difficult for individual workers, even at the level of the firm, to do so, responsive regulation requires the facilitation of mechanisms that enable workers to coordinate and deliberate. The best known such mechanism is the trade union. We do not suggest that trade unions in their traditional forms are the only appropriate legal structure for worker co-ordination and deliberation. We do, however, maintain that the law needs to enable workers to organise deliberative forums that facilitate participation in processes of standard-formation and implementation, whether this be at a national, firm or other decentralised level.
Is WorkChoices responsive regulation or something else? The metaphor of regulatory space allows us to eschew a focus on government or discrete regulatory institutions or agencies as the sole source of norms or standards within a regulatory regime. Instead, we acknowledge that authority and responsibility for a key element of regulation – standard-setting, or the ‘process or set of processes by which norms are established’ – is and indeed ought to be dispersed between a number of public and private organisations. This, in turn, assists us in assessing the extent to which standard-setting under the WorkChoices legislation is responsive in the sense that it engages with the regulatory actors and processes within the regulatory space of work, and avoids either a centered, ‘command and control’ model of regulation and/or an abnegation of public regulation, leaving private law the dominant regulatory mode.
Our strong view is that, in its approach to labour standards (and indeed to many other matters), WorkChoices is not at all responsive regulation. On the contrary, WorkChoices institutes new forms of ‘command and control’ regulation and this is used to reinforce those aspects of private ordering that favour the employer. The reasons for the paradoxically ‘state-centred, unilateral and legalistic’ approach can be found in the nature of the system the government inherited. While the federal government espouses the benefits of the so-called unregulated labour market, one which is primarily governed by relations between individual workers and their respective employers (with ‘third parties’ excluded), it has been faced with the complex and entrenched edifice of the previous system, which we briefly describe below. Embedded in this edifice are many individual and collective workers’ rights and entitlements. The regulatory space created by this system fostered the strength and legitimacy of the Australian trade union movement, to which the present federal government is no friend, and shaped employer responses to labour questions in ways that were relatively settled but constituted constraints on ‘managerial prerogative’.
Over the course of the 20th century, many Australian workers had their pay and conditions of employment set, not through legislation, but through the terms of a federal (or state) award. An award was a legally binding instrument created in the settlement of an industrial dispute between a registered trade union and one or more employers. Gradually, these instruments came to include voluminous terms and conditions of employment. Awards are remarkable in that they not only set standards but also create processes, which deal with workplace elaboration of these standards and with other matters not covered by the award.
The central institution of the previous system, the Australian Industrial Relations Commission (‘AIRC’), and its forerunners were a hybrid. It was not a ‘super-legislator’ which could act on its own motion to create national labour standards; it was activated by the finding of the existence of an industrial dispute, and was limited to solving issues within the bounds of that dispute. Yet despite this systemic limitation, the tribunal evolved to become a powerful regulator, making far-reaching, ‘system-wide’ decisions that have directly or indirectly shaped the working conditions of most Australians. The test case function, where an AIRC Full Bench determines a national standard (for example, in relation to maternity leave or redundancy payments), is an illustration. The federal tribunal’s procedures are rigorous yet flexible: it is not bound by legal form and technicalities, yet it operates as a quasi-judicial body which tests the claims of all parties before it.
Awards were not simply the creation of the tribunal, as there were various processes by which an award could be made at the federal level. The award might emerge from the various processes of unmediated agreement between the union and employers (the consent award), conciliation and, ultimately, compulsory arbitration. Australian awards, even though they were made at the federal level, could be made with an unusually high degree of attention to the particular needs of the various industries and occupations they covered, even to individual workplaces. It was not uncommon for arbitration proceedings to involve worksite inspections or local hearings, to enable the workers and employers affected by the case to be heard, and for the tribunal member to get a better knowledge of the industry they were regulating.
The resulting rules which eventually emerged, with legally binding effect, are often characterised as centralised, inflexible and monolithic. While the number of awards and award content tended to increase over time, this analysis fails to recognise the flexibility inherent in many of the tribunal-generated standards and processes. Examples are the creation of dispute resolution forums in individual workplaces, the technique of enabling employers to derogate from standards if they could mount a successful ‘incapacity to pay’ argument. Further, the tribunal often determined standards in a cautious way, permitting feedback from the field to further inform refinement of the rules as necessary. Also, the tribunal frequently interspersed its formal arbitration of matters, even national test cases, with conciliation and informal negotiation between the parties. An example of this sophisticated and multi-faceted process is the decision which emerged from the 2005 Family Provisions Test Case, discussed below.
Within the conciliation and arbitration system, there were laws concerning the monitoring and enforcement of awards, and a small bureaucracy to back them up. The more powerful de facto system of monitoring and enforcement was conducted by trade unions and their workplace representatives. The tribunal offered a generally quick, inexpensive and efficient mechanism for dealing with problems with the implementation of binding norms. One of the most significant mechanisms was that of ‘dispute notification’. This involved a registered organisation (either union or employer) filling in a simple form with a brief outline of the matter in dispute. Here the system offered those already with award coverage automatic access to an independent tribunal, who could ‘knock heads together’ or, if the matter merited it, move to arbitration so that the tribunal itself determined the outcome.
This system of awards operated alongside contractual regulation, though the two forms of regulation did not ordinarily intersect. Contractual arrangements could be, and very frequently were, superior to award conditions (as in ‘above award payments’) but could not derogate from them. Superior contractual arrangements were concluded at the individual level but also on a collective basis, although the legal status of private law collective agreements has been problematic.
While this system arguably had many responsive features, it was increasingly criticised during the 1980s and 1990s, particularly on the basis that it did not allow for the establishment of standards at a sufficiently decentralised level – awards operated generally at any industry rather than at firm level (although as we have seen awards could provide for local variation) – and because it accorded trade unions a central role at a time when membership of those organisations was decreasing.
The initial response to this criticism was to create a statutory system of ‘enterprise bargaining’ which enabled legally enforceable collective agreements to be concluded at the level of the firm. Unlike prior legal arrangements, these statutory agreements could derogate from awards in certain circumstances. With the enactment of the Workplace Relations Act 1996 (Cth), following the election of the federal Coalition government, decentralisation moved a step further. Statutory agreements that derogated from awards could now been concluded between individuals and employers in the form of Australian Workplace Agreements (AWAs).
Notwithstanding these reforms, the AIRC retained many of its important functions, including the test case function. Further, while the site of standard-setting notionally shifted from central to enterprise level, these enterprise agreements often imported many, if not all, of the significant award conditions into their terms. And despite legislative attempts to limit the content of federal awards, they still sometimes contained elaborate provisions covering most major conditions of employment. The WRA formalised this connection between the enterprise deals and the central system by creating a regime of certification. The AIRC was only to certify those enterprise agreements which passed a ‘no disadvantage test’. This meant that the new agreement could not deviate from the relevant award in ways that disadvantaged workers taking a global view of their previous and proposed entitlements. Thus while the decentred statutory agreements could deviate from award provisions, the award itself underpinned an assessment as to whether or not the worker was better off in all the circumstances. The extreme complexity of properly undertaking such an assessment was an issue not addressed by the legislature, and the ‘no disadvantage test’ has not successfully restrained downward pressure on working conditions in some sectors. However, for our purposes, the link with awards was maintained.
The test case function of the AIRC and its forerunners has been ‘one of the most significant regulatory institutions/processes in Australian social, economic and political history’. In these cases, the AIRC uses its power to vary awards to respond to a claim for uniform standards in relation to a particular aspect of work. All the test cases run since the inception of conciliation and arbitration have been instigated and prosecuted by the Australian Council of Trade Unions (ACTU). The cases were brought on behalf of the many thousands of Australian workers who are members of trade unions affiliated with the ACTU. Just how genuinely representative the ACTU position has been of the interests and aspirations of members is not clear. For example, Baird argues that the predominantly male union leadership neglected to pursue a goal of paid maternity leave. However, unions are at least legally bound to contain democratic processes and structures, and recent test cases have been more attuned to the needs of women workers.
The processes undertaken in determining a test case application are both rigorous and flexible. The most recent test case, known as the Family Provisions Case, is instructive (even though the WorkChoices legislation effectively stymies much of the outcome of that case by failing to insert several of the new standards into the minimum core contained in the Act and preventing its inclusion in awards ). The test case turned on a major contemporary issue – how to balance work with responsibilities associated with the care of family members.
In this test case, despite employer opposition to the major elements of the ACTU’s claim, a lengthy and fruitful conciliation process, superintended by a member of the tribunal, led to agreements on new personal leave standards. The main agreement was to increase the number of personal sick days a worker could use to take care of family members from five to ten days. Agreement was also reached on the continuance of the conciliation process after the arbitrated decision was handed down in relation to the definition of ‘family’, which relegates same-sex partners to the status of ‘household members’ of the employee. The arbitrated decision itself contained the seeds of future regulatory feedback and adjustment of standards, with the AIRC resolving to review the operation of its decision six months after it was made – a resolution that cannot now be carried into effect because of the loss of jurisdiction. The key arbitrated decisions were to create an ‘employee right’ to request part-time work to care for young children, as well as to request a second year of unpaid maternity leave.
In determining the Family Provisions case, the AIRC heard evidence from the ‘real world of work’, courtesy of workers and employers called by parties to give evidence about their personal experience of work/family balance. Academics gave evidence on a wide range of matters: the sociology of work, economics, labour law and so on. Detailed evidence was given by the ACTU and employers about the operation of work and family provisions in the United Kingdom and the European Union generally. Interveners such as the State and Territory governments developed a sophisticated position which they urged the AIRC to adopt, based on original research about the needs of workers and businesses. Non-industrial institutions were also heard: the Human Rights and Equal Opportunity Commission made submissions, as did the Australian Adoptive Parents Association. The AIRC is not bound by the formal rules of evidence, and there was much active questioning of witnesses, parties and intervenors by members of the Full Bench.
The test case process invites public scrutiny: anyone can attend the public hearings and the AIRC website provided daily transcripts of hearings as well as the full texts of voluminous exhibits on-line.
Finally, the outcomes of the test case process are subtle and multifaceted. In addition to the effective use of both conciliation and arbitration, the new standards would have become (if the pre-WorkChoices system had been retained) ‘plugged into’ the Commission’s general dispute resolution processes. Although the ‘right to request’ award provisions only shaped employer responses in a rather vague way, eventually norms of appropriate behaviour would have emerged from the iterative processes of conciliation and arbitration under the supervision of the Tribunal. In other words, the ‘loops’ in the AIRC regulatory system both consolidate and generate developments in labour standards and productivity, but only for so long as the federal tribunal is permitted to continue its dynamic engagement with its regulatory clients.
We note also that the case was not initiated by or dependent on legislative or executive action. The entire procedure was conducted by a specialist body at arm’s length from the formal political process.
The processes just described have been largely dismantled by WorkChoices. However, this dismantling has presented a delicate political problem for the government. Wiping the legislative slate clean and starting again carries political costs because of the threat to existing conditions. On the other hand, to bring about change to such an entrenched and legally embedded system takes a lot of ‘hard’ law. By and large, the government has moved against elements in the previous system, such as the central role of unions and collective processes, through the use of command and control regulation. This is clear from industry-specific legislative developments leading up to WorkChoices. As Howe shows, the building and construction industry laws are punitive measures which are designed to limit union power in that industry. Despite the neo-liberal inspired strictures against ‘third party’ intervention, the system creates a new permanent institution, the Australian Building and Construction Commission, and a bureaucracy to police union activity in the sector. Similarly, the government’s push to de-collectivise labour relations through the use of AWAs has been backed by legislative command in the higher education sector. University funding has been tied to the requirement that Universities offer all existing staff and new staff the option of signing an individual agreement. Again, within direct government employment, the government’s policy is achieved by making the signing of an AWA mandatory for all new staff in certain government departments.
It is therefore understandable, yet paradoxical, that WorkChoices uses ‘command and control’ techniques in many ways, all of which must be understood as devices to achieve the goal of regulating individual employment relations either by private law or even more so, by government command. Alternative modes of norm-setting are accorded inferior legislative status.
One of the most significant features of WorkChoices is that the legislation arrogates to itself the task of setting the minimum floor of conditions below which no worker covered by the system may fall. The test case procedure has been very substantially curtailed, if not effectively abolished. Parliament now creates labour standards, except for pay, which will be fixed by the Australian Fair Pay Commission (AFPC). These standards – relating to basic rates of pay, maximum ordinary hours of work, annual leave, personal leave and parental leave and related entitlements – are described as the ‘Australian Fair Pay and Conditions Standard’ (AFPCS). Under WorkChoices, the AFPCS is the only standard against which workplace agreements or contracts can be measured. Awards are no longer relevant to assessing other forms of statutory agreements; the ‘no disadvantage test’ has been abolished.
Of course, the regulatory meaning which should be accorded to a ‘floor’ of conditions depends on the level at which the standards are set. Very low conditions, for example, create a de facto space within which employer power may be exercised freely. Let us consider the standards set out in the WorkChoices legislation focusing on the time and pay.
The WorkChoices legislation regulates working time in a number of ways. First, through the device of the AFPCS, it creates a number of standards, some of which initially appear to be non-derogable but which on closer examination turn out to be porous. Those standards are expressed in the form of ‘guarantees’ pertaining to: maximum ordinary hours of work; annual leave; personal leave; and unpaid parental leave. The standards prevail over contracts and statutory agreements (now known as ‘workplace agreements’) to the extent that they are provide more favourable outcomes and cannot be excluded.
A second category of standards are default rather than mandatory standards and consist of certain norms which continue to subsist in the award system. Awards continue to exist, although no new awards can be made (other than as a result of award rationalisation ), but their scope has been substantially reduced and it is much more difficult to vary them to improve entitlements than in the past. Although the range of matters that can be included in awards have been reduced (they can no longer address issues such as carer’s or parental leave), they can continue to regulate certain important aspects of working time including ordinary hours of work, rest breaks, variation of hours, public holidays and additional payments for long or unsociable hours. This last matter is particularly important because it has historically been the mechanism for distinguishing between, on the one hand ordinary hours and days of work, and on the other times reserved for rest, family or community activities. The mechanism operates by imposing a financial penalty (in the form of increased wage payments) for requiring employees to work at ‘unsociable times’.
However, the WorkChoices reforms will make even the pared-down awards increasingly irrelevant since their application to employees is permanently displaced (in respect of the current employer) once an employee enters into any form of statutory agreement, except in relation to ‘protected conditions’. The abolition of the ‘no disadvantage test’ means that the nexus between awards and agreements has been broken. The ‘protected conditions’ in awards operate as default terms in workplace agreements and revive when an agreement is terminated. These conditions include additional payments for unsociable hours, but the ‘protected conditions’ can be displaced by express provision in a workplace agreement.
The third category of standards concerns public holidays. The WRA gives employees a ‘right’ to public holidays; this right is circumscribed by an employer’s right to request that an employee work on a public holiday. An employee may not unreasonably refuse to do so. As discussed below, this employee right is likely to prove illusory.
These provisions on working time are enforced by the imposition of monetary penalties in relation to breach of an award or statutory agreement. There is no statutory right to an injunction or specific enforcement to enforce these conditions, although parties to a workplace agreement might construct a dispute resolution clause which provides for such remedies.
As seen above, the WorkChoices scheme for regulating working time is essentially a ‘command and control’ form of regulation and lacks many core features of responsive regulation. The scheme consists of a set of government-mandated norms coupled with penalty provisions. First, the content of the AFPCS temporal provisions as well as that of the public holiday provision have been determined by the government without the involvement of affected persons or their representative organisations (save for the extremely brief Senate inquiry into the draft legislation). There has been no serious opportunity for public deliberation and debate on the appropriateness of the number of mandatory standards, or about the way in which they have been formulated. Moreover, while the default standards are the product of past negotiations between representative organisations, these are a relic of the previous system and as we have stated, there is little capacity for further production of standards through this process. WorkChoices abolishes the sophisticated, participatory, transparent and nuanced processes which have existed to date for setting new standards. While the government could arrange an inquiry (either through parliamentary committees or by appointing an external investigation), there is no obligation on it to do so, nor need any such inquiry be inclusive or transparent.
Secondly, the content of the mandatory standards are static. There is no process analogous to dispute notification to trigger a debate about new forms of standards with public involvement. New standards can be created only by party political means leading to legislative amendments. Again the award ‘test cases’ under the previous system provided such a process.
Thirdly, the element of decentralisation and localisation provided for in the previous system, by making awards industry-based, has been lost. Granted, WorkChoices aims to achieve decentralisation through encouraging negotiation at enterprise level. However, in the absence of mechanisms to support a right to collective bargaining, decisions over the form of workplace negotiations will be the unilateral preserve of the employer, with any prospective employee having to accept the conditions and their legal form if they want the job. Thus, negotiation will frequently be an analogue for private contracting, which, we have already observed, is an inadequate way to regulate time.
Indeed, the privileging of AWAs (evidenced by its priority over awards and collective forms of statutory agreement) not only complicates the relationship between public and private regulation – because the relationship between AWAs and contracts is quite uncertain – but is likely to inhibit or counteract the common law developments making contracts less employer-centric. In short, the legislation interferes with private ordering by conferring additional regulatory power on the employer.
Fourthly, in so far as norm-creation will occur at a decentralised level, there is no mechanism for public evaluation of those norms or for comparing them, identifying poor practices or diffusing good ones. WorkChoices fails to create a meaningful system of monitoring and enforcement of agreements based on the minimum entitlements in the AFPCS. It requires that AWAs (the most likely instruments for the carriage of the new low working time standards) be lodged with the Employment Advocate (‘EA’). However, there is nothing in the legislation that requires the EA to examine the AWAs and then to certify that they do not fall below the (low) standards set in the Act. This replicates the paradox noted above: on one hand the legislation evinces a concern with policing the system to protect workers, and on the other it undercuts any real protection by removing the legal requirement to monitor anything at all.
Fifthly, there are no provisions prompting firms to establish internal compliance mechanisms with the participation of those affected by the standards. This position may be contrasted with occupational health and safety law, where firms must establish autonomous committees to monitor the implementation of the law. Supervision of compliance through non-governmental organisations, most particularly trade unions, has been drastically curtailed through restrictive right of entry provisions. Heavy reliance is instead placed on bureaucratic enforcement though a souped-up labour inspectorate.
Lastly, at a fundamental level, the new standards fail to achieve the goal of ensuring decent work. Not only does it reject a responsive approach to standard-setting, but in the case of some of the temporal standards, it deploys ‘command and control’ regulation to grant power to employers to unilaterally determine particulars of the employment relationship. This is neither good regulation, nor does it represent progress towards the attainment of decent work.
In its WorkChoices policy document and the associated advertising campaign, the government stressed that the ‘command and control’ authority of centrally-made law underpins the new regime in ways, it was implied, which would foster the interests of workers because standards would be ‘protected by law’ and ‘guaranteed’. But rather than strong, if centralised and static, legally binding protection for workers, the WorkChoices hours provisions effectively delegate to one set of private actors, the employer, the task of fixing actual standards. When taken in combination with the legislative attack on collective organisation and representation, and individual job security rights, the traditional Australian system of relational determination of norms has been reconfigured, with the individual employer now the chief regulator in this important field. This represents a major change in the complex interrelationships within the regulatory space.
For example, some of the mandatory standards simply do not deliver on their ‘guarantee’. Take the ‘minimum entitlement’ concerning maximum ordinary working hours set by WorkChoices. It is described as a ‘Guarantee of maximum hours of work’. Section 226 of the Act provides that:
ii. Subject to sub-section (3), if the employee and the employer agree in writing that the employee’s hours of work are to be averaged over a specified averaging period that is no longer than 12 months – an average of 38 hours per week over the averaging period; and
b. Reasonable additional hours.
If s 226(1)(a)(i) stood alone, employers would be prohibited from engaging workers for more than 38 hours in any week, subject to the reasonable additional hours requirement in s 226(1)(b). However, the reference to the 38 hour week is illusory in legal terms, because s 226(1)(a)(ii), which enables the parties to agree to averaging, undercuts its normative content. Employers must not engage workers for more than an average of 38 hours per week, and the reference period for calculation of this average may be any agreed period up to twelve months. Thus, standard hours for workers could be, say, 70 per week for a number of weeks, provided their hours were then reduced sufficiently later in the reference period to bring the average down. The potential for variation in work scheduling is likely to present significant barriers to employment for those with domestic care responsibilities. What looks on its face like a hard rule protective of workers is in fact a gift of wide discretionary power to the employer to fix standard hours in any way deemed fit.
Against this conclusion, it might be said that the legislation permits employers to move from the 38 hour week guarantee to the average of 38 hours over a period of up to a year only if the employee agrees. However, as discussed above, the asymmetries of power – labour market power, information, personal wealth – between many employers and employees mean that the notion of actual bargaining around such terms is a fiction. This is especially so since channels for employee co-ordination have been restricted.
This act of acceptance on the part of the employee to a certain mode of legal regulation is likely to be bound up in the act of accepting the employer’s preferred method of hours regulation under s 226. Indeed, for those prospective employees with little labour market power and/or knowledge, the two actions are likely to be wholly blended into one. A job will be offered with certain conditions, and on acceptance the employee has given consent to a deregulated hours regime pursuant to s 226(1)(a)(ii).
A similar story can be told of the public holiday standards set in the legislation. During the parliamentary process, concerns were raised by members of the Senate that ‘iconic’ public holidays would be lost under the new regime. The debate was cast in terms of a concern that people would have to work on public holidays, without recognising that many people already do such work. As we noted above, the real question is a complex one which should be resolved through an open, rigorous process in which all relevant voices are heard. Like the general hours provision discussion above, the public holiday clauses set no real standard, and devolve wide discretion to the employer at the workplace level.
WorkChoices identifies seven public holidays (which can be added to by State laws). In what can be seen to be a parody of the ‘right to request’ model in relation to leave to fulfil family responsibilities, WorkChoices creates a system where employees can be asked to work on any of these public holidays. They may only refuse if their request not to work is ‘reasonable’.
The Act states that ‘in determining whether an employee has reasonable grounds for refusing a request to work on a public holiday, regard must be had to …’ and there follows a list of twelve factors. Many of the twelve factors will help employers create the circumstances in which they are able to argue that the employee’s refusal was not reasonable. For example, factor (a) is ‘the nature of the work performed’, factor (c) is ‘the nature of the employer’s workplace … including its operation requirements’ and factor (h) is ‘whether the employee has acknowledged or could reasonably expect that the employer might require work on public holidays’.
The idea that workers in non-shift working environments have to show cause why they should not work on public holidays, at an individual level in an unmediated meeting with their boss, is anathema to the previous system of regulation. The requirement that workers must explain, for example, their ‘reasons for refusing the request’ (factor (d)) and their ‘family circumstances (including family responsibilities)’ (factor (e)) shows that the regulatory system is essentially without meaning. How are employers to judge what constitutes a valid employee reason or the personal circumstances which amount to a valid reason? How are non-Christians to justify a desire to take Christmas Day as a public holiday? Are workers without dependents to be refused access to public holidays because they cannot show their domestic labour is needed at home? The legislation, despite all its verbiage and complexity, does not give sufficient direction to its designated decision-maker, the employer.
Under the previous system, eventually norms could evolve around even such badly constructed provisions. Under the auspices of the AIRC, unions could notify disputes about implementation of the standard, conciliation and arbitration would bring to light the issues impacting on production in particular sectors, changing community norms could be acknowledged, and eventually sensible rules would result. Under WorkChoices, employees may seek a review of the employer’s decision in the Federal Court. This is a forbidding endeavour – expensive, complex and unrealistic for most workers. Furthermore, any norms emerging from the Court process are unlikely to be as closely nuanced at those developed through the AIRC processes using the award as the regulatory instrument. No more industry specific rules will emerge; instead, the ‘one size fits all’ determination by a non-specialist tribunal may (if cases come before it) eventually make sense of the Act. The most probable regulatory outcome will be that workers will have to work on public holidays if their employer asks them to and they want to keep their job.
Under WorkChoices, the power to regulate what is perceived to be the most important labour standard (one of the five elements of the AFPCS), wages, has been removed from the AIRC. In contrast to the hours standard discussed in the previous section, minimum wage rates will not be set through legislation but will instead be determined by a new institution, the AFPC.
Little is known of the process of standard setting to be adopted by the AFPC. This is because the AFPC has been given a wide discretion to determine its own modes and procedures. However, there is enough to suggest that the AFPC will lack both transparency and responsiveness. This constitutes a regression since the public process of the AIRC, together with judicial oversight, upholds these values.
The AFPC, which the Government claims is modelled on the British Low Pay Commission (LPC), will set a single adult minimum wage on a periodic basis. It will also adjust award classification wages (the Australian Pay and Classification Scale, or APCS). The AIRC may still make a ‘workplace determination’ where there is industrial action threatening the national economy, but any decision on wages will be constrained by determinations of the AFPC. Does this new system of public regulation of wage standards represent a responsive model of regulation likely to result in dynamic and evolving standards, both in terms of the ‘floor’ to wages in the labour market, and the setting of pay rates above the floor through the classification structure?
The AFPC’s wage setting function must occur within the parameters set by WorkChoices. The Act states that the AFPC’s objective in performing this function is to ‘promote the economic prosperity of the people of Australia’ with regard to ‘the capacity of the low paid to obtain and remain in employment’ and ‘employment and competitiveness across the economy’, as well as ‘providing a safety net for the low paid’. There is no requirement that the AFPC have regard to ‘fairness’ in performing its functions. In setting these parameters, the Government seeks to give effect to its rhetoric that granting employers more ‘flexibility’ with regard to labour standards will generate employment per se, with little regard paid to the quality of employment that might be ‘created’ as a result. So although wages, unlike other aspects of the AFPC, will not be set by the legislation, the AFPC will have much less discretion to ensure that wage levels operate as a component of decent work standards than was the case with the AIRC.
Furthermore, it should be noted that the content of the APCS will not become known until the Award Review Taskforce (‘ART’) appointed by the government has completed the process of award rationalisation and simplification. WorkChoices requires the AFPC to have regard to the recommendations of the ART. It is likely that the AFPC will be working with a much-simplified classification structure, which will mean that the AFPC will have less scope for nuanced approaches to the setting of pay standards above the minimum wage. Under the previous system, the AIRC set minimum wages for the entire range of job grades in each industry. For example, in higher education, the relevant awards set minimum rates for all categories of work ranging from research assistant to professor.
The AFPC lacks responsiveness to the community since there is no requirement that it be constituted by representative groups, nor is it obliged to hear from them. This may be contrasted with the explicitly tripartite structure and consultation requirements of the LPC in the United Kingdom. First, with respect to the structure of the AFPC, WorkChoices simply requires that the Chairman have ‘high levels of skill in business or economics’, while the other four members must have experience in one or more of ‘business, community organisations, workplace relations and economics’. The Government has appointed Professor Ian Harper, a financial economist previously at the Melbourne Business School, as the first Chairman of the AFPC. The other members of the Commission appointed by the Government consist of another academic economist, a businessman, the chair of a welfare agency and a public servant who had at an earlier point in his career served as a trade union official.
Second, the AFPC is not under any legal obligation to conduct its wage setting function through a public, transparent and responsive process, or to have its preferred sources of evidence subject to scrutiny as part of the wage setting process. In defining the AFPC’s wage setting function, WorkChoices distinguishes between the conduct of ‘wage reviews’ and the making of ‘wage-setting decisions’ based on the outcomes of wage reviews. Frequency, timing, scope and manner in which wage reviews are to be conducted are matters to be determined by the AFPC. The AFPC ‘may inform itself in any way it thinks appropriate’, although WorkChoices provides that the AFPC may undertake or commission research, consult with any other person, body or organisation, or monitor and evaluate the impact of its wage-setting decisions.
The Chairman of the AFPC has indicated that, following the practice of the LPC, his preference is for the AFPC to engage in consultations at the places ‘where low-paid workers and their employers work or live’ over ‘formal hearings’. Although the Chairman acknowledges that the latter may nevertheless be one means by which information is gathered, he has promised that these will not take the form of judicial proceedings. However, he has given no indication as to the frequency or nature of any formal hearings the AFPC might hold.
While the Chairman’s statements regarding his plans for consultation appear to facilitate a participative regulatory model, in that they imply some scope for involvement of people who will be affected by the AFPC ’s wage setting functions, this suggests a rather superficial form of participation. Returning to our model of responsive regulation, we are less able to say at this stage that the processes will be transparent and democratically sound. As yet we have no indication as to the role that representative bodies such as trade unions will be able to play in this consultation process. We are unclear as to what extent regulatory actors outside the AFPC will be able to test or contest the evidence provided through the consultation process.
This last criticism can also be levelled against another method that the AFPC will use in performing its wage setting function. The AFPC will also, like the LPC, commission its own research to assist in the carrying out of its functions. It is as yet unclear how much of this research will be carried out ‘in-house’ and how much will be contracted out to external organisations or institutions, let alone which organisations or institutions will be engaged. In the case of the LPC, only a small percentage of its research has been carried out in-house. It has (deliberately) commissioned research from a wide variety of groups, some providing advocacy and advice for the low paid, and other work by labour market consultancy firms and academics.
Of course, the AIRC has always considered research presented to it by the various parties appearing before it through the test case process, commenting on and analysing that research in its decisions. This evidence has therefore (in a responsive manner) come from regulatory actors, including business groups, trade unions and community organisations, rather than being obtained by a state regulator in the first instance. It has been pointed out that the AIRC has in the past refused to commission its own research in relation to the ‘living wage’ test case on the basis that if it did so ‘it would be required to make a number of judgments about questions to be asked and technical matters, raising potential issues in relation to its role as the independent arbitrator of the claims before it.’
The AFPC represents a less effective regulatory institution than the previous system. Wage fixing under WorkChoices will permit less responsive and reflexive regulation, and when the work of the ART is completed, it is likely that the uniquely Australian ‘tapestry’ of wage rates will be gradually displaced by the legal regulation of only one centrally determined minimum wage. This will have negative implications for the capacity of wages to continue as a dynamic and evolving labour standard beyond annual adjustment to the ‘floor’ of wages.
In approaching our analysis of WorkChoices, we have argued that the regulatory framework pertaining to work relationships should promote, not simply work per se, but rather decent work. We have also argued that decent work is unlikely to be achieved by the use of private regulation alone; public regulation of work, and in particular the employment relationship, is needed in order to establish appropriate workplace standards. The lessons of regulatory theory suggest that public regulation should be responsive: standards need to be created through dynamic, participative processes that both engage actors at the local level and provide for continuous evaluation.
WorkChoices fails to promote decent work through responsive regulation, adopting instead ‘command and control’ methods. It establishes labour standards by parliamentary fiat, without sufficient public deliberation. It creates no mechanism for the ongoing revision of standards. It emphasises bureaucratic enforcement rather than the participation of firms, employees and their organisations in compliance. Indeed, WorkChoices uses public regulatory power not to overcome the deficiencies of private regulation, but to exacerbate them.
Judging from the standpoint of responsive regulation, WorkChoices is a retrograde measure. The award-based system it displaces possessed many more responsive features. We do not conclude that the preferable direction of legal reform is to revert to the previous system. The award-based system may be criticised for its constitutional complexity, the proliferation of regulatory instruments, its gendered orientation and, to some extent, the privileging of certain forms of worker associations (registered unions) over alternative and/or complementary types of worker and community self-organisation.
Nonetheless, these deficiencies in the award-system should not obscure its attractive features, which progressive reform of workplace law will do well to build on. These features are epitomised in the test case: an inclusive, dynamic deliberative, evidence-based process of setting new forms of labour standards, a process responsible for many of the major innovations in workplace law in this country. We would hope that any future reform of the WorkChoices legislation would reinstate and update the test case process, in order that Australia’s working people may experience once again reasoned, incremental improvements in their employment conditions.
[*] Law School, University of Melbourne.
[**] Law School, University of Melbourne.
[***] Law School, Latrobe University.
 This Act amends the Workplace Relations Act 1996 (Cth) (‘WRA’).
 ‘Regulation’ has a variety of meanings: Julia Black, ‘Decentring Regulation: Understanding the Role of Regulation and Self-Regulation in a ‘Post-Regulatory World’’ (2001) 54 Current Legal Problems 103, 128-142. In this paper, we generally use the term to refer to state-based law, including law produced by the courts and by legislatures.
 On the propensity for law to fail in these ways, see Christine Parker et al (eds), Regulating Law (2004) 10.
 See generally Joellen Riley, Employee Protection at Common Law (2005).
 And indeed such regulation predates the application to work relationships of modern private law doctrines: Richard Johnstone and Richard Mitchell, ‘Regulating Work’ in Parker et al, above n 3, 101, 103-108.
 Andrew Stewart, ‘Redefining Employment? Meeting the Challenge of Contract and Agency Labour’ (2002) 15 Australian Journal of Labour Law 235.
 This is not to suggest that work performed outside the employment relationship should not be acknowledged in a discourse on workplace rights. For example, volunteer workers who attend a workplace should receive full health and safety protection, and should be protected against unlawful discrimination.
 The Australian Bureau of Statistics reported in August 2005 that there were 8 526 000 Australians in an employment relation: Australian Bureau of Statistics, Employee Earnings, Benefits and Trade Union Membership (6310.0) (2006) Australian Labour Market Statistics <http://www.abs.gov.au/AUSSTATS/abs@.nsf/ProductsbyReleaseDate/99E5614783415356CA25713E000F92B1?OpenDocument> at 3 November 2005. The Bureau reported in April 2006 that there were 10,041,000 Australians in paid work: Australian Bureau of Statistics, Australian Labour Market Statistics (6105.0) (2006) <http://www.abs.gov.au/AUSSTATS/abs@.nsf/ProductsbyCatalogue/E699A367D4B29EADCA256D16007DA2D8?OpenDocument> at 5 April 2006.
 Job creation now forms the most significant rationale for minimum wage fixing under the Australian Fair Pay Commission. The Chairman recently said that ‘[a] key focus of the Fair Pay Commission’s work will be to ensure that minimum wages in Australia, as far as is practicable and over time, do not impede unemployed people from gaining employment and do not induce employers to shed low-paid workers.’: Ian Harper, ‘Ensuring Fair Pay: the First Steps’ (Press Release, 16 February 2006). See also WRA ss 21, 23.
 Guy Standing, Global Labour Flexibility: Seeking Distributive Justice (1999).
 Ibid 3–4.
 Ibid 9. Standing notes that in the 1 9th century, to be ‘in employment’ was ‘a term if not of abuse, at least regret’. See also Anthony O’Donnell, ‘Reinventing Unemployment: Welfare Reform as Labour Market Regulation’ in Chris Arup et al (eds), Labour Law and Labour Market Regulation: Essays on the Construction and Regulation of Labour Markets and Work Relationships (forthcoming).
 Standing, above n 10, 394.
 See, eg, Jill Murray, ‘The Legal Regulation of Volunteer Work’ in Chris Arup et al, above n 12.
 Simon Deakin and Frank Wilkinson, The Law of the Labour Market: Industrialization, Employment and Legal Evolution (2005) 284. On this analysis, the living conditions of those not in paid employment should be supported by the state, and their capacity to participate in the labour market must be developed by state action in relation to education, housing, transport and so on.
 International Labour Organisation, Decent work – the heart of social progress (2004) <http://www.ilo.org/public/english/decent.htm> at 17 April 2006. See also Rosemary Owens, ‘Decent Work for the Contingent Workforce in the New Economy’ (2002) 15 Australian Journal of Labour Law 1, 7-12.
 Hugh Collins, Justice in Dismissal: the Law of Termination of Employment (1992) 16.
 Ibid 18. See also Hugh Collins, ‘Justifications and Techniques of Legal Regulation of the Employment Relation’ in Hugh Collins, Paul Davies and Roger Rideout, Legal Regulation of the Employment Relation (2000) 3.
 The following discussion refers to individual employment contracts. Collective private ordering is an important form of regulation, but it is frequently found not to have legal effect: see below n 56.
 Riley, above n 4, 40-49.
 Hugh Collins, ‘Market Power, Bureaucratic Power and the Contract of Employment’ (1986) 15 Industrial Law Journal 1.
 Mahmud v Bank of Credit and Commerce International SA  UKHL 23;  AC 20; Eastwood v Magnox Electric plc  UKHL 35.
 Johnstone v Bloomsbury Health Authority  1 QB 333 (Browne-Wilkinson and Stuart-Smith LJJ).
 Although there has been some qualified support by a number of Australian judges: Burazin v Blacktown City Guardian Pty Ltd  IRCA 371; (1996) 142 ALR 144; Thomson v Orica Australia Pty Ltd  FCA 939; (2002) 116 IR 186; Heptonstall v Gaskin (No 2)  NSWSC 30. Also, analogous employer obligations may sometimes be able to be derived from Australian tort law: State of NSW v Seedsman  NSWCA 119.
 See, eg, Mayer v Australian Nuclear Science and Technology Organisation  FMCA 209.
 See, eg, State of Victoria v Schou (No 2)  VSCA 71; (2004) 8 VR 120; Song v Ainsworth Game Technology  FMCA 31. For discussion, see Jill Murray, ‘Work and Care: New Legal Mechanisms for Adaptation’ (2005) 15 Labour and Industry 67-71.
 Riley, above n 4, 43-45.
 See, eg, Sidney Webb and Beatrice Webb, Industrial Democracy (1897).
 Simon Deakin and Frank Wilkinson, ‘Rights vs Efficiency? The Economic Case for Transnational Labour Standards’ (1994) 23 Industrial Law Journal 289, 291.
 Lucy v Commonwealth  HCA 32; (1923) 33 CLR 229, 237; Byrne v Australian Airlines Ltd (1995) 185 CLR 410, 428.
 Similar controversy exists around termination of employment.
 Deirdre McCann, Working Time Laws: A Global Perspective (2005) 1.
 Colin Scott, ‘Analysing Regulatory Space: Fragmented Resources and Institutional Design’ (2001) (Summer) Public Law 329.
 John Howe, ‘“Deregulation” of Labour Relations in Australia: Towards a More “Centred” Command and Control Model’ in Chris Arup et al, above n 12.
 For further discussion of ‘command and control’ regulation, see Robert Baldwin, ‘Regulation after ‘Command and Control’’ in Keith Hawkins (ed), The Human Face of Law: Essays in Honour of Donald Harris (1997) 65.
 These problems are discussed extensively in Eugene Bardach and Robert Kagan, Going by the Book: The Problem of Regulatory Unreasonableness (1982).
 See, eg, Neil Gunningham and Peter Grabosky, Smart Regulation: Designing Environmental Policy (1998) 50-56.
 The major source for this conception of regulation is Ian Ayres and John Braithwaite, Responsive Regulation: Transcending the Deregulation Debate (1992). For a discussion in the employment context, see Cynthia Estlund, ‘Rebuilding the Law of the Workplace in an Era of Self-regulation’ (2005) 105(2) Columbia Law Review 319, 356-3 66.
 Ayres and Braithwaite, above n 38, 54-100.
 Ibid 101-132.
 The internationalisation of compliance is discussed comprehensively in Christine Parker, The Open Corporation: Efective Self-regulation and Democracy (2002).
 Clarence Wilfred Jenks, Social Justice in the Law of Nations: the ILO ’s Impact after Fifty Years (1970) 76.
 See, eg, Keith Ewing (ed), Working Life: A New Perspectives on Labour Law (1996) 21. Ewing proposes five underpinning principles for labour law: equality of opportunity, social justice, workplace democracy, protection of civil liberties, fairness and work.
 Scott, above n 33, 331.
 Howe, above n 34, 7.
 For a comprehensive account of the system in place prior to the WorkChoices amendments, see Breen Creighton and Andrew Stewart, Labour Law (4th ed, 2005).
 States systems were also significant, but are being overridden by WorkChoices: WRA sch 8.
 See, eg, Attorn ey-General for the State of Queensland v Riordan  HCA 32; (1997) 192 CLR 1 (Kirby J).
 Jill Murray, ‘The AIRC’s Test Case on Work and Family Provisions: The End of Dynamic Regulatory Change at the Federal Level?’ (2005) 18 Australian Journal of Labour Law 325.
 Laura Bennett, Making Labour Law in Australia: Industrial Relations, Politics and Law (1994) 131-143.
 Creighton and Stewart, above n 46, 153-156.
 Byrne v Australian Airlines Ltd (1995) 185 CLR 410.
 Ryan v Textile Clothing & Footwear Union Australia  VicRp 67; (1996) 130 FLR 313.
 WRA (as in force prior to 27th March 2006) s 89A.
 WRA (as in force prior to 27th March 2006) s 170XA prohibited the certification of agreements which disadvantaged employees ‘in relation to their terms and conditions of employment.’ Section 1 70XA(2) stated that ‘disadvantage’ occurred where the agreement would result, on balance, in a reduction in the overall terms and conditions of employment’ of the worker.
 Richard Mitchell et al, ‘What’s Going on with the “No Disadvantage Test”? An Analysis of Outcomes and Processes under the Workplace Relations Act 1996 (Cth)’ (2005) 47(4) Journal of Industrial Relations 393.
 The following section draws on work originally from Murray, above n 49, 325.
 Marion Baird, ‘Parental Leave in Australia: The Role of the Industrial Relations System’ (2005) 23 Law in Context 45.
 (2005) 143 IR 245.
 Carer’s leave and parental leave are not longer ‘allowable award matters’: Compare WRA (prior to 27th March 2006) s 89A(2)(g) and (h) with the current WRA s 513.
 See generally information about the proceedings: Australian Industrial Relations Commission, Family Provisions Case (2005) <http://www.e-airc.gov.au/familyprovisions/> .
 Howe, above n 34.
 See Higher Education Legislation Amendment (Workplace Relations Requirements) Act 2005 (Cth).
 See conditions of employment offered at Department of Employment and Workplace Relations: DEWR, Salary and Benefits (2006) <http://www.dewr.gov.au/dewr/WorkingInDEWR/AboutWorkingInDEWR/Salaryandbenefits.htm> .
 WRA pt 7.
 WRA pt 7 div 3.
 WRA pt 7 div 4.
 WRA pt 7 div 5.
 WRA pt 7 div 6.
 WRA pt 8.
 WRA s 172(2).
 WRA s 173.
 An additional default entitlement is to meal breaks (WRA pt 12 div 1). These can be displaced by awards or workplace agreements.
 WRA pt 10 div 4.
 WRA s 513.
 WRA pt 10 div 5. In most cases, a substantive variation of an award will be permissible only if it is essential to maintain a minimum safety net of standards: s 553.
 WRA ss 513(1)(a), (c), (f), (g), (i), (j).
 WRA ss 349, 399(3)(b).
 WRA ss 354, 399(3)(b).
 WRA s 3 54(2).
 WRA pt 12 div 12.
 WRA s 719. Damages are recoverable in the case of AWAs.
 The default dispute resolution clause for workplace agreement contains no such mechanism: WRA s 353, pt 13.
 Senate Employment, Workplace Relations and Education Committee, Senate, Provisions of the Workplace Relations Amendment (Work Choices) Bill 2005 (2005) [1.1-3.88]. One week was allowed for public hearings.
 See WRA s 400(6).
 WRA ss 348(2), 349.
 Joel Fetter and Richard Mitchell, ‘The Legal Complexity of Workplace Regulation and Its Impact upon Functional Flexibility in Australian Workplaces’ (2004) 17(3) Australian Journal of Labour Law 276.
 Other than measures that might be taken (but are not required to be taken) by the Employment Advocate. However, unlike the AIRC, the Employment Advocate is not set up to provide for public deliberative processes.
 WRA pt 8 div 5.
 See, eg, Occupational Health and Safety Act 2000 (NSW).
 WRA pt 15.
 WRA pt 6.
 WRA s 226(3) deals with the situation where an employee agrees to an averaging period, but starts work during the time when the averaging period is already running at that firm. In such a case, ‘that averaging period is taken … not to include the period before the employee started to work for the employer’.
 This is defined in WRA s 226(4).
 Research into the content of AWAs supports this conclusion: see the submissions to the Senate inquiry into workplace agreements, the large majority of which were critical of AWAs: Senate Employment, Workplace Relations and Education Committee, Senate, Workplace Agreements (2005) <http://www.aph.gov.au/Senate/committee/eet_ctte/indust_agreements/submissions/sublist.htm> . The majority report itself was also highly critical of AWAs. The report was prepared by Senators from the Australian Labour Party and the Australian Democrats (the Committee was constituted prior to the Coalition taking control of the Senate).
 Workplace Relations Act 1996 (Cth) s 613.
 These are being reviewed by the Award Review Taskforce and are likely to be greatly reduced, possibly to a template of four levels (or some argue for one) for all industries.
 WRA s 23.
 This can be contrasted with the criteria which the AIRC was required to have regard to under the WRA prior to WorkChoices. Section 88B of that legislation identified ‘the need to provide fair minimum standards’.
 WRA s 177. The ART, however, is not created by statute. It identifies itself as part of the executive branch of government: see ART, Terms of Reference (2005) <www.awardreviewtaskforce.gov.au>.
 See National Minimum Wage Act 1998 (UK) c 39, s 7, sch 1. The Blair Government in the UK consulted with both the Trades Union Congress and the Council of British Industry, Britain’s peak trade union and employer bodies respectively, in the appointment of commissioners. The nine commissioners forming the LPC are balanced between three people with a trade union background, three people with an employer background and three academic labour relations specialists: William Brown, ‘The Operation of the Low Pay Commission’ (2002) 24(6) Employee Relations 595.
 WRA s 38.
 WRA s 22.
 WRA s 24.
 See above n 9.
 See Australian Industrial Relations Commission, Safety Net Review – Wages (May 2003), PR002003; Kenneth Davidson, ‘Limited resources for low-wage earners as commission rides a tough timetable’, The Age (Melbourne), 21 February 2006, 1.
 In making these comments we must acknowledge that it is possible that the AFPC will evolve as an institution in ways that are not contemplated or provided for by WorkChoices. After all, the AIRC award-making and ‘test case’ process itself evolved over time without specific legislative authority or direction. Moreover, it is possible that the AFPC may come to accept that the maintaining job quality through the wage classification system can contribute to the attainment of economic prosperity.
 For example, the ‘tapestry’ acknowledged and rewarded differences in skill and educational levels on an industry specific balance – differences which would not necessarily be rewarded in an individual bargaining process.
 See, eg, The Centre for Employment and Labour Relations Law, Submission to the Senate Employment, Workplace Relations and Education Committee Inquiry into the Workplace Relations Amendment (WorkChoices) Bill 2005 (2005) <http://www.law.unimelb.edu.au/celrl/activities/reports.html> .
 Murray, above n 49, 325.