Reviving Australia’s system of enterprise bargaining

Introduction

In 1993, the then Australian Labor Party (ALP) Prime Minister Paul Keating set out his vision for a future of industrial relations focused on enterprise bargaining. He placed industrial relations reform as a central element of a broader micro-economic reform agenda. That agenda was designed to increase the flexibility of the economy, boost productivity, and in turn increase living standards and reduce unemployment. It is uncontroversial to state that much of that micro-economic reform agenda has been delivered and many of the anticipated benefits have been realised. However, what is not so clear is whether the industrial relations reforms outlined by Keating have been achieved. There is no doubt that Australia’s system of wage determination has undergone dramatic change since the early 1990s. However, the original goals that enterprise bargaining would dominate pay fixation and that awards would become a mere safety net have not been fully realised.

This article outlines the increasingly limited reach of enterprise agreements, both in coverage and quality, and the continued pervasive influence of awards in the Australian industrial relations system. It concludes with some proposals for reform to bring the enterprise bargaining system back into line with the goals outlined by Keating in 1993.

Paul Keating’s grand vision

Between its election in 1983 and 1991, the Hawke Labor Government presided over a highly centralised wage-fixing system which sought to limit all wage increases to those provided for in a set of wage principles determined by the Australia Industrial Relations Commission (AIRC) in a series of National Wage Cases. The vast majority of employees were covered by either federal or state awards which contained detailed provisions governing wages and conditions of employment. While many private sector employees received some level of ‘over award’ pay, in practice the only wage increases permitted by the wage-fixing principles during this period were those contained in awards of the AIRC (or its state counterparts).

This centralised system was seen by its critics as diverting attention from relationships at the workplace level, thereby, amongst other things, impeding productivity growth (and thus higher living standards) as well as more generally contributing to rigidities in the Australian economy (see e.g. Industrial Relations Study Commission, 1989).

In 1991, in at least partial recognition of these criticisms, the ALP Government and its Accord partner, the Australian Council of Trade Unions, persuaded the AIRC to adopt principles for bargaining at the enterprise level. These were subsequently reinforced by changes to the legislation to create a new form of statutory workplace agreement (known as Certified Agreements).

In April 1993, weeks after the ALP Government had been re-elected, Prime Minister Paul Keating made a major speech to the Australian Institute of Company Directors (AICD; Keating, 1993). Setting out a programme of micro-economic reform, Keating said there was a need to accelerate industrial relations reform ‘so that all the other elements of flexibility in the economy can work in greater harmony’.

Keating described the model of industrial relations he said the government was working towards. It was, he said:

a model which places primary emphasis on bargaining at the workplace level within a framework of minimum standards provided by arbitral tribunals.

Thus, the goals set by Keating included the following:

most employees would have their pay and conditions determined by enterprise agreements;

Legislation over the last 30 years

It is interesting to ask the question – to what extent have governments (of either political persuasion) legislated in line with the goals set by Prime Minister Keating?

Over the course of the last 30 years there have certainly been several major changes to the legislative framework governing industrial relations in Australia. All have retained an award system while providing scope for bargaining at the workplace or enterprise level. Nevertheless it is reasonable to infer that throughout this period the majority of employees remained covered by the provisions set out in the applicable award and that these continued to be legally enforceable unless overridden by means of a formal agreement (which generally required approval by an industrial tribunal or similar body). ¹

How have the NDT and the BOOT been applied in practice?

Given the significant roles they have played in the industrial relations system over most of the last 30 years it is important to consider how the NDT and the BOOT have operated in practice. The way in which these tests have operated has greatly affected the ability of organisations to make agreements that reflect their own circumstances rather than simply follow the arrangements set out in awards.

There has been some diversity of approach between different tribunal members – at least until late 2016 when the FWC adopted a more centralised approach to the agreement approval process (the ‘administrative triage’ process).

A number of issues have arisen in how to approach the application of the NDT or the BOOT. One concerns what weight, if any, should be given to the preferences of the employees covered by the agreement. A related issue is what weight one should give to the practical effects of any changes made to the terms and conditions of employment embodied in an agreement.

These issues can be demonstrated by considering the following example: take the fairly standard example of an award that has a ‘span of hours’ provision that provides that ordinary hours of work cannot commence before 6:00 a.m. Any hours worked before that time are not ‘ordinary hours’ and therefore attract overtime rates. Imagine a situation where the employees working for a particular employer would prefer to start work earlier than 6:00 a.m. (perhaps so they can finish their shift earlier in the afternoon). The employer in question is willing to start the shift at the earlier time, but is not willing to pay overtime rates, as this would be uneconomic. The employees and the employer make an enterprise agreement changing the span of hours so that it commences at 5:00 a.m., thereby enabling them to start their shift at that time. Let us assume there are no other differences between the agreement and the award. Should the FWC find that the agreement passes the NDT (or the BOOT)?

Under an approach that considers the preferences of the parties and the practical implications of the relevant agreement, the agreement in question would pass the NDT. When compared to the award, the only change is one that better meets the preferences of the employees, by enabling them to start work at 5:00 a.m. rather than 6:00 a.m. From their perspective they are ‘better off’, let alone not ‘disadvantaged’. I will call this approach the ‘evidence-based approach’ as it relies on evidence of employee preferences and/or the practical impact of the agreement on the actions of the parties involved.

Applying the BOOT in the example given might be more problematic. Even though the employees who made the agreement prefer to start work at 5:00 a.m., it is conceivable that a ‘prospective employee’ might prefer to start work at 6:00 a.m. The problem with ‘prospective employees’ is that they only exist in theory at the time the agreement is made – and there is no way of testing their preferences. One possible way around this might be a provision in the agreement that would allow any future employees to choose whether they wish to start work at the earlier time. Only if this is their preference (as opposed to something mandated by their employer) would they be paid at the ordinary rate.

This evidence-based approach gives weight to the preferences of the employees concerned as well as the practical effect of making the agreement – that the employer would be willing to start the shift at the earlier time in line with the employees’ preference if the enterprise agreement was approved. Under this approach the FWC would need to make an evidentiary finding that these are the facts.

However, there is an alternative approach. Under this approach one focuses simply on the text of the agreement and compares it to the text of the award. There is no regard to any evidence that may be available about the employees’ actual preferences or any practical effect of approving the agreement. This approach simply notes that under the award employees receive overtime rates of pay for time worked between 5:00 a.m. and 6:00 a.m., while under the agreement they receive ordinary rates. This is a disadvantage, and under this approach – if there are no other countervailing benefits – the agreement fails the NDT. I will call this the ‘text-based’ approach.

The evidence-based approach is consistent with a philosophy that the parties themselves are generally best placed to judge their own interests. Under this philosophy the role of the NDT (and/or the BOOT) is only to ensure that the employer is not able to use its power to force employees to make agreements against their own interests. Under this approach the issue before the tribunal with the change in the span of hours is an evidentiary one – is this something the employees actually want as opposed to being pressured into against their will?

Under the second, text-based approach the employees’ preferences are not considered to be relevant, nor are the practical implications of the agreement – that is, that the employees would now be able to start work at an earlier time. After all, as the proponents of this approach would point out, there is nothing in the text of the award that prevents the employer from starting work at 5:00 a.m. (even though it may not be willing to do so in practice).

In fact, both approaches have been used by members of the FWC at various times. However, it is the text-based approach that has generally been supported by full bench decisions and has come to dominate the way the FWC now deals with the BOOT.

One issue that has arisen that illustrates the differing approaches of Commission members (and the preference of the full bench for the second approach) is that of ‘voluntary additional hours’.

Many employers regularly need to find additional labour often at short notice, for a variety of reasons (for example, sick leave, unpredicted increases in workloads, etc.). Many employers will use casuals to perform such work (especially where the work is relatively unskilled) rather than giving it to their permanent workforce as overtime, because of the higher rate of pay that overtime attracts. This is despite the fact that members of the permanent workforce would in many cases like to be given the opportunity to work additional hours. It is simply cheaper for the employer to employ casuals rather than pay overtime to their permanent workforce, even though all other things being equal they might well prefer to give their existing employees the opportunity to earn more. One way of providing a solution to this dilemma that has often been suggested is by allowing for voluntary additional hours to be worked at ordinary rates of pay (while compulsory overtime is still paid at overtime rates). This makes it economic for employers to give the extra work to those in their permanent work force who wish to do that work.

This issue came before the Commission in 2003 in relation to the approval of a certified agreement for a group of security officers. Senior Deputy Polites approved the agreement at first instance. However, it was subsequently overturned by a full bench. In its decision (‘Security Officers’) the majority of the full bench said (at [79]) (footnotes omitted):

We fail to see how the distinction between hours performed beyond ordinary hours at the employer’s direction and hours performed with the voluntary agreement of the employee is available for the purpose of applying the no-disadvantage test, given the terms of the Award. To the extent it was relied upon, Polites SDP relied on an irrelevant matter and misconstrued and misapplied the no-disadvantage test. That test is to be applied by reference to the terms and conditions of the competing instruments – the Award and the Agreement. (MSA Security Officers Certified Agreement, 15 September 2003, PR930723)

For example, in 2010, in Fanoka (Application by Fanoka Pty Ltd T/A Fairview Orchards & Anors, 1 March 2010 [2010] FWA 2139), Senior Deputy President Richards approved 102 single enterprise agreements in the agricultural industry which featured voluntary additional hours clauses. The agreements were subject to the NDT because they had been made during the ‘bridging period’ under the Transitional Act.

In considering whether the clauses met the NDT, Richards SDP heard evidence from employers that they limited employees to working 38 hours a week because slim margins prevalent in the sector meant they could not afford to pay overtime.

Employee evidence suggested that they would be keen to take on extra hours, even without overtime rates, provided it was up to them whether or not to take on the work.

Richards SDP found that, on balance, the clauses satisfied the NDT, taking into account that they would:

enable employees electing to work extra hours to earn more than they otherwise would, allowing them to offset periods where work was scarce;

Richards SDP focused on the practical effect of the agreements and the ability of the employees to exercise a voluntary choice about whether to work the additional hours. He said (at [49]):

… the voluntary additional hours clauses in the agreements have tangible value to the employees to be covered by the agreements, particularly so where the arrangements are only able to be initiated by the employee and where there is no meaningful loss of an opportunity for paid overtime (for reason there is no demonstrable pattern of paid overtime in the industry sector in which these agreements are to operate). That is, the voluntary additional hours clauses only allow employees voluntarily to take advantage of hours of work which would otherwise not be accessible to them.

However, it was the text-based approach that was upheld by a full bench of the FWC later that same year in dealing with an appeal against a decision by Commissioner Smith not to approve an agreement involving BUPA Care Services (BUPA).

In its decision ([2010] FWAFB 2762) the full bench noted that the BUPA agreement contained a preferred-hours arrangement. Under the relevant clause employees might wish to work additional hours at particular times or on a continuous basis or during particular roster periods due to ‘personal circumstances’. Examples of personal circumstances were family and/or carers’ responsibilities and/or financial circumstances.

Under the clause in question an employee could at his or her initiative advise the employer in writing of the additional hours of work that he or she wished to be rostered to work. Where an employee requested to work additional hours on an available shift, the employer would endeavour to roster the employee to work on that shift. Employees would be paid at ordinary rates (plus any applicable shift penalties) for any additional hours worked at the employee’s request under the clause. Additional work at the employer’s request or direction would be paid at normal overtime rates.

The full bench, in upholding Smith C’s decision to refuse to approve the BUPA agreement, referred to the previous full bench decision in Security Officers. It said (at [25]):

We consider the views of the majority in the Security Officers case are apposite to the appeal before us. Where there is a relevant reference instrument, the application of the “no-disadvantage test” requires a comparison of the terms and conditions of employment in the enterprise agreement against the terms and conditions of employment in any relevant reference instrument and an assessment of whether the terms and conditions in the enterprise agreement result, on balance, in a reduction in the overall terms and conditions of employment under any relevant reference agreement. The “no-disadvantage test” does not involve an analysis of matters other than the terms and conditions of the enterprise agreement against those in any relevant reference instrument. The effect the terms and conditions may have on the actions of the employer or employee is not relevant to the “no-disadvantage” test.

Interestingly, there are provisions in the Explanatory Memorandum to the FW Act which suggest that the evidence-based approach should be used in applying the BOOT. The relevant paragraphs relate to the application of the BOOT to individual flexibility arrangements (IFAs). IFAs allow employers and individual employees to make written agreements that can vary the application of awards (or enterprise agreements) if certain conditions are met, including that the employee must be ‘better off overall’ than under the applicable enterprise agreement.

Paragraph 867 of the Explanatory Memorandum gives the following example of an IFA that would pass the BOOT: an arrangement allowing an employee to start work early 2 days a week in exchange for giving up the overtime rates that work would normally attract.

An illustrative example follows: Josh works as a membership consultant at a gymnasium. Under the enterprise agreement applying to his employment, the ordinary hours of work are 37 hours each week to be performed in a span between 8am and 6pm each day. Hours worked outside this span attract penalty rates. Josh’s employer usually requires membership consultants to work from 9am to 5.30pm. In his spare time, Josh coaches an under-12s footy team. To do this, he needs to be able to leave work at 4pm on Tuesdays and Thursdays each week. He wants to start work at 7.30am on these days, but usually this would attract a penalty under the terms of the agreement. The agreement allows the employer and an employee to make an individual flexibility arrangement that varies the terms of the agreement dealing with hours of work and penalty rates. Josh approaches his employer and asks whether the employer will make an individual flexibility arrangement with him under which the employer agrees that Josh can work from 7.30am to 4pm on Tuesdays and Thursdays. Josh agrees that he will not be paid a penalty on these days, even though he starts work at 7.30am. Josh is genuinely happy to agree to this arrangement because it enables him to balance his work and personal commitments. The employer agrees to this arrangement.

While this discussion pertains to IFAs it is very hard to see why a similar approach should not be taken to applying the BOOT when dealing with applications for approval of enterprise agreements, at least where there is scope for individuals to exercise a choice. In particular it suggests that one needs to go beyond simply looking at the text of the relevant instruments and give weight both to employee preferences and the practical effect of the agreement on the actions of the parties involved.

Leaving aside the issue of the evidence-based approach versus the text-based approach, the BOOT is a significantly more prescriptive provision than the NDT. The difference between being ‘better off overall’ and ‘not disadvantaged’ is a relatively minor one. The biggest change introduced by the BOOT is that each and every employee who is covered by the agreement must be better off when compared to the award. Previously, under the NDT, it was open to the Commission to have regard to whether employees in general were disadvantaged by the agreement. Second, the BOOT must be met in relation not only to existing employees but also to any ‘prospective’ employee that is any employee who might be employed under the terms of the agreement. Both these changes, especially when combined with the adoption of the text-based approach, in practice make it very difficult to make an enterprise agreement that does anything more than ‘add on’ to awards.

The full implications of the BOOT only really became clear with the finding in 2016 by a full bench that an agreement made by Coles Supermarkets with the Shop Distributive and Allied Employees Association (SDA) failed the BOOT. This decision ([2016] FWCB 2887) focused primarily on the fact that at different times the rate of pay in the agreement was either higher or lower than in the award.

[11] As one would expect as a matter of simple logic, the more hours that are worked during times when the Agreement rates are higher, the better off an employee will be. Conversely, the more hours worked when the Award rates are higher, the worse off the employee will be compared to the Award. In other words, if an employee works predominantly at nights or on weekends, the higher base rate under the Agreement will be counterbalanced by lower penalties payable under the Agreement at these times.

The National Secretary of the SDA, Gerard Dwyer, said at the time that this decision created a ‘new environment’ for bargaining. In particular by requiring, in effect, that the position of every single worker be better off at every single hour under an agreement, it effectively ended the widespread practice of using enterprise bargaining to ‘roll up’ penalty rates to secure higher base rates. This, he said, had been an effective avenue for securing superior wages and conditions for workers as the economy has moved away from the traditional 5-day working week (Workplace Express, 2016).

The evidence suggests that the BOOT, at least as applied by the FWC since The Coles decision, combined with the text-based approach, together with the adoption of the administrative triage process (which has ensured these approaches are applied consistently by the FWC) has made it increasingly difficult for organisations to make enterprise agreements that diverge significantly from arrangements set out in the underlying award.

Trends in enterprise agreement coverage

The goal set by Paul Keating of close to 100% of employees under federal awards being covered by enterprise agreements has never come close to being achieved.

The federal department responsible for industrial relations (currently the Attorney-General’s Department) publishes regular data on trends in coverage by federal enterprise agreements. The number of employees covered by current federal enterprise agreements indicated by this data is set out in Figure 1.

OPEN IN VIEWER

Figure 1.

Number of employees covered by federal enterprise agreements.Source: https://www.ag.gov.au/industrial-relations/industrial-relations-publications/Pages/historical-table-current.aspx

This graph shows that enterprise agreement coverage rose consistently from the early 1990s, before plateauing from around 2010 to 2014, and has then been on a substantial downward trend. Total employee coverage is now at around the level it was 10 years earlier. It is worth noting that the total number of employees in Australia has grown throughout the period and was more than 60% larger in 2019 than it was in 1992 (Australian Bureau of Statistics (ABS) 6202.0 and 6203.0). Moreover, in the years immediately preceding 2010 Australia moved to a largely national system with the transfer of most employees from state systems to the federal system. This means that many state-based agreements were converted to federal agreements around 2009 and 2010.

The graph shows that the decline is primarily a private sector phenomenon. The number of private sector employees covered by federal enterprise agreements peaked in 2012 at 1,939,000 and is now around 30% lower at 1,375,000 – lower than it was prior to the introduction of the FW Act. Analysis conducted by Alison Pennington (2018) shows that the coverage rate for private sector workers (taking into account the growth in the overall number of employees during the period) fell from 19% in 2013 to 12% in 2017.

At the same time, the number of agreements that the Commission has found fail the BOOT – at least without the provision of undertakings – has grown significantly. This appears to reflect the more rigorous application of the BOOT in recent years (rather than any change in the nature of agreements being submitted for approval), especially since the Coles decision and the introduction of the triage process. While up until 2016 most agreements were approved without undertakings, since then over two-thirds of agreements have required undertakings to be approved (refer Image 5, FWC Annual Report 2018–2019). The effect of undertakings is often to require replacing provisions in the agreement with those set out in the award.

Concluding comments: Can enterprise bargaining be revived?

The evidence is that enterprise bargaining, at least in the private sector, is failing. The reduction in the number of agreements being made is undoubtedly due to a number of factors, including the decline in union coverage. However, the increasingly rigid application of the NDT and the BOOT is making it more and more difficult to make agreements that are anything more than ‘award add-ons’. This reduces the incentive – especially from an employer perspective – to make enterprise agreements.

Pennington (2018) suggests a range of reforms to revitalise collective bargaining. These include a shift to bargaining at the industry level, (further) restrictions on the ability to terminate agreements and increased union rights. Such changes might indeed increase the number of employees covered by agreements and would quite possibly lead to an increase in wages (Pennington’s main concern). However, they would do little to achieve the micro-economic goals set out in Keating’s AICD speech. In particular, they would do little to prevent agreements being simply ‘add-ons’ to awards.

A (possibly complementary) approach would be to revive enterprise bargaining by restoring the flexibility that the system was originally meant to encompass. In particular the BOOT should be replaced, and the NDT restored. The FWC should be able to use its judgement to approve enterprise agreements that overall do not disadvantage the employees. Whether in the text of the statute or in the accompanying explanatory memorandum and/or second reading speech, it should be made clear that the Commission should be able to have regard for evidence concerning the practical effects of the agreement for the employees as well as the preferences of those employees. Moreover, it should be made clear that the Commission should be able to make a judgement on the effect of the agreement on the employees as a collective – rather than insisting that every single employee be made better off. No doubt any such changes would be criticised by some as an attack on the safety net. However, I think it is much more likely that by making the system more flexible it will encourage the sort of ‘win--win’ agreements that can lead to increased productivity, increased wages and better conditions for workers.