Abstract
This article considers the issue of classification of platform workers in Australia and it is motivated by the uncertainty that persists with this issue as well as the negative impacts that may arise for workers and platforms in such an environment. We contend that awaiting a decision of the High Court, that may provide clarification on the correct classification of platform workers, is not a preferred option for resolving the issue. Introducing a third category of worker to the existing categories – employee and independent contractor – has the potential to introduce its own complexities. Another option is a legislative intervention whereby existing legislation would be amended to deem platform workers to be employees. We argue that a better alternative is for the Australian legislature to enact a legislative definition of employee similar to that in the Employment Relations Act (New Zealand) s 6(1)(a). A principle-based approach to drafting the provision, and a purposive approach to its interpretation appear to be effective means of addressing the indeterminacy that has pervaded the platform worker classification issue in Australia.
Introduction
The platform economy has significantly transformed the way people undertake work. In Australia, businesses have engaged workers as either employees or independent contractors. The increasing prevalence of platform workers raises questions as to the appropriateness of the current binary worker classification of either employee or independent contractor. According to Casale and Fasani (2021), platforms seek to characterise themselves as ‘mere intermediaries’ and describe their main function as ‘to bring people together’. This may be partly motivated by seeking to avoid obligations that arise upon workers being characterised as employees.
Stewart and Stanford (2017: 421) note that while platforms may make it easier for buyers and sellers to connect, the platform model of work can facilitate ‘a ruthless race to the bottom’. In this context there is an urgent need for the current uncertainty about the correct classification of platform workers to be resolved. A proportion of platform workers appear to be dependent on precarious work for reason of ‘gig work’ being easier to obtain than long-term employment (Singer, 2014) and, in this sense, they may be characterised as vulnerable workers (Parliament of Australia, 2021: 90-92). The vulnerable workers characterisation appears to apply to those platform workers who often do not possess the skills required to negotiate their own terms and conditions of engagement and who are engaged in the work predominantly as a matter of survival. Some platform workers are impacted by sociolegal, cultural and linguistic barriers, which causes intersectional vulnerability and, in turn, a power imbalance between the platforms and workers (Marmo, Sinopoli and Guo, 2022: 16). Such workers may be accepting a ‘contract of adhesion’, set by one party on a ‘take it or leave it’ basis (Owens et al., 2011: 164).
Zwick (2017: 679) argues that platforms have enacted ‘the neoliberal playbook’ to the extent they have misclassified workers, undertaken ‘regime shopping’ and employed ‘the most economically vulnerable’. A possible effect of this is that it places low skilled workers in a vulnerable position by being provided with less choice when negotiating the terms and conditions of the precarious platform work which they undertake. In this setting, the worker may consider that the process is less of a negotiation and is, more accurately, their begrudging acceptance of the work conditions offered, in the absence of the availability of other work options. There are several references to the low level of bargaining power for platform workers in the literature (e.g. Aloisi, 2018; De Groen, Maselli and Fabo, 2016).
This article is focussed on how the current uncertainty, about worker classification, for platform workers and platforms under current Australian law may be resolved. The article highlights the indeterminacy arising from the multi-factor test, which distinguishes employees from independent contractors based on a non-exhaustive list of factors, with relative weight ascribed to them depending on the facts of each case. This test is difficult to apply in the classification of platform workers.
Zhang (2022) highlights a consensus among the academic community for the establishment of a new test to determine labour protection in the sharing economy. To provide appropriate protections for platform workers, the current practice of worker classification must be reformed. In this article, we examine the classification of workers under the multi-factor test and consider the implications of the current approach for platform workers. Ultimately, we recommend a solution to resolving the uncertainty that appears to pervade the worker classification issue.
The remainder of the article is organised as follows. First, it sets out the method and contribution. Second, it briefly describes the significance of the worker classification issue. Third, it summarises application of the multi-factor test in the classification of workers, over several decades. Fourth, it contemplates the role of soft law and extra-legal solutions. Fifth, it reviews how the Fair Work Commission (FWC), and the Federal Court, have applied the multi-factor test in recent decisions relating to unfair dismissals. Sixth, after briefly considering Australian legislative responses and how the classification issue can be resolved, it reviews a recent New Zealand decision on the classification of platform workers. Finally, we set out our conclusions.
Method and contribution
Recognising that the law on worker classification is unlikely to be completely neutral or value free, we examine the issue of platform worker classification through a critical legal studies (CLS) lens. CLS theorists have set out to demonstrate the incoherence and the manipulability of labour law, and its contradictions which ‘it relies on and vainly attempts to resolve’ (Conaghan, 1987: 338). Galanter (1974: 149) refers to the legal system having an architecture that confers ‘interlocking advantages on overlapping groups’ referred to as ‘haves’. The legal reform that is, ultimately, recommended in this article can be evaluated in terms of whether it would provide positive legal effects and protect the interests of those who interact with it. There is evidence that protecting the rights of platform workers transcends issues that are traditionally determined along political party lines, as reflected in the strident opinion piece by British Conservative MP, Damian Collins, who sets out a compelling case in favour of the government taking action to protect the rights of platform workers (Collins, 2022).
Given the nature of the research problem, we analyse relevant case law, focussed upon decisions of courts and the FWC on platform worker classification. The analysis of case law is justified on the basis that it highlights the indeterminacy of the current law on the issue and the current uncertainty that has arisen from the application of the multi-factor test. As well as providing some insights into the prevailing indeterminacy in the categorisation of platform workers, the decisions analysed may also prompt the development of an appropriate solution. There is a concern that the reliance on common law tests has the potential to exclude a wide range of workers from the benefits of protective regulations, despite a view—originating from principles of logic and fairness—that such workers should rightly have access to these benefits (Stewart, 2002).
The research method used in the article is unique, in the context of the extant literature on the issue, and the ultimate contribution is the development of novel recommendations on how the indeterminacy in the classification of platform workers can potentially be resolved. An intervention to resolve the issue appears aligned with the broad intention of policymakers.
Scope
De Stefano (2016) distinguishes between crowdwork and work-on-demand. Whereas crowdwork often involves ‘microtasks’, ‘work-on-demand via app’ relates to more traditional working activities. The focus of this article is on work-on-demand via app. Whereas work-on demand via app is usually performed for individual consumers, crowdwork is more often used by enterprises and institutions (Cavallini and Avogaro, 2019: 179). The article is focussed upon the classification of platform workers in Australia and, consistent with this, only platforms that operate, or have operated, in Australia are discussed. After briefly reviewing the development of the control and multi-factor tests, the Australian case law referred to is focused upon cases involving food delivery platform workers.
The significance of worker classification
In Australia, the classification of workers—as either employees or independent contractors—has been determined upon the application of the common law multi-factor test. Australia's use of the terms ‘employee’ and ‘employer’ has the effect of limiting the rights and entitlements of many workers who are outside the common law employee definition, typically independent contractors (Forsyth, 2020a: 15). Similarly, businesses have limited responsibilities in respect of workers who are independent contractors rather than employees (Hardy, 2016: 86). The employment status of a worker determines whether they are entitled to the National Employment Standards, under Chapter 2, Part 2 of the Fair Work Act 2009 (Cth), which employers cannot displace. Whether a worker is an employee is an important issue, as employers have an obligation to comply with the National Employment Standards and other entitlements which apply to employees, but not to independent contractors (Sutherland and Riley, 2016: 389). There are several consequences of how a worker is classified, including those related to taxation, superannuation, claims of underpayment, adverse action, unfair dismissal, and allegations relating to sham contracting (de Flamingh and Cameron, 2015: 70).
The binary concept of the employment relationship is determinative for courts in their decisions on the liability that a business has to its workers (Hardy, 2016: 86). This is evident in a comparison of two High Court decisions - Zuijs v Wirth Brothers (High Court of Australia, 1955) and Stevens v Brodribb (High Court of Australia, 1986) (‘Stevens’). In the former case, it was found that the injured worker—a trapeze artist—was an employee and thus entitled to be compensated by the employer. In the latter case, it was found that the injured worker—the driver of a logging truck—was an independent contractor and thus not entitled to be compensated by the principal.
As well as being an important issue for workers and businesses, the classification of workers is also important in relation to liability that arises from workplace accidents. There were five platform worker deaths in Australia within a 2-month period in 2020 (Bonyhady and Chung, 2020). Worker classification is of great importance in terms of the rehabilitation of an injured platform worker as well as to the entitlement to compensation for families of workers who die while working for a platform.
The following section of the article considers the development and application of the multi-factor test. Arguably, one of the causes of the uncertainty arising from the application of the multi-factor test is that some doubt might exist as to the materiality of the factors that are to be considered. As the multi-factor test has developed from the control test, consideration could be given to whether and how the multi-factor test could be further developed to better resolve the issue of classification of platform workers. Alternatively, a third category of worker could be introduced. Although, at this writing, the Government's reform approach is anticipated to include creating a new class of ‘employee-like’ workers (Crowe, 2023), such a new category could create further regulatory avoidance options, by increasing the opportunities for employers to misclassify workers in order to avoid employment responsibilities (Stewart and Stanford, 2017: 230).
The multi-factor test in the classification of workers
As outlined in this section of the article, the issue of whether a worker is engaged as an employee or independent contractor has been a matter of dispute before the Australian courts and industrial tribunals since the 1940s. The distinction was described by Windeyer J (at [217]) as ‘rooted fundamentally in the difference between a person who serves his employer in his, the employer's business, and a person who carries on a trade or business of his own’ (High Court of Australia, 1963).
It was noted that ‘the question … as to how the existence of a relationship of employment is to be determined … has not squarely arisen in [the High] Court for 20 years’ (High Court of Australia, 2022 HCA 1) (‘CFMMEU [2022]’) (at [99]). Initially, the High Court considered the issue in seeking the meaning of ‘wages’ to determine whether a payroll tax was leviable. In Federal Commissioner of Taxation v J Walter Thompson (Aust) Pty Ltd (High Court of Australia, 1944), the court held that the artist performing in cooperation with others under the control of the producer was an employee. The basis of the control test was the relationship between ‘master’ and ‘servant’. ‘Control’ influenced the decision in Queensland Stations Pty Ltd v Federal Commissioner of Taxation (High Court of Australia, 1945). In this case, the drover was an independent contractor on the basis that he was paid upon the completion of a specified result, with the payer having no control over specifics of how the result was brought about.
In Stevens, the control test alone was held to be inadequate to determine whether a worker was an employee. Mason J noted (at [24]) that ‘the existence of control, whilst significant, is not the sole criterion by which to gauge whether a relationship is one of employment.’ This approach, of having regard to relevant indicia under a multi-factor test, rather than control being the determinative factor, was endorsed in subsequent decisions.
In Stevens, Mason J considered control or lawful authority to command as a prominent factor, but not the sole criterion (at [517]). His Honour noted a non-exhaustive list of potential factors for the determination of an employee or a contractor, such as the mode of remuneration, the provision and maintenance of equipment, the obligation to work, the hours of work and provision for holidays, the deduction of income tax, and the delegation of work by the putative employee. Stewart (2018: 56) notes that there is neither a set number nor combination of indicia that a court must consider, and no single factor is determinative.
The High Court endorsed the multi-factor test in Hollis v Vabu (High Court of Australia, 2001). The case dealt with compensation for the appellant, Hollis, who was injured by a bicycle courier working for the respondent. The court held that Vabu effectively employed the bicycle courier and was thus vicariously liable for his negligent actions. The joint judgment (at [11]) cited Stevens (at [29]), referring to ‘the totality of the relationship between the parties.’
Stewart and Stanford (2017: 426) note that, in recent years, Australia's Federal courts have been more inclined to have regard to the ‘substance or practical reality of an arrangement’ and to not limit the analysis to ‘the formal terms agreed by the parties’. Due to the relative weight that courts and tribunals may ascribe to the various indicia, there appears to have been a marked increase in this ‘grey zone’ between the employee and independent contractor categories (Stewart and McCrystal, 2019: 8). While written terms of contract have evidential value, a business may be structured in a way that attempts to ensure that the engagement of its workers is not subjected to the accountability and liability imposed under the industrial law and in particular the Fair Work Act 2009 (Cth). This was evident in Fair Work Ombudsman v Quest South Perth Holdings Pty Ltd (High Court of Australia, 2015) where the High Court held that the business had set up a, triangular, sham contracting relationship to engage a worker. By contrast, in Building Workers Industrial Union of Australia v Odco Pty Ltd (Federal Court of Australia, 1991), the Full Federal Court held that building workers hired through a labour hire company were independent contractors. Uncertainty that arises from the substance-based approach to the employee-contractor divide has intensified with the rise of platform work (Stewart and Stanford, 2017: 2).
In Construction, Forestry, Maritime, Mining and Energy Union & Anor v Personnel Contracting Pty Ltd (Federal Court of Australia, 2020) (‘CFMMEU [2020]’), Lee J (at [80]) identified three broad categories of tensions that arise from the intuitive nature of the multi-factor test. First, how to apply the ‘control’ indicium to multilateral relationships; second, to what extent is the question of whether the putative employee is conducting a business determinative of characterisation; and third, what weight should be given to the contract in the characterisation (CFMMEU [2020]).
In the appeal case (High Court of Australia, 2022 HCA 1), the multi-factor test was seemingly of lesser relevance to the determination of worker classification than it had been in earlier decisions. A 22-year-old backpacker, Mr McCourt pursued work with Construct Pty Ltd, a labour hire company. At the times of engagement, he signed a contract stating that he was a ‘self-employed contractor’. Subsequently, under a labour hire agreement he worked as a labourer on construction sites run by Hanssen, a builder, over a period of less than six months. Although there was no contractual relationship between McCourt and Hanssen, there was a labour hire agreement between Construct and Hanssen. When Mr McCourt's contract ended, the CFMMEU argued, on McCourt's behalf, that as an employee he was entitled to compensation under the Fair Work Act 2009 (Cth).
The joint judgment of Kiefel CJ, Keane and Edelman JJ noted that ‘such a test is apt to generate considerable uncertainty, both for parties and for the courts’ (High Court of Australia, 2022 HCA1, at [33]). The importance of the contract between the parties was emphasised by their Honours, who held that where the validity of a written contract was not in dispute, ‘there is no occasion to seek to determine the character of the parties’ relationship by a wide-ranging review of the entire history of the parties’ dealings (High Court of Australia, 2022 HCA1, at [59]). The primacy of the contract is such that ‘the ultimate characterisation of a relationship … must be concerned with the rights and duties established by the parties’ contract, and not simply an aspect of how the parties’ relationship has come to play out in practice but bearing no necessary connection to the contractual obligations of the parties’ (High Court of Australia, 2022 HCA1, at [61]). The court recognised that an employment relationship will not always be defined exclusively by a contract between the parties (High Court of Australia, 2022 HCA1, at [41]). The High Court held, however, that because the terms of the relationship between the two parties were comprehensively committed to a written contract, the validity of the terms was not challenged as a sham, and the terms of the contract were not otherwise varied, the rights and duties of both parties could be found exclusively in the written contract (at [43]). The approach taken in CFMMEU [2022] can be seen as quite distinct from the tendency of the High Court to place greater weight on the multi-factor test in its earlier decisions since Stevens.
CFMMEU [2022] was heard together with ZG Operations Australia Pty Ltd v Jamsek [2022] HCA 2 (High Court of Australia, 2022 HCA2) (‘Jamsek’). The two cases were considered to have raised similar points of law, but the outcomes were different for the putative employees. In CFMMEU [2022] it was held that the labour hire worker was an employee, while in Jamsek [2022] the delivery drivers were held to be independent contractors.
In Jamsek, two truck drivers who were previously employed by ZG Operations, before being made redundant, were subsequently re-engaged as independent contractors. Both had operated their own business partnerships with their spouses since the mid-1980s. The parties had individually purchased and maintained their own trucks with operatable insurance and they had engaged accountants to handle their taxation affairs. The parties worked predominantly for ZG Operations, making daily deliveries, but they were free to pursue other work after hours and on weekends. When they were terminated, Jamsek and Whitby argued that they were employees and, therefore, entitled to various benefits such as long service leave, and annual leave under the Fair Work Act 2009 (Cth), long service leave under the Long Service Leave Act 1955 (NSW), and superannuation under the Superannuation Guarantee (Administration) Act 1992 (Cth)
In Jamsek [2022], the High Court held that, provided the validity of the contract was not challenged, only the rights and obligations of the parties under the contract were relevant to the classification of the worker. The surrounding circumstances of the contract relationship were not relevant for determining whether the workers were contractors or employees. The High Court unanimously held that Jamsek and Whitby were independent contractors. A consequence of this determination was that they were not entitled to the benefits provided to employees under the Fair Work Act 2009 (Cth). Their entitlement to superannuation was still to be determined, due to an extended definition of an employee in the context of superannuation.
The plurality's reasoning in CFMMEU [2022] and Jamsek [2022] elevate these cases to the most significant decisions concerning the employee/independent contractor dichotomy since the multi-factor test was introduced in Stevens in 1986. Over the ensuing decades, the High Court's reasoning in Stevens was endorsed in several cases up until CFMMEU [2022] and Jamsek.
As the workers in CFMMEU [2022] and Jamsek [2022] were not platform workers, the uncertainty that persists with classifying this type of worker is likely to continue unless an appropriate reform is implemented. On one reading, the comments of Steward J (at [220]), in dissent, in CFMMEU [2022] highlight the need for a legislative intervention: any alteration of the law should be left to the legislature. Factors relevant to such decisions have included the effect on competing interests that should be consulted before any alteration of the law; the existence of significant economic implications of any change; the enactment of legislation evidencing parliamentary attention to the subject; the perceived undesirability of imposing retrospective liability, especially criminal liability, on persons; and the desirability, in particular cases, of not making any change until after intensive analysis of social data and public consultation, facilities typically unavailable to a court.
Soft law and extra-legal solutions
It is recognised that, in practice, regulation is not limited to black letter law, such as legislation, and that policymakers have a wide range of avenues available to them to regulate labour standards; this has been described as a spectrum of labour regulation, rather than a dichotomy between state regulation and public regulation (Ogus, 1995). Examples of light touch regulation techniques that policymakers can use, or promote, include financial subsidies, codes of practice, and best practice statements or guidelines (Howe and Landau, 2007). The influence of soft law, at the international level, is embodied in institutions including the International Labour Organisation and the Organisation for Economic Cooperation and Development, in terms of their guidelines for multinational enterprises (Verge, 2000).
Stewart and Stanford (2017: 432) argue that the development of platforms is such that a regulatory innovation is required to avoid labour protections being undermined by ‘the blurring of the relationship between intermediary and worker that has been a central feature of most platform-based business models.’ In this context, Australia could develop a clearer legislative test or a stronger common law precedent for worker classification that recognises the impact of the platform economy and its effect on entitlements for platform workers and the responsibilities of platform businesses.
There have been developments that may have the effect of improving working conditions for platform workers. For example, in June 2022, the Transport Workers’ Union and Uber were signatories to a statement of principles for greater worker protections; the four agreed principles were based on setting a minimum wage, dispute resolution mechanisms, voices for workers, and the enforcement of those standards (SBS, 2022). This is broadly a type of ‘soft’ regulation, which may be viewed, overall, as a positive development.
Stewart and Stanford (2017), refer to a need for researchers to ‘be eclectic and open-minded in pursuing any potential avenue for extending regulatory protections to gig work and workers.’ This implies that soft law could play a part here. In an opinion piece, the CEO of Uber suggested there should be a ‘third way’ for platform workers, arguing that new laws were required, as the current system in the United States, creates uncertainty and risk for the company (Khosrowshahi, 2020). This could suggest that some platforms see regulatory changes as inevitable, and that those platforms would prefer a third category of worker rather than platform workers being classified as employees.
Gyulavari and Kartyas (2022) argue that collective bargaining is a ‘must’ for platform workers, to enhance workplace stability and reduce uncertainty from the perspective of workers and managers. We contend that industry-wide collective bargaining would constitute an improvement and that, if it was introduced in the current environment, it should include all engaged platform workers, irrespective of whether they are employees or independent contractors.
Stanford (2019: 192) suggests that the ambiguity in labour regulation could be resolved by broadening the definition of employee to ‘include any workers whose work arrangements and pay are clearly dependent on the decisions of major firms.’ It is also recognised that there are substantial costs involved in enacting changes to the existing legal framework, which consist of the parliamentary resources required, and the need for businesses, workers, and others to educate themselves about the changes (Stewart et al., 2015). For this reason, if enacting or amending legislation is the preferred approach to the worker classification problem, the legislative change should, ideally, be one that is relatively simple, and which achieves the intended policy outcome while keeping such costs to a minimum.
The FWC on the classification of platform workers
This section of the article examines cases on the classification of platform workers, and highlights the prevailing uncertainty in distinguishing between employees and independent contractors.
In cases of the unfair dismissal of platform workers, it is only where the worker is an employee that they would have an action in unfair dismissal. Joshua Klooger v Foodora Australia Pty Ltd (Fair Work Commission, 2018) (‘Klooger’) was the first case that considered an unfair dismissal remedy under the Fair Work Act 2009 (Cth) for a platform worker where the worker was held to be an employee. The applicant delivered food for Foodora on his bicycle and he sought a remedy when his contract was terminated, with the purported notification of dismissal provided by email (at [114]). The respondent, Foodora Australia Pty Ltd (‘Foodora’) contended that the applicant's activities showed ‘all the hallmarks of an enterprise that the applicant was conducting for his financial benefit’ (at [49]). It also submitted that because the applicant was allowed to choose which tasks or jobs he would perform, he was akin to a contractor, rather than an employee (at [53]). The services contract was drafted in a way that attempted to characterise the work relationship between two parties as one of principal and an independent contractor and considerable effort had been made to avoid the applicant being characterised as an employee (at [70]), although it was found that the terms of the contract nonetheless contained many provisions that would normally be found in an employment contract (i.e. a contract of service) (at [71]).
The determination of whether the applicant was an employee or independent contractor required ‘standing back from the detailed picture, and properly viewing all of the accumulation of detail from a distance [to obtain] an informed, considered and qualitative appreciation of the whole picture’ (at [101]). The FWC held that the applicant was an employee, upon the application of the multi-factor test (at [102]). The respondent had ‘considerable capacity to control the manner in which the applicant performed work’ where its batching system was set up in such a manner that the worker was required to perform a certain number of deliveries to maintain high rating (at [73]). This was conceded to resemble a level of control that an employer would generally exercise (at [74]). Notably, the worker could not pick and choose the hours of work, where to work, or the speed at which deliveries should be made (at [74]).
The FWC acknowledged that the applicant carried out similar tasks for different businesses other than Foodora (at [75]), but this was considered equivalent to having more than one job and it did not preclude an employee from being characterised otherwise (at [76]). It was noted that the applicant did not have a substantial capital investment in the equipment used for his work and that the bicycle he used for his delivery work was also used for other activities at [78]). Furthermore, the FWC held that the applicant's services did not require a high degree of skill or training (at [78]).
Klooger provides some guidance for determining whether a platform worker is an employee or an independent contractor, as does the FWC conclusions on the limited level of control that the worker had (at [74]). However, the case can be contrasted with Amita Gupta v Portier Pacific Pty Ltd; Uber Australia Pty Ltd t/a Uber Eats (Fair Work Commission, 2020) (‘Gupta’), in which the Full Bench of the Fair Work Commission unanimously held that Gupta, a platform worker, was not an employee of Uber Eats for the purposes of the Fair Work Act 2009 (Cth). Similar to Klooger, this case dealt with an unfair dismissal remedy under the Fair Work Act 2009 (Cth). However, in this case, the applicant, Gupta, owned and drove her own vehicle.
The case was heard on appeal to the full bench of the FWC. Ross P and Hatcher VP applied the multi-factor test, while noting it was not necessary to consider other authorities. The judgment in Gupta, however, highlights the difficulties in applying the current law to platform workers: ‘the application of the test in borderline cases such as (this) … requires … evaluative judgments … and … may not produce any single clear answer’ (at [55]).
In finding that Gupta was not an employee, Ross P and Hatcher VP considered the following facts but said they were neutral as to the question of whether she was an employee or independent contractor. First, the use of one's own vehicle and mobile phone was common to both employees and independent contractors (at [65]). Second, the requirement to complete jobs satisfactorily through the ratings system could also apply to both employees and independent contractors (at [66]). Third, the mode of payment was not considered determinative of the work relationship between parties (at [67]). Ross P and Hatcher VP held that Gupta was not an employee, as the relationship lacked several of the hallmarks of an employment relationship (at [70]). Overall, the respondent had insufficient control over when and how long Gupta performed her work; she could accept work from other competitor delivery apps; and she did not represent the respondent's business.
In a minority decision, Colman DP agreed with the majority that Gupta was not an employee, but was of the view that she was not a party to services with the respondent as an independent contractor (at [81]), noting that ‘the fact that an individual is undertaking work does not mean that she must be doing so in the capacity of somebody's employee or contractor. Whether a person is either of these things depends on the evidence’ (at [82]). Colman DP did not elaborate further on this point.
Subsequent to the FWC holding that Gupta was not an employee for the purposes of the Fair Work Act 2009 (Cth) in Gupta, Ms Gupta appealed the decision to be heard before the Federal Court. The court reserved its decision, with the two parties reaching a confidential settlement out of court (Federal Court of Australia, 2020a) reported to be $400,000 (Bonyhady and Duke, 2021).
In CFMMEU [2020], although the worker was held to be an independent contractor, Allsop CJ had nevertheless noted (at [28]) that ‘the notion that Mr McCourt was an independent contractor … would defy any rational principle and common sense.’ In that case, if unconstrained by authority, it is arguable that Allsop CJ would have concluded that the worker was classified as a casual employee (High Court of Australia, 2022 HCA1 at [31]).
Other cases suggest persistent uncertainty as to whether platform workers are employees. The FWC postponed hearing the appeal in Diego Franco v Deliveroo Australia Pty Ltd (Fair Work Commission, 2021) (‘Franco’) until the High Court decided on two appeal cases on worker classification. Mr Franco was held to be an employee pending the appeal but subsequently was held to be an independent contractor by the FWCFB (Deliveroo Australia Pty Ltd v Diego Franco [2022] FWCFB 156). The High Court appeal cases may have had the effect of being a constraint on a potential finding of Franco being an employee, notwithstanding that those cases did not consider platform workers.
Given that different conclusions have been reached on similar facts, previous cases on the correct classification of platform workers provide insufficient clarity for litigants, tribunals and lower courts. The uncertainty that persists, after several decisions on the issue, may indicate that another High Court decision, by itself, may not resolve the issue of the correct classification of platform workers.
Australian legislative responses
As highlighted above, an employee who is unfairly dismissed is afforded statutory protection under Part 3-2 of the Fair Work Act 2009 (Cth), whereas independent contractors are outside the scope of Part 3–2. It is noted, however, that, depending on the facts, independent contractors have access to other types of remedies under the Fair Work Act 2009 (Cth) (e.g., via a general protections claim). In 2017, a Senate Committee considered the issue of ‘sham contracting’, whereby an employment relationship is mischaracterised as an independent contracting arrangement (Education and Employment References Committee, 2017). Although sham contracting is defined under the Fair Work Act 2009 (Cth), the legislation does not apply any specific definition or test, with legislation deferring to common law tests and interpretations of employee and independent contractor. The Fair Work Act 2009 (Cth), sets out—in ss 357 to 359—relevant defences and penalties that apply to corporations who breach the ‘sham contracting’ provisions. The Education and Employment References Committee (2017) was critical of ‘inadequacies in the Fair Work Act which … (are) inadvertently encouraged by taxation laws.’ The Education and Employment References Committee (2017) referred to ‘the absence of a definitive test at common law differentiating independent contractor from employee relationships.’ In 2020, a separate Senate Select Committee was appointed to review Job Security (Parliament of Australia, 2021).
On 4 September 2023 the Fair Work Legislation (Closing Loopholes) Bill 2023 (the Bill’) was introduced. A purpose of the Bill was to allow the Fair Work Commission to set minimum standards for some workers in the gig economy. In its original form, it included a proposal to introduce an unfair deactivation and unfair termination regime; allow operators of digital labour platforms to enter into a collective agreement with organisations representing employee-like platform workers; and expand the existing general protections in the gig economy. On 8 November 2023, the Senate passed single issue IR Bills that were split off from the Bill (Borys, 2023). Tony Burke noted that the ‘amendments guarantee that minimum standards are set in a way that is fit-for-purpose for the type of work and meets the needs of (platform) workers, and those who rely on the services’ (Marin-Guzman, 2023). On 8 February 2024, the Senate passed the Bill, by 32 votes to 29.
How to resolve the classification issue
The correct classification of platform workers is becoming of increased importance due to the growing prevalence of platform work (Ichino, 2018: 13). Nonetheless, there has been a reluctance, or perhaps, a lack of political will by policymakers and regulators to seek to resolve this issue (Forsyth, 2020b). The rapid growth of platform work in Australia has taken place in a legislative vacuum in which existing protections provided to employees have not been afforded to platform workers (Barratt et al., 2023).
In the context of greater public awareness of platform workers’ rights, or lack thereof, the issue may become a matter of normative demand. Improving the working conditions of platform workers appears to have attracted increased interest recently, and there is support for ‘increased protection for gig economy workers who are at a huge disadvantage when dealing with … delivery companies’ (The Sydney Morning Herald, 2022). A proposal to extend the legislative boundary of who is an employee by way of a ‘balanced, measured government intervention … in collaboration with stakeholders’ was recommended by Industrial Relations Victoria (2020: 188).
The overarching problem is that to the extent that platforms workers are not employees under current law, they are not entitled to the conditions that apply to employees, including access to unfair dismissal and the National Employment Standards. There is no legislative definition of employee in the Australian labour law context, and the platform worker classification issue is not necessarily resolvable by enacting a prescriptive legislative definition.
In this context, a possible solution to classifying platform workers is to deem them—via the definitions section in the Fair Work Act 2009 (Cth) (section 12)—to be employees for the purpose of the Act. A separate decision for policymakers would be whether platform workers should be considered employees for the purpose of taxation law. For the sake of consistency, it would seem logical for the employee designation to also apply in the taxation law and superannuation context. This is especially the case given that the taxation law consequences of being one or the other appear to indirectly contribute to the classification issue.
The proposal to deem platform workers to be employees can be characterised as a compromise that could be of some benefit to both platform workers and platform operators. On the one hand, platform workers would be afforded the protections that apply to employees, including having an action in unfair dismissal, under the Fair Work Act 2009 (Cth). From the other perspective, the platforms would, prima facie, not face any significant change to their business model in terms of how they operate. Under this proposal, the new status of platform workers would resemble that which applies to long-term casuals. A long-term casual is defined as an employee who has ‘been employed by the same employer on a regular and systematic basis for at least 12 months’ (Fair Work Ombudsman, Dictionary).
There is a, more specific, potential solution involving deeming, that is directed at achieving the intended outcome irrespective of whether a platform worker is considered an employee or an independent contractor. Specifically, the legislative rule that applies in relation to the superannuation guarantee, as set out in section 12(3) of the Superannuation Guarantee (Administration) Act 1992 (Cth), could be used as the criterion for determining that a platform worker is entitled to a minimum level of entitlements and protections. The specific effect of the provision is that where a person works under a contract that is mostly or entirely for labour, the person is an employee for the purposes of the superannuation guarantee. The adaptation of this legislative rule to the Australian labour law context would be that, where a contract between a platform and platform worker is mostly for labour, the worker would be eligible for a minimum level of protections and entitlements. Policymakers would ultimately determine whether the protections and entitlements included all or some of those that apply to employees.
Deeming workers to be employees, for the purpose of some provisions of the Fair Work Act 2009 (Cth), is currently being used in Australia in relation to certain textile clothing and footwear outworkers (Fair Work Act 2009 (Cth), Division 2). Deemed employees are also within the definition of ‘employee’ in the Industrial Relations Act 1996 (NSW) s 5(3). Workers can also be deemed by the FWC to be subject to a Modern Award. This deeming approach provides the worker with entitlements provided to employees. It is also noted that the Superannuation Guarantee (Administration) Act 1992 (Cth) provides for the common law meaning of employee as well as an expanded definition of the term, and section 12(3) states that ‘if a person works under a contract that is wholly or principally for the labour of the person, the person is an employee of the other party to the contract.’ In Jamsek v ZG Operations Australia Pty Ltd (No 3) (Federal Court of Australia, 2023), the fact that Jamsek and Whitby operated a partnership, was one of the reasons that they were held to be independent contractors. By contrast, it is likely that many platform workers are contracting with platforms as natural persons.
The classification of platform workers in New Zealand
This section of the article highlights a decision from New Zealand on platform worker classification - E Tū Inc v Rasier Operations (New Zealand Employment Court, 2022). Given that Australia is undergoing reform that is relevant to the classification of platform workers, the recent experience in New Zealand can be considered by Australian policy makers. This is in the context of New Zealand being a common law country that has taken a different legislative approach to distinguishing between employers and independent contractors in comparison with Australia. It is the legislative approach that allowed the New Zealand court to conclude that several Uber drivers were employees. Given that Australian policymakers are intending to address the same issue in the Australian context, the New Zealand experience here may be instructive.
The Employment Relations Act (New Zealand) (“ERA”) s 6(1)(a) defines an employee as ‘any person of any age employed by an employer to do any work for hire or reward under a contract of service…’ Per ERA s (2), in deciding whether a person is employed ‘the court or the Authority … must determine the real nature of the relationship.’ Indeed, Chief Justice Inglis described ‘the real nature of the relationship’ as ‘the key question’ in the case (at [23]).
Her Honour noted ‘the open-textured way in which s 6 is drafted (at [14])’ and stated that ‘…the question to be asked and answered is whether ERA s 6, construed purposively, was intended to apply to the relationship at issue when viewed realistically’ (at [17]). Her Honour's purposive approach to interpreting the legislation allowed ‘regard to be had to the different ways in which employment relationships manifest in practice and develop over time. If it were otherwise, an important piece of social legislation which touches on the lives of millions of New Zealanders, and which is designed to regulate labour relations within the community and protect vulnerable workers, would ossify’ (at [14]).
The matters considered relevant to the real nature of the relationship included: the nature and operation of Uber's business, its impact on the plaintiff drivers, who benefited from the work of the plaintiff drivers, and who exercised control over their work (at [25]).
Although Uber has characterised itself as a facilitator of connecting drivers with customers, her Honour held (at [57]) that being such a facilitator was not incongruent with being an employer (New Zealand Employment Court, 2022). Rather than simply connecting individuals, the platform ‘creates, dictates and manages the circumstances under which its business is carried out, and driver labour is deployed in order to grow that business.’ All of which is consistent with an employment relationship (at [52]).
It was also noted that Uber drivers are not ‘truly free’ in decisions to accept or reject work, as rejection would attract consequences, unilaterally imposed by Uber, which would limit the driver's pay (at [59]). It was held that ‘Uber exerted strict control’ and that the drivers were subject to a ‘significant degree of subordination and dependency’ (at [63]). The ‘choice’ that drivers have was compared to any employee having a choice to be absent from work and face disciplinary action as a consequence. Her Honour noted that although developments in technology may have resulted in ‘less direct, but equally effective’ ways of exerting ‘authority’ and ‘control’ over an employee, this does not render these ‘classic hallmarks of an employment relationship’ any less relevant to the ERA s 6 analysis (at [64]).
The relationship was in marked contrast to that which usually applies to an individual operating their own business (at [50]). As Uber was the party in control of marketing, pricing, and setting the terms of service provided to drivers, it is the ‘only party running a business’ (at [51]). Arguments on the ‘flexibility and choice’ that being an Uber driver affords as being indicative of a non-employment relationship (at [53]) were rejected by her Honour (at [54]), noting that flexibility is a characteristic of modern employment relationships and that casual employees, for example, have a choice about when they work, with no legal obligation to accept work they are offered.
Notably, her Honour highlighted that per ERA s 6, primacy could not be afforded to party documentation nor to the descriptor used for the relationship between the parties, as this approach would undermine the purpose of the provision (at [79]). Her Honour noted that the documentation was construed in a way that suited the defendant's business interests and this was different to the relationship's context and how it operated in practice.
The New Zealand experience serves as an example of how a clear, and principle-based, legislative definition of employee, in that jurisdiction, can provide a foundation for more certainty in the classification of platform workers. Specifically, the enactment of a statutory definition of employee in Australia, similar to that in ERA s 6. This could, in turn, provide an appropriate setting for a subsequent Australian court decision that provides greater certainty on the issue of classification of platform workers. This would overcome platforms constructing their contractual and working relationships in a manner where they can ‘sidestep employment’ (Barratt et al. 2020). A prescriptive legislative definition of employee should be eschewed, as it would likely be counterproductive to resolving the classification issue. Indeed, a focus on overly prescriptive legislative rules can result in ‘a self-perpetuating spiral of ever greater specificity and detail’ (Gribnau and Dusarduijn, 2021). In a discussion of the relative merits of principles in comparison to rules, Black (2007: 11) noted that principles ‘have the benefit of congruence: of communicating the regulatory objectives and promoting behaviour which will achieve those objectives, and of minimising the scope for “creative compliance”.’
Notably, a rule of statutory interpretation in Australia is that statutes must be construed purposively (Mason, 2016). The Bill that was introduced in September 2023 includes a measure that is aligned with our recommendation. Specifically, it refers to ‘inserting into the FW Act an interpretive principle for determining the ordinary meanings of ‘employee’ and ‘employer’ for the purposes of the FW Act. This would enhance fairness by requiring consideration of the real substance, practical reality and true nature of the relationship by reference to the totality of the relationship between the parties’ (Fair Work Legislation (Closing Loopholes) Bill 2023).
Conclusion
This article is motivated by the compelling need to clarify the classification of platform workers in Australia from the perspective of workers, platforms, policymakers, and the general community. An important consideration here is whether the objects of the Fair Work Act 2009 (Cth) are being met, including ‘providing workplace relations laws that are fair to working Australians’, per s 3(a), and taking ‘into account Australia's international labour obligations’, per s 3(b). We consider s 15AA in the Fair Work Legislation Amendment (Closing Loopholes) Bill 2023 to be a positive recent development in clarifying the classification of platform workers. Section 15AA requires that the issue of whether a worker is considered an employee is to be determined by ‘ascertaining the real substance, practical reality, and true nature of the relationship’. Furthermore, s 15AA(2) provides that ‘the totality of the relationship’ … ‘must be considered’ and that ‘regard must be had not only to the terms of the contract governing the relationship’ between the parties. Thus, s 15AA would allow the courts more scope to consider both pre and post contractual conduct, which might be considered relevant to the extent that ‘contracts of adhesion’, referred to above, are used for the engagement of platform workers. This is aligned with the recommended principle-based approach to classifying platform workers.
In light of recent developments, Australia and New Zealand appear to be on a pathway to convergence in their approaches to classifying platform workers, due to Australia adopting a similar approach to that used in ERA s 6(1)(a) in defining ‘employee’ in the Fair Work Act 2009 (Cth). This appears unsurprising, given that the platforms would be operating very similar or identical business models in both jurisdictions and would, presumably, expect the operation and effect of the law on classification of platform workers to be similar. Although it is unclear to what extent the New Zealand approach may have influenced Australian policymakers, we consider that the New Zealand provision is of importance in achieving greater certainty for platforms and workers and has given effect to a more equitable outcome than the status quo to date in Australia. Nonetheless, it may be premature to conclude that the classification issue in New Zealand is completely resolved, given that E Tu Inc is a decision of a lower court.
Clarifying the classification of platform workers is important in terms of the legislative protections that are potentially available to them. These protections include those related to unfair dismissal, and WorkCover insurance. The latter is important in the case of injury, illness, and death arising out of, or in the course of, platform work. Another anticipated benefit of legislative reform, directed at clarifying the classification issue, is its potential to reduce transaction costs for workers, platforms, and regulatory agencies.
It would appear that Australian policymakers have clearly articulated the reasons for the law change and have highlighted how it would constitute an improvement to the status quo. The legislative reform may achieve the desired policy outcome within a short timeframe, while reducing uncertainty for the parties to dispute cases related to platform work – platforms and workers.
This article has set out a case for the enactment of a principle-based definition of ‘employee’, to address the indeterminacy that has pervaded the classification of platform workers in Australia. Such a reform could have the effect of some platform workers being classified as employees. Although this would be a new development, in the Australian context, such an outcome would appear uncontroversial from the perspective of the Australian community. The resolution of the issue through legislative intervention accords with Australia better fulfilling its role under the relevant ILO Conventions to protect vulnerable workers—including the platform workers who fall within this definition.
