Employee or not employee? That is (still) the question for workers in the gig economy Open Access

The rising numbers of people working in the gig economy and nonstandard work, contributing to increased labour market dualisation have received significant attention in recent years (Baber, 2024; Palier and Thelen, 2010; Rubery et al., 2024; Todolı'-Signes, 2017). Albeit gig work is often oversimplified to describe a multitude of employment arrangements (Kalleberg and Dunn, 2016), in this article we examine drivers providing delivery services for a pizzeria outlet – engaging in the “on demand, in person” classification of gig work ^[1]. We draw on a case study of a Supreme Court ruling in Ireland in the long running case of Karshan Ltd. t/a Domino's Pizza v the Revenue Commissioners ^[2], which drew attention to the issue of employment status, finding that the drivers were employees rather than contractors. However, our analysis has relevance for all types of nonstandard work as we uncover nuanced insights into the role of state economic and social institutions in interpreting and applying employment status.

The changing nature of employment in a neo-liberal, digital era has highlighted the uncertainty regarding core features of the standard employment relationship (SER), particularly the concept of employment status (Doherty and Franca, 2020). Employment status is central to this paper, both in terms of construct definition and the distributive power it bestows. Whether individuals are considered employees working under a contract of service or independent contractors under a contract for service determines eligibility for a range of employment, welfare and social protection rights (Deakin, 2020). Employment status is also critical for determining tax treatment of earnings in contributing to the exchequer for the state (Eurofound, 2018) yet inconsistent interpretations of employment status have emerged across these spheres, particularly in liberal market economies. For over a century a series of “tests” and “indicators” have been developed through cases in the courts to help guide decision makers in determining whether a contract of employment exists (Botero et al., 2004; Deakin, 2020). These are conventionally known as the control, integration, economic reality and mutuality of obligation tests (Deakin and Morris, 2005). There are many limitations to these however and it is widely recognised that there are “grey zones” of work where individuals do not fit neatly into either category (Burchell et al., 1999; Bogg and Ford, 2021). This has raised questions of whether a different legal status and regulatory apparatus is required beyond the binary distinction between employees and independent contractors (Prassl and Risak, 2016). In some countries, “intermediate categories” grant certain workers some labour law protections, not afforded to the “self-employed”, but do not equate fully to those guaranteed to “employees” (Doherty and Franca, 2020). These include “workers” in the UK and “economically dependent self-employed” status in Spain for example, albeit it is argued that the cost of this can be to lower standards for all workers (Grimshaw et al., 2016). The binary classification remains a feature in many contexts including Ireland yet there is no consistent definition of employed or self-employed in Irish or EU law.

Issues of employment status predate the contemporary “gig economy” and have long been tested through the courts in many countries, with varying outcomes (Burchell et al., 1999; Fudge et al., 2003); however, the rise in such work has heightened attention to these issues. Over the last number of years, several “landmark” cases globally involving companies such as Deliveroo, Uber and TaskRabbit have seemingly provided solutions to employment status (McGaughey, 2019; Prassl and Risak, 2016; Todolí-Signes, 2018). However, the impact of such cases has been constrained often resolving some issues yet leaving many questions unanswered. Invariably one of the reasons for this non-resolution is due to the different roles of state institutions in interpreting and applying employment status. A key limitation of much of the existing analysis of judicial rulings is that they do not capture the full complexity of employment status beyond the courtroom (Wu and Wang, 2025). Thus, while the legal institutions, instruments, interpretations and outcomes of these cases have been widely examined, the procedural tools and powers employed by other state institutions have been less explored. The role of institutions has long been studied by industrial relations (IR) scholars (Dunlop, 1958; Fox, 1985; Durkheim, 1984), however the integration of economic and social issues within those institutions has been underexamined (Jackson and Muellenborn, 2012). In this article we set out to rectify these gaps by analysing the role of state institutions, beyond the courts, in interpreting and applying employment status. Our findings show a lack of integration between the economic and social institutions of the state, concurrent with the demise of the SER, leads to an imbalance in worker outcomes and contributions thus furthering labour market dualisation. We find that a worker can simultaneously hold a different employment status across different spheres thus occupying both insider and outsider positions in the labour market.

The remainder of this article is structured as follows. First, we outline the centrality of employment status for workers in the welfare state. We then set out the role of different Irish state institutions in determining and/or interpreting employment status. Next, we briefly outline our methodology. We outline the Karshan case and explore the responses of these institutions to the Supreme Court judgement. We discuss the impact of the judgement illustrating the inconsistencies revealed among these state institutions. We contribute to ongoing debate on precarisation arguing that such inconsistencies contribute to the unresolved problem of employment status, maintaining a fragmented and two-tiered concept of what constitutes employment and discuss the implications for labour market dualisation. We conclude by discussing avenues for future research.

The integration of economic and social spheres has long been central to the welfare state. In this paper we draw on Garland's (2014, p. 330) definition of the welfare state as “a distinctive form of governmentality: a specific mode of constituting the economy, assuring social security, and guaranteeing social provision”. Garland (2014, p. 328) redresses modern interpretations of the welfare state as either being centred outside the labour market or “a thing of the past”. Instead, the welfare state is conceptualised as a fundamental dimension of modern society and integral to the economic functioning and social health of industrialised capitalist societies. Barr (2001) too argues that the welfare state is essential and enduring while Le Grand (1982) notes that the chief beneficiaries of the welfare state are not the poor but rather the middle classes and those in employment as well as employers and corporations. Hemerijck and Huguenot-Noël (2022) argue that rather than being viewed as a “burden on the taxpayer”, the welfare state should instead be understood in normative terms as productive social investment.

Our focus in this paper is on the links between the welfare state and the SER, the emergence of which are inextricably linked. As Supiot (1999, p. 8) argues, “the working relationship became the site on which a fundamental trade-off between economic dependence and social protection took place. While it was, of course, the case that the employee was subjected to the power of another, it was understood that, in return, there was a guarantee of the basic conditions for participation in society”. In early welfare states, social rights were predominantly derived from formal employment relations, and protection from market forces was primarily achieved through the establishment of the SER characterised as a “socially protected, dependent, full-time job” (Frericks et al., 2014).

This progress was in part supported by the “contractualisation” of the employment relationship, with the terms “contract of employment” and “employee” becoming commonplace. Deakin (2002) argues that contractualisation had two important implications for the employment relationship. Firstly, by placing limits on the employer's legal powers of command. Secondly, supporting the use of the employment relationship as a “vehicle for channelling and redistributing social and economic risks, through the imposition on employers of obligations of revenue collection, and compensation for interruptions to earnings” (Deakin, 2002, p. 193). Recent research has called for greater attention to the uneven distribution of risks and rewards within gig work (Kougiannou and Mendonça, 2025) and the establishment of employment status is at the crux of this.

The state plays a key role in navigating the complexities of employment status. In understanding the role of the state, it is important to engage in nuanced analysis of the various spaces and actors involved. There is a need to separate out levels of analysis between “the state” as a unitary abstraction and the state as a collection of institutions organised around various projects and potentially competing agendas (Martínez Lucio and Mac Kenzie, 2017, p. 2999). These include socially embedded institutions such as parliament, the judiciary, dispute resolution institutions, social protection and tax agencies, the roles of which we examine in this paper. The advent of the gig economy poses fundamental challenges for the institutions of the welfare state. State policies of deregulation have been a driver of precarious work, and the effectiveness of labour law depends on coverage, scope, enforcement as well as interaction with other labour market institutions (Prosser, 2016). Gig workers are disadvantaged by an absence of employer contributions to tax and social insurance protection – a feature that also disadvantages the state in terms of reduced tax intake (Moran, 2009). Indeed, most of the assumptions of the welfare state no longer hold good (Garland, 2014) and thus there is a need for welfare states to adapt to changes in social and economic life, including changes in types of employment. Despite criticisms regarding its methodology, placement of certain countries and omission of factors like gender, Epsing Andersen's typology of welfare states remains fundamentally important, particularly in analysis of precarious work (Kim et al., 2012). Within this typology Ireland was classified as a liberal welfare state (Esping-Andersen, 1990, pp. 35–78) where benefits such as health insurance and pensions, are linked with employment, and means testing is used to determine eligibility for state services. However, as Cousins (1997) suggests, the Irish welfare state is a moveable feast owing to historical connections with the UK and contemporary globalisation. McLaughlin (2001) notes the Irish regime has developed into a hybrid of liberal welfare and corporatism. Ireland, alongside many other states has, since the early 1980s, allowed for deregulation of the labour market to increase competitiveness within the global arena. According to McLaughlin (2005) Ireland has both moved towards the European model of social pacts in terms of process but also widened the gap with the European model in relation to social outcomes. While Irish social partnership did not share all the features of social democratic European corporatist models, it was also substantively different from the neoliberalist approaches evident in the UK (McLaughlin and Wright, 2018). Albeit individual employment rights in Ireland have developed substantially through adoption of EU regulation, in terms of the centrality of the employment relationship, the Irish approach shares significant parallels with the UK in determining employment status and by extension welfare rights and entitlements. Therefore, while employment rights have expanded (led from the EU), the ability access those rights has remained largely state determined. Indeed, this has been a critical concern and one of the reasons for delays in the agreement of the Directive on Platform Work by the EU (Prendergast, 2024).

Deakin (2020) notes the stability of the tax base and the capacity of the state to deliver public goods is also dependent on employment classification. Self-employed workers tend to be paid less than employees and avoidance of paying tax is also potentially greater, which leads to a revenue drain for the state (Dhorajiwala, 2023) and puts pressure on governments to reclassify these as employees (Fudge, 2017). In many countries there has been recognition of the impact of bogus self-employment both for workers and the state (Thörnqvist, 2015). In Ireland, a report from the Oireachtas (Parliament) Committee in 2021 made several recommendations to address this issue (Houses of the Oireachtas, 2021). These included public information campaigns, the establishment of a dedicated employment status unit to provide determinations on cases and the development of a standard definition of the word “employee” and “worker” to apply to all pieces of employment legislation. This report highlighted the numerous challenges that confront the state regarding employment status classification including ensuring employment rights and social protection for workers, compliance by employers and securing correct levels of tax revenue to the state. The issues raised draw attention to the role of the state in shaping the labour market through its' institutions as a legislator, policymaker, revenue collector, coordinator of welfare supports as well as its' dispute resolution institutions.

While the role of the state as policymaker has been subject to significant debate, the role of individual state bodies has remained underexamined. The way in which these institutions interpret employment status is critical to the functioning of the welfare state but with significant potential for increasing dualisation within the labour market. Labour market dualisation theory refers to the increasing division within labour markets, characterised by a distinction between “insiders” and “outsiders”. Insiders are typically those in stable, well-protected jobs, often with strong employment rights, while outsiders are those in precarious, less secure employment situations. This theory has gained prominence in the context of labour market reforms across various countries, particularly in Europe, where economic pressures have led to a growing divide in job security and benefits (Berglund et al., 2020; Ciarini and Rizza, 2021; Porte and Emmenegger, 2017).

The interpretation of employment status shapes whether workers are “insiders” safeguarded by employment legislation and social protection (Hipp et al., 2015). Meanwhile outsiders categorised as self-employed (although that does not reflect the reality of their work situation) find themselves in “limbo” with limited access to protections from the welfare state while bearing a disproportionate burden of social and economic risks in the absence of an employer (see Table 1). The mechanisms through which workers are positioned as either insiders or outsiders include the legislative powers and procedural tools used to determine employment status by employment, welfare and tax institutions of the state. Such determinations influence worker contributions and outcomes, as examined in our empirical analysis. Palier and Thelen (2010) argue that increasing labour market dualism is explicitly supported by state policy and welfare state reforms and in distributional terms, social protection systems can exacerbate rather than mitigate labour market divides. Similarly, Emmenegger (2012) argues that the widening divide between labour market insiders and outsiders is not inevitable but rather the result of political choices and institutional design. Hipp et al. (2015, p. 365) call for further investigation into the complex interplay of institutions including the interplay of “a couple of wisely chosen” institutions for “subsets of workers”. This is the approach we follow in this paper where our analysis focuses on the interplay of state institutions in determining employment status through our case study of Karshan delivery drivers.

There is no single statutory definition of employee and so the institutions of the state are first faced with determining employment status led by the decisions of the judiciary, for example precedents set by the higher courts. The Workplace Relations Commission (WRC) is the independent statutory dispute resolution body with a quasi-judicial role and responsibility for adjudicating on employment rights in the first instance. The Labour Court is the sole appellate body for employment law cases from the WRC adjudication service. Established under the Industrial Relations Act 1946 and amended by the Workplace Relations Act 2015, the Labour Court is not part of the civil courts system, however its judgements are legally binding (Wallace et al., 2020). In the realm of tax and welfare, the institutions of the state are also tasked with determining employment status. The Revenue Commissioners key role is the collection of taxes and duties. In categorising workers, Revenue makes decisions about whether a person is self-employed or an employee for tax purposes and has responsibility for the collection of Pay Related Social Insurance (PRSI) for the Department of Social Protection (DSP). PRSI contributions determine eligibility for a range of state welfare supports. The Scope section of the DSP makes decisions and provides information and advice on PRSI classifications and contributions. Decisions of Revenue can be appealed to the Tax Appeals Commission, an independent statutory body which adjudicates on tax disputes in accordance with finance and tax legislation. Table 2 summarises the key state institutions and their role in determining the employment status of workers.

In 2001, an Employment Status Group was set up to develop a Code of Practice on determining employment status due to concerns regarding misclassification of employees. Such concerns have continued, and the code was updated in 2007, 2019 and 2021 with input from the DSP, Revenue Commissioners and the WRC. The code is “intended to be a living document” which will continue to be “updated to reflect future, relevant changes in the labour market, relevant legislation and caselaw” (DSP et al., 2021, p. 4). The code is “aimed at investigators, decision-makers and adjudicators in the DSP, the Office of the Revenue Commissioners, the WRC, their respective appeals bodies and the courts”. Thus, the institutions of the state recognise the importance of having clarity and coherence on employment status yet, for the most part, they operate independently of each other. It is explicitly noted within the code that the decisions of the different institutions “are not binding on each other” (DSP et al., 2021, p. 11). Given the erosion of the SER and the complexities of establishing employment status, there is considerable potential for inconsistencies among these institutions in determining the entitlements of nonstandard workers including gig workers. Such workers are often vulnerable, have little power resources and are arguably most in need of protection from market forces and a welfare safety net to enable them to fully participate in society. It is notable that the Code of Practice has not been placed on a statutory footing despite indications from the DSP that this was due to be done in 2019 and recommendations from the committee in 2021 that this must be achieved by the end of 2021 (Houses of the Oireachtas, 2021). In fact, the committee explicitly stated its' concerns in relation to platform workers noting that current employment law:

While the institutions of the state essentially operate independently, the civil courts hear appeals across both realms of regulation – employment rights and tax – and it is within the courts where issues in relation to employment status converge. The complexities of different work arrangements mean that the employment status is inexorably decided on a case-by-case basis and interpretation of key tests and the application of precedents to individual cases within institutions is critical. As the highest judicial authority in Ireland, Supreme Court decisions exert significant influence on all other institutions in this regard. In granting leave to hear the Karshan case the Supreme Court noted it “had not yet had an opportunity to clarify the law in this area, specifically in relation to the gig economy and workers engaged under contracts similar to those used by Karshan”. To hear an appeal of a Court of Appeal judgement (as in Karshan), the Supreme Court must be satisfied that the decision involves a matter of general public importance or that the interests of justice necessitate such a further appeal. In Karshan the Supreme Court found that “a number of issues of general public importance do arise for appeal” and that these issues “undoubtedly go beyond a discrete issue of tax payments … particularly due to the apparent existence of similar umbrella contractual relationships in Ireland” (see Prendergast, 2022). Thus, when the issue of employment status was before the Supreme Court in the Karshan case the outcome was keenly awaited.

In this article we adopt a case study approach to analyse the role of state institutions (and the attendant procedural tools at their disposal) in interpreting and applying employment status. Case study research allows the researcher to look at a phenomenon in depth and in context (Farquhar, 2012; Yin, 2014). Furthermore, Argyrou (2017) suggests that legal case studies can offer rich contextual insights that reveal the complex interactions between law and society to uncover the real-world impact of legal rules and practices. From an employee perspective this is important as through legal case studies, the shifting balance of power among IR actors becomes more visible and open to analysis. Our analytical approach is through means of document analysis which is “a systematic procedure for reviewing or evaluating documents – both printed and electronic” (Bowen, 2009, p. 27). It is an approach used in qualitative research that enables the researcher to examine and interpret data to elicit meaning, gain understanding and develop empirical knowledge (Corbin and Strauss, 2008). It is appropriate for our case study analysis as it enables an overview of interpretation of employment status by the judiciary but also deeper insights into the role of other institutions that shape whether individuals are insiders or outsiders.

In selecting documentary sources, we were guided by the intended purpose of the document as well as its source. We selected documents that offered insights into the determination of employment status from state institutions that affect labour market dualisation. Our key sources were publicly available documents produced by state institutions and other relevant actors. Our primary document was the text of the Karshan ruling which offered nuanced insights into the legal issues of employment status including from relevant precedent cases. This was supplemented by documents including codes of practice, guidelines and public notices in relation to the implications of the Karshan case published by institutions identified in Table 2. These institutions were identified as the most relevant economic and social institutions of the state in relation to determining employment status. We also examined responses to the Karshan ruling published by key IR actors including trade union and employer organisations. We followed an inductive approach to analyse each source. First, we read through the entire document to establish key meaning. Next, guided by the literature review and the key elements outlined in Table 1, we reviewed each source to identify meaningful and relevant passages of text. We then reviewed these key passages of text to establish insights to our research question – how do state institutions' interpretation and application of employment status impact worker outcomes? Our analysis drew attention to the rules and procedural tools utilised by these institutions to determine employment status and the purposeful and explicit separation of decisions between them. Finally, we organised insights provided by our analysis as illustrated in Figure 1.

In October 2023, the Supreme Court addressed the perennial complexities of employment status in its judgement on the long-running Karshan v's Revenue Commissioners case. Albeit the case centred on tax categorisation of workers rather than employment law per se, the ruling was widely lauded as an important step forward in securing employment rights (Prendergast, 2023; O'Donovan, 2023) and seen as critically important for gig workers in Ireland (Doherty, 2024). The country's largest trade union the Services Industrial Professional Technical Union (SIPTU) described the judgement as providing “ammunition” to thousands of workers to claim employment rights albeit it was recognised that individual cases would still need to be argued at the WRC (Malone, 2023).

The crux of the case concerned the employment status of pizza delivery drivers working for Karshan (Domino's Pizza) to determine the taxation of the drivers' income under the Taxes Consolidation Act, 1997. While Karshan argued that the drivers were independent contractors, the Revenue Commissioners argued they were employees and thus taxable for income tax and national social insurance purposes. This was despite an overarching written agreement which Karshan required drivers to sign confirming that they understood all delivery work undertaken was “strictly as an independent contractor” and a requirement to acknowledge that Karshan had “no responsibility or liability whatsoever for deducting and/or paying PRSI or tax on any monies [they] may receive under this agreement”. However, as demonstrated in the Autoclenz ^[3] case in the UK Supreme Court, the written agreement is only a part to consider and the “true agreement will often have to be gleaned from all the circumstances of the case”. Other factors may be taken into consideration such as the reality of the relationship and the differences in bargaining power between the parties to an employment contract. The Karshan case had first been decided by the Tax Appeals Commission who determined that the drivers were employees and on appeal, the High Court agreed. However, the Court of Appeal determined that the drivers were independent contractors, primarily basing that decision on the fact that there was no ongoing “mutuality of obligation”. The test of whether there is an obligation for an employer to provide work and a worker to perform work has long been a stumbling block for workers engaged in nonstandard or casual arrangements given the intermittent nature of their work (MacMahon, 2019).

The Karshan judgement provided an in-depth analysis of previous jurisprudence and legislative history dating back to UK laws such as the Master and Servant Acts and Employers and Workmen Acts. It explored in detail the development of tests for determining employment status and provided important clarity on this in its' conclusions. The concept of “mutuality of obligation” took central importance in the case. A core question was whether this requirement imposes a sine qua non of the employment relationship, an ongoing obligation of some kind. However, J. Murray noted that the concept of mutuality of obligation was a “wholly ambiguous label” that was “overused and under analysed” in employment law. In a critical development for assessing the employment status of individuals, J. Murray suggested that use of the phrase be discontinued noting that while it was clearly relevant to questions of employment status, it was not a sine quo non. The question of whether a contract is one “of service” or “for service” should be resolved, it was noted, by reference to five key questions as outlined in Table 3.

These questions derived from a “reduction of the existing case law” and “the desirability of avoiding confusion in the future as to the need for mutuality of obligation” (2023, IESC 24).

In the aftermath of the Karshan judgement, Revenue developed updated guidelines for determining employment status for tax purposes making extensive reference to the judgement throughout (Revenue, 2024). In essence, the new guidelines incorporated the five key questions as outlined in Table 3. Importantly, it was explicitly emphasised throughout that these guidelines related only to employment status for tax purposes. The implications were evident – “It is clear that there are a number of workers across a number of sectors who will need to be treated as employees for tax purposes, where previously they have been treated as self-employed. For those re-classified as employees for tax purposes, the business will have an obligation to operate PAYE” (Revenue, 2024, p. 25).

Extensive reference has also been made to the Karshan ruling in employment rights cases before the WRC where employment status needs to be determined. The institutions of the state – DSP, WRC and Revenue once again came together to update the Code of Practice of Determining Employment Status (DSP et al., 2024). Crucially however, this remains separate to Revenue's own guidelines on determining employment status for tax purposes. While the Code of Practice and Revenue's guidelines set out an identical purpose - to outline the implications of the Karshan Supreme Court judgement in determining employment status, a disjointed approach remains. In fact, Revenue's guidelines overtly state “it cannot be assumed that the decision of one State body will be replicated by either or both of the other two” (Revenue, 2024, p. 4). Figure 1 illustrates the fragmentation that exists in the procedure for how each body determines employment status. The distinct fragmentation of what constitutes employment is echoed on the citizens information website which explicitly states, “you could be for example, an employee for tax purposes and self-employed in respect of your employment rights and social welfare entitlements ^[4]”.

As Doherty (2024, p. 18) notes “the higher courts dealing with point of law appeals, frequently refer to tax/social insurance and employment rights jurisprudence interchangeably”. Yet the other institutions of the state, while on the one hand working together to provide guidance and clarity on employment status (through the code of practice) continue to operate independently and unequivocally maintain a fragmented and two-tiered concept of employment for tax purposes and for social and employment rights. Dhorajiwala (2023, p. 768) notes that “whether employment rights accrue to a particular work relationship, or an individual is assessed as owing tax, the central question is whether the nature of their relationship vis-à-vis another entity is one of employment”.

Ostensibly, this case illustrates the divide between tax and employment legislation. We contribute to previous analysis by Dhorajiwala (2023) in the UK context highlighting the capacity of legal institutions to endorse divergent concepts of employment. Indeed, there is extensive literature on gig work focusing on legal analysis and efforts to resolve inconsistencies in how employment is conceptualised – the contract of employment and attendant tests for employment status (Deakin, 2020); the categorisation of different types of gig work (Duggan et al., 2020); the adaptation of employment status to reflect different worker categories (Todolı'-Signes, 2017) as well as platforms as labour market intermediaries (employers) that are or should be party to the employment contract (Baber, 2024). This legal focus is rational and important given that the employment contract is a legal instrument central to much economic and social life. However, it also means that the rule-making instruments of other state institutions, that may contribute to labour market dualisation has been neglected. Our analysis shows the contributions of such institutions, beyond the courts in maintaining inconsistent and fragmented interpretations of what constitutes employment. In doing so, we contribute to previous calls for research to examine the complex interplay of state institutions (Hipp et al., 2015) as well as that calling for analysis beyond the courtroom (Wu and Wang, 2025). We now proceed to examine how the mechanisms of dualisation are illustrated through this case.

The Karshan case had the potential to be a watershed judgement for clarifying what is at the very core of the welfare state – employment status. It has been presented as such by employer, trade union and government groups and has undoubtedly provided welcome guidance in relation to dismantling the requirements for mutuality of obligation which had long proved a barrier for nonstandard and gig workers. However, the reactions of state institutions in the aftermath of the case give less cause for optimism in relation to securing employment rights and social protection for workers. Despite calls from the Oireachtas Committee report (2021) to put the code of practice on a statutory footing and despite members of the Irish Congress of Trade Unions (ICTU) on the employment status working group arguing for the designation of workers as employees unless proven otherwise (Higgins, 2022), there has been little change to the state institutions individualised approach. There has arguably been a deliberate effort to keep a separate and fragmented approach to determining employment status for tax and employment rights and welfare purposes rather than developing a coherent and consistent approach. The way in which each institution explicitly limits their guidelines on interpreting employment status to their own specific context illustrates this. The result is a potentially more ambiguous and opaque translation of employment status, leaving workers to self-navigate their way through understanding the application of both employment, social protection and taxation rights. In the aftermath of Karshan, workers may expect their status to be consistently applied across each of these spheres, however, these remain misaligned.

This misalignment is illustrated in Figure 1. It shows how the state, on the one hand emphasises cooperation and coordination of each body with responsibility for determining employment status, but at the same time elevates the powers of Revenue in allowing them to have separate guidelines for determination. Indeed, the state also did not grasp the opportunity for radical financial reform of tax and welfare systems to align the treatment for everyone who provides labour as has been called for in other jurisdictions (Freedman, 2020). The lack of cohesive agreement among different institutions on what constitutes employment status remains, adding layers of complexity to its' definitional uncertainty. The institutional actions demonstrate how the legislative powers and procedural tools used by state social and economic institutions maintain a divergence in employment status and contribute to labour market dualisation. Furthermore, they entrench labour market dualisation by explicitly providing that a worker can simultaneously be an employee and an independent contractor across different realms of employment, tax and welfare.

A second key contribution of our paper relates to the welfare state. We add to Garland's (2014, p. 349) assertion that a welfare state's capacity to adapt is only as great as the capacity of the nation's political institutions to manage change effectively. The welfare state emerged during a specific socio-economic context and was rooted in assumptions centred on the contractualisation of employment and the SER (Frericks et al., 2014). However, decades of free-market policies, deregulation and privatisation have fragmented the SER both in definitional and distributional terms challenging this core of the welfare state. This does not mean the collapse of the welfare state. On the contrary, the welfare state is a necessity in modern societies albeit the relations between the capitalist economy and the states' social protections are both combined and contradictory (Streeck, 2011). Our analysis shows that the state institutions' capacity (or willingness) to adapt to social and economic change is limited. The Karshan case provided an opportunity to redress the growing difficulties for nonstandard workers of establishing employment status, which to an extent, it did. However, it stopped short of introducing a new approach that would better capture the realities of individual's working relationships in changing economic and social contexts. In fact, the judgement is explicitly cautious on this noting “if new rules that fundamentally depart from the established general law of contract are to be suggested for the construction and application of written agreements governing the exchange of labour, these need to be rigorously justified, and precisely defined”. To date precise definitions in relation to employment status have proven somewhat elusive in the Irish regulatory space.

It is not only workers impacted by the ambiguities of employment status. Reduced tax liability through avoiding employment status also serves to reduce tax income to the exchequer. While the extent of such costs is difficult to quantify and is largely unknown, ICTU (2015) estimated that bogus self-employment had cost the exchequer €600 million in recent years. Freedman (2020, p. 194) has called for radical reform of the tax and social insurance systems to “align the treatment of all those who supply labour, whatever legal arrangements they happen to use” as this would increase tax equity and remove the incentive for businesses to resist employment status for tax purposes. From a legal perspective, Lasek-Markey (2024) points out, that while the outcome of Karshan did not endorse the creation of a third category of worker (beyond binary employee/self-employed), it did effectively enable it by finding that the same working arrangements may be treated differently for different purposes. As O'Sullivan et al. (2020) note, while the state exerts influence on labour market arrangements through the welfare system, studies on work and employment have been criticised for their lack of attention to welfare. Here, we further that argument in illustrating how the tax system and its attendant bodies too are often ignored in studies relating to the regulation of work and employment.

Although there have always been divisions within the labour market, it has been recognised that work has become increasingly insecure for many individuals and inequalities in unemployment risks, wages and social protection between labour market insiders and outsiders have considerably grown in recent decades (Hipp et al., 2015). Our third key contribution relates to dualisation theory, illustrated in Figure 2 which presents a model showing the interconnectivity of labour market inputs, state institutions and worker outcomes, in determining both economic contribution and social protection.

At its core, the figure illustrates a cyclical relationship whereby workers contribute to the system through taxation and participation in the economy, while state institutions provide protections, rights and benefits in return. For workers the outcomes of this process should deliver income security and social stability. The model emphasises how this exchange is not ad hoc but is governed by legislative powers and procedural tools (captured centrally in Figure 2), which act as a framework ensuring fairness, enforcement and accountability. Rather than showing these elements in isolation, the figure highlights their interdependence. Workers' contributions sustain the system, institutions safeguard rights and resolve disputes and legislation provides the authority to regulate and enforce standards. As our model illustrates the application of legislative powers and procedural machinery by different institutions mediates the impact of protective rights for workers. Arguably, the SER has been used as a mechanism to ensure a cycle where economic participation and social protection reinforce one another, creating stability both for individuals and for the wider state. However, the determination of employment status through differing procedural machinery by different institutions breaks this cyclical and interconnected link.

The dualisation theory has long highlighted the role of political and institutional factors in shaping labour market outcomes with “insiders” protected at the expense of “outsiders”. Much research has focused on addressing these protective gaps and/or re-balancing inputs and outcomes (Grimshaw et al., 2016). Palier and Thelen (2010) highlighted a core problem in France and Germany's welfare reforms was the mismatch between the growing number of workers not contributing into the system in proportion to what they would likely be drawing out of the system, which was explicitly supported by state policy. The Karshan case demonstrates the complexity of addressing the balance of inputs and outputs cannot solely be resolved through worker classification.

Our analysis highlights the fact that rather than the redistribution of economic and social risk as envisioned by the welfare state, fragmented employment combined with inconsistencies in employment status means that workers can bear the costs of contributing to the exchequer with little benefit in return in terms of welfare protection and employment rights. Furthermore, any suggestions of using an approach that gives an assumption of employee status unless proven otherwise have been met with strong opposition – both in Ireland and in most of the wider EU (Higgins, 2022). While literature has focused on how state policies extend dualisation, it has to date to our knowledge conceptualised “insiders” and “outsiders” as distinct groups of workers (albeit Hipp et al., 2015 emphasise the considerable heterogeneity within these groups). However, what our analysis shows is that a single individual worker can simultaneously be an insider contributing into the system but an outsider unable to draw out of the system, institutionally entrenched by the procedural tools and powers of the state (see Figure 2).

While the Karshan ruling was widely welcomed as a step forward for gig workers, the issue of employment status remains far from resolved. The transposition of the EU Directive on Platform Work will further shape the state's approach but for now there remains no singular route to determining employment status. The continued separation of individuals' employment status for tax purposes from employment rights and social protection disrupts the balance and reciprocity inherent in the welfare state and institutionally extends and entrenches labour market dualisation. From a policy perspective there is much scope for reform to address the issues raised. Greater co-ordination of employment status among state institutions is critical. In this context, a singular code of practice, placed on a statutory footing whereby decisions made in the first instance are binding across the relevant institutions would address dual employment status. An assumption of employee status, as has been adopted in Spain (Todolí-Signes, 2021) would also provide more consistency albeit this has to date been met with strong opposition (Higgins, 2022).

Our study is empirically limited in its reliance on secondary data and a single case study focused on delivery drivers in one organisation. Extending the methodology to include primary interviews with the stakeholders involved in the case would shed light on the rationale and decision-making structures within different institutions. Our theoretical model offers scope for future research to examine the impact of the procedural tools and powers of state institutions in other contexts. This includes not only other categories of gig workers and country contexts but other avenues of work research where the decisions of economic and social institutions of the state are intertwined. This could include, for example research on employment permits and pension entitlements. Future research could extend the proposed model through comparative cross-national analyses of tax and welfare coordination mechanisms and administrative discretion in classification decisions, particularly in other liberal market economies. This would encourage a renewed emphasis on examining procedural machinery and rules in broadening or limiting the powers of state institutions.