REGULATING PLATFORM WORK: REFLECTIONS ON THE PROPOSED EU DIRECTIVE

Algorithmic management as a kind of a new organizational system and technology at work brings several challenges into law, especially labour law – from processing of personal data, assignment of work and assessment of the work results to the safety and protection at work. Proper assessment of risks of algorithmic management by the platform provider is therefore a truly necessary step before its introduction in order to avoid the negative effects of breaching the law.

In this context in December 2021, the EU submitted a proposal for a directive on improving working conditions in platform work (“Directive proposal”) that sets several duties to the digital labour platforms (providers) concerning algorithmic management, i.e. automated monitoring and decision-making systems. According to the Directive proposal the platforms shall provide information on working of algorithmic management, consult its aspects with trade unions or platform workers. Also a duty to ensure a human review of significant algorithmic decisions should bring a human aspect to an otherwise machine-like environment. The question is if these duties in the Directive proposal are sufficient or rather lenient. Either way, there is still a lot of work to be done by the national legislation of each EU Member State, their courts and administrative authorities.

Our research and poster is based on case-studies using comparative-analytical methods. The case-studies stem from judgments, decisions of the administrative authorities (and other sources describing working of a particular algorithmic management used by the platform) on this topic, especially from the EU and the USA, to find out acceptable settings of algorithmic management and formulate basic principles that should be respected. In particular, the paper will devote close attention to the discriminatory algorithmic settings or settings that can violate the principles of equal treatment.

The discrimination and unequal treatment can manifest itself in various forms within the algorithmic management of platforms. From initial selection of platform workers, work allocation or access to some types of assignments, the amount of remuneration to the type of contract that is concluded between platform provider and the platform worker. The fact that some platform workers carry out activities as self-employed persons without legal protection and entitlements and others as employees may itself be discriminatory or unequal, e.g. a platform chooses one business model in some countries but different in others, or it disadvantages minorities. From a technical point of view, however, the fundamental question is whether it is possible for the system to encompass and avoid all possible forms of discrimination, unequal treatment and, strictly speaking, also bias and prejudice usually held by people.

Research Context and Objective

In this study, we examine why AI-based algorithms frequently fail to fulfil the intended tasks coherently. Artificial intelligence (AI) has become a promising field of innovation, representing broadly “a system’s ability to interpret external data correctly, to learn from such data, and to use those learnings to achieve specific goals and tasks through flexible adaptation” [1]. AI includes machine learning (ML), defined as using large datasets by machines to solve various problems [2]. ML relies on algorithms, that is, a formal set of logical rules to process data.

The rapid adoption of AI applications in many social and economic domains resides in their potential strength to efficiently recognize patterns in large datasets and predict future trends and trajectories [3]. However, meanwhile AI-based applications increasingly pervade businesses, personal, and social lives., repeated incidents of algorithmic failure continue to manifest. For instance, “GPS provides faulty directions, or biometric systems misrecognize people” [4]. Moreover, the increasing power and pervasiveness of AI-based algorithms risk amplifying the frequency and severity of failure occurrence [4]. This evolution of AI could lead to considerable implications for business and the wider society [5, 6].

AI, ML, and algorithms originate in the computer science discipline, primarily addressing algorithmic failure as a technical topic. Most computer science contributions examining AI failure focus on failure detection mechanisms and technical approaches to improve algorithms. Exemplary, [7] classify failures of AI applications for autonomous weapons, employment automation, control failures, and self-replicating machines. Technical approaches to AI failures are algorithm-centric, assuming that improved algorithms and technical mechanisms provide solutions and safeguard against future failure occurrences. However, research shows that the success or failure of technology is not merely a technical issue but has a broader management angle [8].

Against this background, we investigate the failures of AI-based algorithms from a management perspective. Relatedly, [9] argues that technology projects fail as they do not correspond to the problems they were supposed to solve or the needs they were supposed to fulfil. Scholars commonly represent AI, ML, as the next step of technology evolution, enabling AI-based algorithms to learn from data autonomously [10]. In management terms, it is crucial to understand if AI failures occur due to an incompatibility between given problems and AI- based algorithms. To do so, we need to view algorithmic failure in its entirety, including managerial problems and their situating context. The managerial perspective helps develop a better understanding of non-technical aspects of AI failure and, in turn, positions AI-based algorithms more effectively within an organizational technology architecture [11].

Consequently, we leverage in our study problem-solving literature to discern common problems for which AI-based algorithms are deployed. This theoretical lens helps to understand the distinct dimensions of based algorithms for problem-solving, such as their formulation, situating context, and temporal characteristics. In parallel, we analyze constituent features of algorithmic problem-solving capabilities, such as how their inherent rules have been formulated and how they seek to solve problems. We specifically focus on identifying compatibilities and incompatibilities between managerial problems and algorithmic problem- solving approaches.

We bring perspectives from intelligence literature to examine if the algorithm exhibits the intelligence type specific to the problem. In so doing, we develop a well-grounded understanding of why algorithms fail. It is increasingly recognized that algorithms, being a set of rules, are well-suited for formally well-defined problems and seek optimal solutions [12]. However, despite the prevalent use of algorithms for well-defined problems, failures are common. We, through our study, highlight that problems’ formulation, situating context, and temporal nature can make them dynamic and evolutionary, thus leading to the need for distinct intelligence types relative to the ones exhibited by algorithmic capabilities, thereby leading to failures.

We aim to make several potential contributions to research and practice. First, we bring a managerial perspective to the algorithm failure literature by highlighting that problem formulation, often discussed in management, plays a central role in frequent failures of AI- based algorithms. We also contribute to problem-solving literature by conceptualizing algorithmic failures as a context and highlighting specific problem characteristics that are not relevant for traditional problems but become relevant due to AI emergence. Second, we highlight that algorithm failures cannot be effectively managed if treated as mere technical problems. Awareness of problem characteristics and incompatibilities with algorithm capabilities can help develop management practices to avoid failures. Third, our study highlights the non-technical aspects that need to be considered for designing better AI-based algorithms for practice.

We will use secondary data sources to compile documented AI-based algorithmic failures. We will follow two-step methods to analyze data. First, we will use topic-modelling techniques to identify failures’ themes. Second, we will code all information regarding problem dimensions, algorithm capabilities, and intelligence required for specific problems and intelligence exhibited by specific algorithms. Finally, we will synthesize insights from the previous two steps to develop a framework for algorithm failures and offer detailed propositions for research and practice.

When you try to build an IKEA cabinet, how often are you bothered by its instruction manual? While holding the wooden plank with one hand, screwdrivers in the other, you still have to find a way to flip the pages of the paper-based manual, only to discover that you have absolutely no idea which steps to follow. No matter how experienced you are with the work, the inconvenience affiliated with the paper-based instruction and the limited information provided by this type of form constrict the way we perform tasks.

Now consider the industrial context. Frontline workers, such as maintenance inspectors, face the same issue in their daily work. Workers are usually in the field with their hands full of tools, communication devices, and paper-based instruction manuals, but they still need their hands to be available for physical tasks (Coon, 2018). In addition to the inconvenience of carrying paper-based manuals, workers have to divert their attention from the task to read and reread the instructional steps, which disrupts their workflow. Moreover, while the reality is three-dimensional, the information provided by paper-based instructions is limited to two-dimensional pages; thus, extra effort is required to interpret and apply them to the working environment (Wuttke et al., 2022). These issues may seriously hinder workers’ performance. To reduce the inconvenience posed by paper-based manuals, augmented reality (AR) smart glasses (i.e., AR in short) are introduced in companies to improve workers’ performance. However, due to the novelty of AR technology, current streams of literature focus more on the technical side of AR instead of the behavioral or managerial sides (Kim et al., 2016). Therefore, in this study, we aimed to bridge this gap by investigating the effect of AR on workers’ performance and exploring its underlying mechanism as well as its implications.

Nowadays, firms increasingly use AR to improve business (Wuttke et al., 2022). For example, Boeing has reported that AR reduces 35% of the training time new employees need to learn how to assemble aircraft wings, while the logistic giant DHL has increased productivity 25% by implementing AR-guided picking (Porter & Heppelmann, 2017). From the perspective of Industry 4.0, digitalization with technologies like AR inherently offers new ways to address the future of work (Olsen & Tomlin, 2020). In these industrial cases, it is obvious that AR has the potential to benefit business. However, seldom has any research provided empirical evidence to support the effect of AR. There are some exceptions. For example, Schein and Rauschnabel (2021) studied AR in a manufacturing context, but they mainly focused on the barriers concerning AR adoption from the perspective of technology-resistance using the survey method, while our study uncovers the short-term effect of actual AR usage in a field experiment during the AR implementation phase. Moreover, Wuttke et al. (2022) and Gürerk et al. (2018) examined the learning effect of AR on workers’ performance, while our study reveals the underlying mechanism of AR usage on performance from the aspect of cognitive processing.

Essentially, AR enables futuristic ways of information-delivery and transforms analytical data into a virtual layer in the real world. For example, in the industrial maintenance context, AR improves how workers visualize and consequently access all the information, how they perceive and follow guidance and instructions, and how they interact with the working environment. Based on information processing theory and cognitive load theory, AR-based instructions require less attention-split to access information than paper-based instructions, indicating a significantly lower extraneous cognitive load. More importantly, AR can provide information within the workers’ immediate field of vision while freeing their hands, which may enhance their ability to process information and facilitate their work (Coon, 2018).

We collaborated with China Southern Airlines and conducted a field study to investigate the use of AR in an airplane maintenance context during the AR implementation phase. We designed a within-subject experiment and tested the inspection performance of 80 workers before and after using AR. All the participants were first observed during their routine inspections with paper-based instructions. After proper training of AR usage and pilot inspections (at least three times to ensure familiarity with AR-based How AR Improves Workers’ Performance 2 instructions), all the participants were observed again during inspections using AR-based instructions. By collecting third-perspective video recording data and survey data, our objective was to examine the impact of AR’s short-term usage on workers’ information processing efficiency and their inspection effectiveness.

The results of our experiment suggest that, after short-term AR usage, there is a significant increase in the effectiveness of the inspections. The effect of AR on this improvement is mediated by the efficiency of information processing. This means that inspectors process information more efficiently when using AR- based instructions (versus paper-based instructions), which improves their inspection effectiveness. This mediation effect is stronger when inspectors perceive the instructions to have a lower extraneous cognitive load, because the way inspectors perceive how instructions are presented (i.e., via AR or paper) significantly moderates the effect of using AR on the efficiency of information processing.

Our study provides managers with insights regarding how AR improves workers’ performance. Our study results indicate that there is an immediate gain in short-term performance improvement after AR is used a few times. Specifically, with a lower extraneous cognitive load in presenting information, AR enhances workers’ efficiency in information processing, consequently improving inspection effectiveness.

Big Data (BD) has the potential to help firms overcome the subjective limitations of humans and lead to ambidexterity and sustainable financial results. At the same time, firm ambidexterity is a unique precursor to adopting BD. This research-in-progress aims to reconcile these differing views by identifying the interplay between an incumbent firm’s ambidexterity, BD, and top management roles. We adopt an explorative case study design to conduct semi-structured interviews with data scientists, top managers, and decision-makers within an incumbent firm. The interviews will be transcribed and coded in qualitative data analysis software to uncover richer findings (NVivo). We aim to lay the foundation for future research avenues around BD, top management’s roles, and ambidexterity in incumbent firms.

Introduction and Background

The ongoing war for talents shapes the future of organisations and is a major challenge for the entire field of talent management (TM). TM thereby involves (1) recruitment, staffing, and succession planning, (2) training and development, and (3) retention management. Nonetheless, organisations struggle substantially in managing the scarcity of talents effectively. The growing capabilities of artificial intelligence (AI) show potential in creating a strategic advantage in this war for talents (Brock & von Wangenheim, 2019). Given the rapidly increasing number of studies examining how AI impacts TM, the research lacks a comprehensive review. To address this void and thus support scholars and practitioners by providing an overview of extant knowledge on this topic, we conducted a systematic literature review (SLR) on the use of AI in TM.

Method

We based our SLR on the PRISMA approach (Page et al., 2021). To identify relevant literature, we searched in different electronic databases. This provided us with 3,714 publications from which 47 were eventually analysed in detail. We took a stepwise approach to select a conclusive final sample (see Figure 1). Based on previously defined criteria, we excluded non-fitting articles. Each step reduced our initial sample and eventually yielded a preliminary sample of 37 relevant articles. Based on these articles, we conducted backward and forward search, which resulted in additional 10 relevant articles. Consequently, in sum we received a final sample of 47 articles published between 2015 and 2021 on the influence of AI on TM.

Findings and Conclusion

We identified three overarching streams in the literature regarding AI in TM. Literature addressed AI use for recruitment, training and development, and retention management whereas the first stream was the most prominent. Regrading recruitment, literature emphasized prerequisites for and barriers to AI adoption. Cost-effectiveness of AI use, employee readiness, top-management support, and technology vendor support (Pillai & Sivathanu, 2020) are found to be crucial prerequisites for AI adoption. In contrast, security risks, privacy concerns, and high in-house technological complexity hinder AI adoption in recruitment (Pan et al., 2021; Pillai & Sivanthanu, 2020). Further, literature showed how AI can benefit recruitment efficiency across various TM activities such as outreaching to candidates, job-candidate matching, screening of candidates, candidate assessment and evaluation (Black & van Esch, 2020). For instance, machine learning (ML) algorithms trained based on video interviews, have been successfully employed to predict applicants’ personality traits (Hickman et al., 2021). However, literature also showed the importance of both, AI expertise and domain expertise for successful AI application (van den Broek et al., 2021). Despite possible use cases in training and development, like using AI to reveal potential skill gaps (Malik et al., 2020) or the introduction of AI based animated characters to provide feedback on learning progress (Vrontis et al., 2021), research in this area is scarce. In relation to retention management and the prediction of employee turnover through AI applications (e.g., ML for patterns detection and neural networks) is a well-researched phenomenon (e.g., Choudhury et al., 2021; Teng et al., 2021; Yuan et al., 2021). Additionally, AI has been employed to understand employees’ mood swings, enabling targeted countermeasures to prevent turnover (Kshetri, 2021).

The second stream comprises literature on individuals’ perception of AI application in TM. Research addressed user acceptance and willingness to use AI applications. Literature on user acceptance showed that AI in the application process is accepted when applicants are informed about its use early in the process. However, AI evaluation and selection decisions are more likely to be accepted by applicants, recruiters, and managers when matched with human decision-making (Laurim et al., 2021). Literature on user’s willingness to use AI showed contradictory results, as known AI use reduced willingness to apply for a position (Mirowska, 2020). Incorporating trendiness, fair treatment, and intrinsic rewards offered for using the AI, mitigated the resentments and increased users’ willingness to use AI (van Esch & Black, 2019).

The third stream targets algorithmic decision making in TM, addressing particularly AI bias and fairness and AI decision’s effects on users. Extant literature showed how biased decisions for example regarding gender manifested in ML training data (Köchling & Wehner, 2020). Particularly, algorithmic decisions were perceived less fair than human decisions and were perceived reductionist (Newman et al., 2020). Regarding algorithmic decision-making, literature identified a potential overconfidence in the algorithms’ objectivity, potentially even devaluing human decision- making, which was perceived subjective and deficient to the algorithms (Giermindl et al., 2021).

With our SLR, we offer a comprehensive overview of the extant studies on AI in TM. Thus, we enable scholars to identify possible future research options and practitioners to build decisions on the adoption and implementation of AI for TM on a scientific basis. Thereby, we contribute to both, research and practice.

Historically, mechanisation of production has always been accompanied by questions about its impact on the incentive to reallocate resources, with a natural focus on the substitutability of labour (Mokyr et al. 2015). However, labour substitution is only one of the effects of automation. In this paper, we study whether the adoption of robot technology influences the rate and direction of innovative activities.

In essence, robots are capital goods. However, contemporary robots are depicted as increasingly ‘malleable’, or flexible, capital goods – multi-purpose equipment capable of executing different tasks with little re-programming. Growing robot flexibility is a clear trend, as robot technology is augmented by other technologies characterising the fourth industrial revolution (Benassi et al. 2022; Martinelli et al. 2021), both hardware (e.g., sensors, or additive manufacturing technologies) and software (e.g., artificial intelligence algorithms). Robots become a component in larger systems, such as cyber- physical systems and advanced digital production technologies (UNIDO, 2019). As such, it is possible to hypothesise that robot adoption will induce changes in firms’ behaviours that go beyond the well- known replacement and productivity effects on employment (Autor, 2019) and that are more ‘enabling’ in nature. At the same time, current robots are “the most recent iteration of industrial automation technologies that have existed for a very long time” (Fernandez-Macias et al. 2021) that continue to operate in specific and constrained environments. Hence, their enabling capability might be limited if firms are not able (or do not plan) to exploit it. We shed some new light on this by measuring how product innovation and R&D expenditure change when robots are adopted at the firm level. Doing so, the paper contributes to the growing, yet nascent, strand of studies analysing firm-level data on robot adoption with a unique perspective on the nexus between the adoption of industrial robots and product innovation performance.

We exploit a unique dataset of Spanish firms, coming from the Survey on Firm Strategies (Encuesta Sobre Estrategias Empresariales, or ESEE) and implement an event-study approach (a generalised diff-in-diff model) to relate different indicators of product innovation to robotisation. We show that robot adoption is negatively associated to product innovation in the long term. We isolate the effect of large- vs small-scale investments mechanisation and find that the negative association with product innovation disappears for large-scale investments. Firms that are located in the top quartile of the investment distribution experience a positive increase in R&D expenditure (but not innovation), while firms in the bottom quartile display a negative relationship with both product innovation and R&D. We interpret the findings along a few lines of reasoning and converge on the idea that a conditional (on the scale of investment) substitutability exists between robotisation (process change) and the introduction of new products. In particular, implementation costs and the returns to learning- by-doing in process technology following robot adoption can divert resources away from product innovation. Furthermore, robots – even when flexible – might display enabling capabilities only when introduced in flexible production processes. More ‘classic’ and standardised mass production processes might not benefit from robots’ full potential.

We take a step further by discussing whether the types of robots under analysis are the ‘right’ robots to induce innovation. In fact, not all instances of process mechanisation and robotic equipment might be malleable enough to shape technological opportunities and to affect the incentive to engage in new product discovery, design, and development. The specific type of robots adopted do matter. In particular, innovation-inducing robots are those characterised by the feature of being research tools, invention machines, or IMIs. These types of robots are used to aid the search process over, for example, the space of materials to be employed or the space of designs to be trialled and prototyped. Industrial robots such as the majority of those captured by our data might not completely lack the capability to enable new activities; however, they are not IMIs, and have less scope for what concerns facilitating innovation-related search. New IMIs, such as certain types of AI algorithms, are mainly software technologies, which are used in knowledge-intensive domains and are not yet seamlessly integrated in the architecture and functionalities of industrial robots. By contrast, robots are employed in the manufacturing sector to increase the rate of execution and the precision of factory floor tasks under specific conditions.

To our knowledge, this paper is the first expanding the literature on automation to the microeconomics of innovation and firms’ strategic decision making. While exploratory in kind, our results suggest that non-linear mechanisms are at work within companies when robots are used to re- organise production activities. We conclude the paper discussing the implications of our findings for policy.

This work demonstrates an example of how arti- ficial intelligence, and specifically natural language processing (NLP), may be applied to automate as- pects of legal practice. After demonstrating this application, I consider the implications of legal automation more generally.

In common law jurisdictions, like the U.S., U.K. and Australia, judges and lawyers construct their arguments by drawing on judicial precedent from prior opinions. Judges cite precedent in their opin- ions and apply it to the facts of a case to build incrementally towards a final judgement. Lawyers use precedent in their legal briefs to argue why one party to the case should prevail.

U.S. case law currently consists of around 6.7 million published judicial opinions, written over 350 years. The process of extracting the correct precedent from this daunting corpus is a funda- mental part of legal practice. It is estimated that American law firm associates spend one-third of their working hours conducting legal research (Las- tres, 2015). Lawyers rely on legal research plat- forms to access and search legal precedent, which charge $60 to $99 per search, a cost that is ordi- narily passed on to clients (Franklin County Law Library, 2020). Access to justice continues to be a serious global problem. For example, “86% of U.S. civil legal problems reported by low-income Amer- icans received inadequate or no legal help” (Legal Services Corporation, 2017). Meanwhile, attorney fees continue to rise and are approaching $300 per hour in the U.S. (CLIO, 2020). Thus the price of legal advice is becoming increasingly unaffordable and access to justice is diminishing accordingly.

This paper presents a novel NLP approach to predicting judicial precedent relevant to a given legal argument by training NLP models on legal arguments made by U.S. federal judges. Given the time, expertise, and costs associated with identifying relevant precedent, this task represents a major barrier for widespread access to justice. The goal of this work is to aid attorneys in drafting legal briefs, reducing time and money spent on legal research while increasing access to justice and improving the quality of legal services.

Middle management in organizations has been acknowledged as pivotal for stability and legitimizing change initiatives. Research so far has built on the assumption of vertical layers in organizations and elaborated middle management in terms of intra-organizational sensemaking, politics and information processing. In the digital era, these assumptions and the role of current theories is likely to change. Data lakes and analytics capabilities provide all layers in the organization unlimited access. And the pace of change leads to major challenges in terms of intra-organizational digital practices and extra- organizational networking. Research so far has put premium attention to strategic change and change of work at the operational level. Middle management requires more attention to keep up with major changes.

Our objective in this abstract is to explore new directions for understanding the reshaping of middle management in the digital era.

In this policy paper, we investigate the impact of Artificial Intelligence (AI) in the workplace on the quality of jobs and the wellbeing of workers. Job quality is a multidimensional concept that includes all features of jobs that impact workers objective and subjective wellbeing (Nurski & Hoffmann, 2022). While labour regulation mainly focusses on the physical and contractual working conditions, two other job quality dimensions – job content (or job design) and the social environment of work – are the main determinants of worker’s behaviour, attitude, and wellbeing at work (Humphrey et al, 2007). As we show in this paper, AI will likely impact exactly these two dimensions of job quality, necessitating closer attention of policymakers.

Jobs and their characteristics are shaped by institutional antecedents (features of the labour market and the welfare state) and organisational antecedents (features of the organisation’s structure and culture). While AI has some impact on the functioning of the labour market through its role in the matching process, most of its impact will take place inside organisations. Therefore, in this contribution, we analyse the impact of AI on job quality by investigating how it acts on the organisational antecedents at the firm-level.

We argue first that job design originates from the division of labour and specialisation in the firm, both horizontally and vertically, in the production process and the governance process (Mintzberg, 1979). Job design then further shapes the other dimensions of job quality such as the social environment and the contractual and physical working conditions. Next, we construct a framework for assessing the impact of AI on job quality through its effect on the functions of the organisation, based on six AI use cases. Besides the traditional use case of automation of production or service tasks, we find five more use cases for the automation of management activities, also known as algorithmic management: (1) algorithmic work method instructions, (2) algorithmic task coordination, (3) algorithmic scheduling, (4) algorithmic surveillance of effort and performance, and (5) algorithmic staffing (including selection and recruitment).

Through an extensive literature review, we collect existing empirical evidence on these six AI use cases, and we show how they impact each dimension of job quality. Using the job demands – control/resources model (Karasek, 1979, and Demerouti et al, 2001), we first assess how AI either increases or decreases job demands (like work intensity and complexity) and job resources (like autonomy over planning and method, and skill discretion) for each use case (see also Nurski, 2021). We then show how these changes in job design spill over to the social and physical environment of work and finally put pressure on contractual employment conditions as well. We exemplify each use case by building persona’s that bring together empirical evidence from different sources into an illustrative story, easily understandable for both policy makers and business managers.

We finish this contribution by discussing how the previously described effects of AI on job quality are not technologically predetermined but are the result of choices of the technology designers (AI developers) and job designers (managers). Certain features of technology design might moderate the job quality impact, namely transparency, fairness, and human influence (Parent-Rocheleau and Parker, 2021). We illustrate why and how technology design might fail, either through incompleteness of data or because of the designer’s intention when specifying the algorithm’s objective function. We examine how technology design and implementation can be improved through worker participation. We briefly discuss the potential pitfalls and shortcomings of the proposed AI Act (European Commission 2021a) and the proposed platform work directive

Agriculture is making leaps in digitalization and artificial intelligence (AI) systems with autonomous machines, sensor data, and decision support systems (Liakos et al., 2018; Smith, 2020). Understanding and improving how farmersinteract with AI requires research that looks beyond AI in laboratory settings and into the application of AI in the field (Huysman, 2020; Jussupow et al., 2021). One key issue is explainability which paves the way for successful AI deployments (Gregor & Benbasat, 1999; Thiebes et al., 2021). Explainability refers to the effectiveness of AI’s explanations (e.g., user interfaces, documentation, or manuals). This study focuses on the comprehensibility of explanations and specifically user interfaces for end-users. End-users often cannot comprehend how AI systems reach their decisions (Waardenburg et al., 2020). However, explainability is crucial for using AI in joint decision-making (Asatiani et al., 2021).

Human-AI joint decision-making happens through configurations of Human-AI agency, which are continuously and mutually shaped (Suchman, 2007, 2012). Recent research found that a translator role is required who mediates between end-user and AI system (Gal et al., 2020; Jussupow et al., 2021; Waardenburg et al., 2022). The translator role addresses comprehensibility in domain-specific contexts. What remains unclear is how human-AI joint decision-making occurs when explanations influence it. Research into how AI explanations are embedded in the organization and integrated into decision- making procedures is lacking. How humans engage with AI systems and make sense of explanations in the domain context has seen little empirical work until now (Abdul et al., 2018; Benbya et al., 2021). These issues are urgent for small businesses, where human actors rely on AI explanations. Therefore, this study asks: How do configurations of human-AI joint decision-making emerge, and how do explanations influence these configurations?

The rise of artificial intelligence (AI) is a potential source of competitive advantage for firms to shape and re- design their organizations. However, the introduction of AI within firms has raised the usual question: “Will these machines substitute or complement humans in the workforce?”. Once an organization decides to introduce an AI machine, it wants to optimize its usage in such a way that performance is maximized and costs are minimized. However, this optimization problem is more complicated than one might initially expect because an organization’s internal processes often consist of different jobs that require different skills from the employees. Routinized jobs consist of tasks that can be codified and hence automated. By contrast, non-routinized jobs can be classified in two broad sets of tasks, which are proven to be challenging to automate. The first set is about creative tasks that are ‘abstract’, and require problem-solving capabilities, intuition, creativity, and persuasion. These tasks are typically allocated to workers with high levels of education and analytical capability, and they place a premium on inductive reasoning and communication ability. The second set includes manual tasks requiring situational adaptability, visual and language recognition, and in-person interactions. Specifically, manual jobs consist of both routine tasks and non-routine tasks.

To the best of our knowledge, no existing research compares the micro-level performance and cost impacts of the introduction of an AI machine on different types of jobs. The different types of jobs will result in different human behaviors, incentives and effort because the more the job consists of routine tasks the more the worker feels her job at risk of being automated. Thus, managers face a tradeoff in their decision making. On the one hand, they want to introduce the machine to improve firm performance and reduce costs. On the other hand, the introduction of the machine may lead employees to sabotage it, thus increasing the organization’s costs. In this work, we provide an answer to these organizational problems by showing how the different types of jobs perform after the introduction of the machine. We investigate our research question by means of an agent- based model to simulate the actions and interactions of two autonomous agents (i.e., the machine and the workers within the organization) and estimate the impact of AI on different types of jobs. Agent-based models are suitable for our research purpose as large-scale longitudinal data that trace interactions between AI machines and humans are not available.

Our simulation results show that, after the introduction of the AI machine, manual jobs outperform routinized and creative jobs in terms of both performance and costs. First, in the case of routinized jobs, since around 50% humans at any simulation time aim to sabotage the machine, the AI machine cannot increase its intelligence and performance. Thus, the manager cannot fire any worker, because the performance levels of the humans will remain higher than the machine’s performance level. Second, in manual jobs there is a peak in costs just after the introduction of the machine caused by recruitment costs; however, long-term costs are the lowest because of a relatively easy replacement of low-performing humans. In comparison with routinized jobs, workers are less likely to feel their jobs at risk, and thus to sabotage the machine. The overall human- machine performance is thus driven by human incentives and human labor quality. Third, AI machine performance in creative jobs is the highest in the long run. However, the joint human-machine performance is lower than in the case of manual jobs because of less machine assistance to humans (i.e., the machine cannot perform a large part of the job). This lower complementarity translates into very low chances of sabotage but less replacement of low-performers.

This study contributes to the emerging AI literature by showing how managers can cope with tradeoffs faced when deciding whether to introduce AI in their organization. Our work provides managers with guidelines on which job types benefit more from the introduction of an AI machine, and how synergies (conflicts) between AI and jobs positively (negatively) influence an organization’s overall performance.

Technology is changing the way entrepreneurs manage their human resources. Many employers have already started to dismiss the completely human exercise of their managerial prerogatives, totally or partially delegating them to more or less smart machines. Data collected through people or workforce analytics practices are the fuel to fill the tank of algorithmic management tools, which are capable of taking automated decisions affecting the workforce. Notwithstanding the advantages in terms of increased labour productivity, recurring to technology is not always risk-free. It has already happened, also in the HR management context, that algorithms have revealed themselves as biased decision- makers. This problem has often been exacerbated by the lack of transparency characterising most part of automated decision-making processes. Moreover, this issue is worse in the employment context because it increases the already existent information asymmetries between entrepreneurs and workers. These are the main reasons why it has been underlined how workforce analytics and algorithmic management practices may implicate an augmentation of managerial prerogatives unheard in the past. It has also been stressed that this should entail an update – or even a rethinking – of employment laws that, as they are today, may be inadequate to address the issues posed by the technological revolution.

This paper tries thus to understand, mainly looking at the Italian and other EU civil-law based legal systems, whether there are rules that may foster transparency and prevent abuses of employers’ managerial prerogatives potentially arising from the increasing recourse to algorithmic management practices. In other words, this article will try to examine whether there are any existing regulatory techniques that may be helpful in alleviating the issues of lack of transparency and augmentation of managerial prerogatives. In order to perform this task, I will analyse three different case studies of algorithmic management devices developed and deployed by Amazon in the US, to understand whether the implementation of these specific tools in the EU may have been legally feasible from an employment and data protection laws perspective, analysing three discrete legal issues, which are often at stake in employment litigation:

1. limits to employers’ monitoring and dismissal powers (case study: the algorithm that tracks and fires employees for productivity);
2. non-discrimination (case study: the machine learning recruiting tool that does not like women); and
3. classification of workers (case study: the invisible algorithm that guides and disciplines couriers). After the case study analysis, this article points to three types of regulatory techniques that may alleviate the issue of algorithmic opacity and, as a consequence, the augmentation of managerial prerogatives deriving from the always more massive use of algorithmic management devices. These three regulatory techniques are:
   1. information and access rights, to be exercised before a claim has been brought with a view to gather evidence to be used within a trial;
   2. rules that, within a trial, totally or partially switch the burden of proof to the employer or set presumptions in favour of the employee; and
   3. rules that, within a trial, grant employment judges with broad powers to gather evidence.

All these regulatory techniques strongly incentivise employers to recur to only those algorithmic tools with a decision-making process that can potentially be made transparent to their employees and, in case of a trial, to employment judges. Therefore, the employment legal system already knows how to foster transparency in the workplace and consequently uncover the violation of rules that already limit abuses of managerial prerogatives by employers. In light of the pervasive use of new technological tools to manage human resources, a more massive recourse to these regulatory antibodies can constitute an effective policy recommendation to better face the challenges posed by the algorithmic revolution.

As the scope of implementation of artificial intelligence (AI) in organizations gets wider, the question of how AI will change the role of managers and their perceived power among the employees becomes more relevant. While people-focused tasks, such as emotional support, conflict management and mentoring have always been important to manager’s performance, their importance might significantly increase once AI takes over the analytical tasks (Huang & Rust, 2018).

To perform people-focused tasks well, managers require specific soft skills, such as the political skill, defined as the “ability to effectively understand others at work and to use such knowledge to influence others to act in ways that enhance one’s personal and/or organizational objectives” (Ahearn et al., 2004, p.311). Managers with well-developed political skill can facilitate a higher level of subordinates’ job performance compared to mangers with less-developed political skill. This facilitation is partially induced by manager’s perceived power (Treadway, 2011). We focus on two sources of manager’s perceived power – reward power that is based on employee’s perception that manager has the ability to control once’s rewards (French & Raven, 1959) and social power that is defined as “the global perception by a follower of his/her supervisor potential to influence important organizational actors and the organizational decision-making process” (Chénard-Poirier et al., 2021). We suggest that employee’s behavior depends on the combination of these types of power.

Assigning AI the core managerial tasks, such as employee’s performance assessment, may significantly change managers’ autonomy and, as a result, their perceived power (Jarrahi et al., 2021). We aimed to explore the extent to which the allocation of employees’ performance evaluation to AI impacts the different types of managers perceived power driven by their political skill. Our findings demonstrate the importance of managers social power and political skill in the era of algorithmic decision-making.

Employee retention is becoming increasingly important, given the scarcity of high-skilled employees. Thus, practitioners and scholars are trying to understand and predict employee turnover as accurate as possible (Ben-Gal et al., 2021; Choudhury et al., 2021; Farrell and Rusbult, 1981; Oswald, 2020; Wang and Zhi, 2021; Yuan et al., 2021; Zhao et al., 2018). Complex machine learning (ML) models can offer powerful support in this context, as they enable high accuracy predictions (Hong et al., 2007). However, these models are often based on black-box models and provide limited interpretability of the results (Shrestha et al., 2021). Hence, to use black-box models, more interpretable approaches (Ben-Gal et al., 2021; Mitchell et al., 2001), i.e., explainable AI (XAI) methods (Guidotti et al., 2018; Hamm et al., 2021; Barredo Arrieta et al., 2020) are necessary. Understanding the models for turnover prediction – respectively the underlying patterns- is particularly important for Human Resource Management (HRM) to take appropriate countermeasures (Oswald, 2020). Therefore, we evaluate the use of three XAI methods in the context of turnover. As basis of our exemplary illustration, we use the publicly available IBM HR Analytics Employee Attrition & Performance data set (Subhash, 2017), which is a synthetic data set created by IBM that includes administrative data, performance data, data on job satisfaction, and data on individual characteristics (e.g., age and gender) of 1470 fictional employees (see Appendix for an exhaustive list of features) (Subhash, 2017). The goal of our exemplary use of XAI was to predict turnover (0 = no turnover, 1 = turnover) based on these characteristics and make the prediction interpretable.

For this purpose, we selected a sample of 7 frequently used ML models (e.g., Hastie et al., 2009; Wu et al., 2007) and evaluated their accuracy using 10-fold cross-validation (Berrar, 2018). The results regarding the achieved mean accuracy are shown in Figure 1. The Random Forest (RF) model achieves the highest accuracy. After the final model optimization, we achieve a f1-score1 of 84%. Due to the comparatively high accuracy, this model can provide a solid basis to predict employee turnover.

However, two fundamental questions remain. First, what are the general reasons for turnover in the organization? Second, how to prevent a turnover of an identified employee? Technically, these questions address two different types of explanations: Global interpretability (the average model behavior) and local interpretability (so the model’s prediction for an individual observation – here a particular employee) (Molnar, 2020). To answer these questions, we employ Local Interpretable Model-Agnostic Explanations (LIME) (Ribeiro et al., 2016), Shapley Additive Explanation (SHAP) (Lundberg and Lee, 2017), and Partial Dependence Plots (PDP) (Kamath and Liu, 2021).

For global interpretability, we use a SHAP summary plot (Molnar, 2020), as shown in Figure 2. This ranks the features according to their influence on the model output. Since the influence of the feature value is different for each instance, depending on the value of the other features, SHAP displays this with a scatter plot. For example, the feature StockOptionLevel has the highest average impact on the model, where especially a low level of stock options links to a turnover prediction of the model. In comparison, the feature TotalWorkingYears has a lower impact on the model. By analyzing these features and their model impact, the key drivers of turnover in the organization can be detected and used by the HRM for data-driven decisions (Oswald, 2020). In this case, the following question could be derived: What level of stock options is appropriate to prevent potential turnover?

Therefore, we take a more granular insight with PDP (Kamath and Liu, 2021; Molnar, 2020) to investigate the dependence between the feature StockOptionLevel and the model output (Figure 3). The plot shows that the curve flattens significantly from the StockOptionLevel of 1 onwards and thus a further increase in the level of stock options rarely changes the average response of the model’s decision.

Besides PDP, we used LIME and SHAP as they offer additional insights. Both XAI methods allow the analysis of individual observations (i.e., employees in our case) and thus enable local interpretability (Molnar, 2020). They display, in descending order, the strongest influence of the features on an individual model prediction (Figure 4). Both methods come to fairly close results and thus could be used to improve our understanding of potential reasons of a single turnover decision.

Finally, we demonstrated the use of XAI applications for decision making in HRM. XAI enables to make the decisions of complex ML models understandable. This inside can be used in two ways: First, to get a global understanding of turnover in the organization, and second, to understand possible reasons for the supposed turnover of a particular employee. This leads to the limitations with the presented XAI methods. In addition to the methodological challenges of the correct application (Choudhury et al., 2021), these methods do not provide any information on whether the taken actions actually change the decision of the model and thus, of the employees (Slack et al., 2020; Fernández-Loría et al., 2022). Hence, it is not possible to read off or determine what quantity of change, for example in the level of stock options, changes in the end the decision. However, the presented XAI approaches are a first step to get insights into complex AI models and to a data-driven assessment of which actions are most likely to be successful.

The social media entertainment (SME) industry operates at the intersection of Hollywood and Silicon Valley and is populated by professionalized content creators who develop their businesses based on followers on social media platforms (Cunningham & Craig, 2019). As the industry enters its second decade, we are seeing increasing efforts to regulate it from two angles: Content moderation by social media platforms (Gillespie, 2018) and the requirement for transparency of sponsored and branded content, meaning that creators are required to disclose whether they are being paid to promote a product or service (Abidin et al., 2020). Given the growing literature on the precariousness of creators’ work (Duffy et al., 2021; Arriagada & Ibanez, 2020), it is astounding that there is a lacuna in the labor law perspective of content creators’ employment status. Therefore, this study aims to fill this gap by bringing social media content creators into the employment law perspective.

First, drawing on content analysis, the study examines the Terms of Service (ToS) documents of the industry-leading social media platforms (Instagram, TikTok, and YouTube) that refer to the employment status of content creators. Second, drawing on an extensive literature review of the substance of creators’ work, the study looks at work activities through the lens of the International Labour Organization’s (ILO) Recommendation No. 198, which provides indicators of the existence of an employment relationship.

Analysis of ToS documents shows that platforms disguise work through clauses and language use. Unlike platforms such as Uber and Deliveroo, social media platforms do not classify creators as “independent contractors” but equate them with users of the platforms: “users who may provide content to the service.” Direct payments from the platforms to creators from the TikTok Creator Fund and the YouTube Partner Program are discursively framed as “rewards” or “bonuses.” The analysis shows that social media platforms engage in discursive work to obfuscate the work of content creators by portraying their activities as regular user activities and promoting the connotation of content creation as a leisure activity that could potentially be monetized. The latter has implications for the employment status of creators and also for a broader labor law perspective, as there is no case law on litigation brought from workers against platforms over the existence of an employment relationship, as has been the case for other categories of platform workers worldwide (De Stefano et al., 2021). Second, analyzing the substance of creators’ work through the lens of ILO Recommendation 2006,No. 198 shows that determining employment status is ambiguous. Control is one of the most germane indicators of an employment relationship (De Stefano et al., 2021).

However, creators are not under direct control, yet platforms do govern them with algorithms (Cotter, 2021; Bishop, 2019) and content moderation (Caplan & Gillespie, 2020). The second important indicator is whether the work is carried out within specific working hours. In the case of creators, the boundaries between work and leisure are conflated, as they create professional content on holidays and weekends and generally do not have set working hours. The next important indicator is whether the work requires the provision of tools. Creators, like other platform workers, work according to the BYOD (bring your own device) principle. However, as De Stefano et al. (2020) argue, the platform itself could be considered a work tool. Finally, another important indicator is the periodic payment of the worker. Creators who collaborate with brands do not receive regular pay, but are paid by “gigs,” and often the goods represent a form of payment. Creators who are part of the YouTube Partner Program or the TikTok Creator Fund, however, are paid regularly every month, in YouTube’s case on the 12th of the month.

Viewed through the lens of ILO Recommendation 2006, No. 189, social media content creators are much closer to true self-employment, as they are not under the direct control of a supervisor and are free to decide which clients they work for and set their own individual rates. Creators’ working hours are not clearly defined, and they use their own work tools. This indicates that they are de facto and de jure self-employed and not in dependent employment, a legal status that allows platforms to avoid their responsibilities to workers. Hence, creators lack labor and social protections like health insurance, lending, pensions, and other benefits tied to the employment relationship.

I conclude by discussing that in light of changing circumstances of work, we should think about providing adequate working conditions and social protections beyond the standard employment relationship, and argue for a platform-specific regulatory approach, as the work experiences of Uber drivers operating on “lean platforms” (Srnicek, 2017) and content creators operating on “advertising platforms” (ibid.) differ and cannot be regulated according to the same principles.

Theoretical Framework:

Behind the development of artificial intelligence is a growing body of digital human labour – often called microwork – remaining largely invisible to the public. This work is typically organized on digital labour platforms which mediate labour supply and demand. Research shows that within these mediated work environments, microworkers often work independently, have limited connections and communication with managers or colleagues, receive little feedback on their performance, and lack access to standard labour protections (Piasna et al., 2022). As microwork is one of the myriad forms of platform work, the working conditions found in microwork also include conditions of precarity found in platform working more generally, such as an uncertainty of available work, inconsistent workflows, and low payments per task (Graham et al., 2020; Heeks et al., 2021). Currently, the European Union’s proposed efforts to regulate platform working conditions, across the various forms of platform work, focus on issues such as the misidentification of worker employment statuses and collective representation (European Union, 2021; Georgiou, 2022). While these issues are especially salient for more observable ‘on-location’ forms of platform work such as ride-hailing or food delivery (Howcroft and Bergvall-Kareborn, 2019), research also shows that working conditions and worker characteristics differ across different forms of platform work (Eurofound, 2018). This raises two questions: how well do current ‘top-down’ strategies address relevant concerns for workers in other, less observable forms of platform work; do such legislative efforts entail contextual blindspots, creating opportunities for ‘bottom-up’ refinement via input from active microworkers within the EU.

Methodology:

Data is collected through semi-structured, in-depth interviews (N=82 of 160) with European microworkers active on six prominent digital labour platforms. Interviews were conducted online (n=72) and in-person (n=10); questions covered worker characteristics/background, experiences of microworking and microworking conditions, whether workers intend to pursue microwork in the future, and worker perspectives on improving these experiences. Analysis involves cataloguing microworkers’ priorities and concerns for improving microworking conditions. As the reality of microworking conditions is constructed through workers’ experiences, relevant investigation of these experiences requires familiarity with the nature and structures of platforms and microwork. This study, therefore, takes a Grounded Theory approach in line with the ontological and methodological positions of Straussian rather than classical Glaserian grounded theory (CGGT).

Preliminary Findings:

Microworkers often encounter similar conditions across different microworking platforms (e.g., lack of available work, lack of communication with requestors or platform, relatively low payouts for tasks). Nevertheless, workers differ in how they experience these conditions (e.g., as stressful, empowering, engaging or exhausting); some do not qualify microwork as work at all, while others are frustrated by the (lack of) career opportunities in platform working. However, workers consistently note how current strategies for improving platform working conditions may be misguided, ultimately creating disadvantageous conditions for workers seeking or limited to non-traditional forms of employment through microworking. Through interviews and policy analysis, these findings present active microworker concerns to provide a complementary ‘bottom-up’ perspective to refine proposed strategies for improving platform working conditions.

Gig work represents some of the extremes in temporary and precarious work (MacDonald & Giazitzoglu, 2019). Digital labour platforms often act as intermediaries that help workers find the next ‘gig’ (Duggan et al., 2020; Prassl, 2018), though not all platform workers are gig workers as some are hired as employees (Zekić, 2019) and not all gig workers are platform workers as some find gigs through other means than a platform (Koutsimpogiorgos et al., 2020; Watson et al., 2021). The rapid growth of gig work (Mastercard & Kaiser Associates, 2020) presents paradoxes allowing flexibility in determining when, how, and where to work (Gig Economy Data Hub, n.d.; Watson et al., 2021), but being equally characterized by job insecurity, the lack of professional mentorship, career advancement, and day-to-day work support (Kost et al., 2020). Further, an unstable work supply and career track has led scholars to pay attention to gig workers’ feelings of isolation and their struggles to obtain the reassurance, encouragement, and understanding as they deal with platform or client organizations (Ashford et al., 2018; Caza et al., 2018; Petriglieri et al., 2019).

This growing focus in scholarship on isolation has hinted at the roles that collectives and communication play in enabling a sense of belonging among gig workers (Ashford et al., 2018; Caza et al., 2018; Petriglieri et al., 2019). Referring to the idea of social identification, Ashforth et al. (2008) argues: “it is an essential human desire to expand the self-concept to include connections with others and to feel a sense of belonging with a larger group” (p. 334). Because the ‘traditional’ organizational loci for identification are insecure or altogether absent, this need for belongingness may be higher for gig workers than traditional workers (e.g., Ashforth et al., 2008), thus raising the question whether conventional notions of identification apply in the gig work context.

Our study, which is part of a larger research project on social identification in gig work, explores what is understood by ‘communities’ in the gig work context, what functions these communities perform, and the communication processes that both facilitate and inhibit identification with communities. Recent scholarship has indicated the importance of communities as an avenue for gig workers and professionals to connect with each other (Caza et al., 2018). However, research has not explored this question in depth yet. Guided by this nascent scholarship (e.g., Caza et al., 2018; Jenkins, 2014), our research explores communities as a particular locus for social identification and examines the role(s) of communication in identification as a process (Scott et al., 1998) in the gig work context. It is reasonable to expect that differences communicative characteristics in gig work (e.g., being able to casually speak to and receive advice from fellow workers) may lead to diverse needs that may explain identification with a specific type of community.

For this study, we conducted in-depth semi-structured interviews with 25 gig workers living in the Netherlands. Of these, 14 identified themselves as women and 11 as men, aged between 18 and 55 years. Nearly half of the interviewees work hybrid or from home (e.g., marketing, writing services, illustration, or IT), four outdoors (i.e., food deliverer and flyer distributor), and four on location (i.e., waiter, babysitter, store host, and Airbnb host).

Preliminary results support the centrality of communities as a particular locus for social identification. Gig workers were found to identify with different types of gig communities, reflecting Delanty’s (2018) distinction between location-based, virtual, and imagined communities. Accordingly, communities ranged from being bounded and concrete (e.g., a platform or client organization) to being permeable (e.g., Facebook groups and subreddits) to being hardly definable and abstract (illustrators or starting videographers on Instagram, or food deliverers in the world).

Depending on the gig worker’s communicative context, communities may serve a range of functions, encompassing professional day-to-day functions (e.g., quick answers on work-related questions, new work opportunities, salary benchmarking), professional relational functions (e.g., work inspiration, self-branding), and cognitive functions (e.g., overcoming loneliness, career sensemaking). Results also show that social interaction, even if very brief, can make an important difference in gig workers’ sense of belonging, and that their social identification can even be so strong that it gets described in terms of family metaphors.

This study contributes to current gig work research by further emphasizing the importance of communities for gig workers, suggesting how scholarly work on gig work can benefit from a social identity perspective, and expanding knowledge on social identification at work beyond the organizational boundaries.

The expansion of platform work has attracted the attention of journalists and academics, the public and regulators. Yet, despite the significant amount of literature and commentary produced over the last few years, there continues to be a knowledge gap about the gender dynamics that are present in the platform economy and about women’s experiences of platform work across different economic and social realities.

Much of the literature as well as some the regulatory initiatives and the labour inspections undertaken in Europe and the United States, have focused on male-dominated sectors such as ride-hailing and delivery. Even still, in these regions and in other parts of the world, little is known about the experiences of women gig workers who work as riders or drivers or the barriers they face. Furthermore, platform work has expanded and is likely to continue expanding into traditionally female-dominated sectors such as domestic work, home care and beauty services. Yet again, relatively little attention has been paid to these sectors.

To fill this gap, Digital Future Society did some exploratory research and published the report “Home care and digital platforms in Spain” in 2021. The research was based on desk research and semi-structured interviews with key informants, including platform CEOs and founders, as well as representatives of trade unions, academics and international organisations such as the ILO.

The report zooms in on platforms mediating home care services and offers an overview of the main platforms and their business models. The report also provides an analysis of the potential role of platforms in formalising and professionalising the home care sector. This is an ambition frequently expressed by platforms themselves in their marketing materials and in interviews and mirrors the discourse of digital platforms entering the sector in other countries such as India, the United States and Kenya. As discussed in the report, both domestic work and the overlapping field of home care are highly feminised, highly informal and precarious sectors, and they have historically been socially undervalued and invisible. Therefore any new actor that offers the potential of disrupting this sector, making it more “formal and professional” merits close attention.

My proposal is to share the work we have done in Spain at the Reshaping Work conference to generate a discussion around the impact platforms are having in highly feminised sectors such as care and domestic work sectors across Europe (and beyond).

With the increased reliance by firms and consumers on online platforms, the economy is transforming in a ‘platform economy’ with platform companies emerging as the new, global superpowers (Kenney & Zysman, 2016). A recent trend over the past decade is the rise of many platform apps that connect freelance workers to urban residents that demand ‘gigs’ on-site (e.g.taxi rides, food delivery). We call these labour platforms (Koutsimpogiorgos et al., 2020). Although “Big Tech” platforms operating in Europe mainly originate from the US and to a lesser extent China (Kenney and Zysman, 2020), labour platforms in Europe more often originate from Europe itself (De Groen et al., 2021).

Labour platforms are not as “asset-heavy” as traditional firms, resulting in swifter diffusion across countries. Platform focus on developing and improving software, which can then be replicated across borders supporting a rapid internationalization process, akin to a ‘born global’ strategy (Zander et al., 2015).

We ask the question what differences exist among European cities and countries in terms of the entry of new labour platforms and the international success these platforms achieve over time. We aim to investigate the urban and national conditions that have supported the entry of labour platforms over the past decade, and to analyze the sources of platforms’ success, or lack thereof, in terms of their internationalization over time. We proceed in three empirical steps.

1) Mapping

This study starts from mapping the geography of labour platforms in Europe using the CEPS database released in December 2021. It covers 278 onsite labour platforms active in Europe in 2021. We will extend the database by collecting data on the city of origin. This allows us to map the geography of labour platforms in European both at the urban and national level.

2) Entry analysis

We perform an entry analysis by counting the annual number of new labour platforms founded between 2000 and 2020 in each major European city exceeding 500,000 inhabitants adding up to 110 cities (www.geonames.org). To this end, we extent the original CEPS data with location-specific data on the founding city of each platform. We expect the number of labour platforms per capita to scale-up with city size as large urban areas provide labour platforms with a thick two-sided market to kickstart their business. Furthermore, liberal market economies compared to coordinated market economies (Hall and Soskice, 2001) may experience more entry of new platforms, as platforms in such countries may benefit more from venture capital, strong IPR protection, flexible labor market regulations and low taxes. We thus explain the annual urban entry rates of new labour platforms by urban and national characteristics.

3) Internationalization analysis

Finally, we investigate the geographical conditions that support internationalization of labour platforms, while controlling for platform-level characteristics as founding year and corporate vs. cooperative form. We expect that the larger the city in which a platform enters first, the higher its level of internationalization later-on. This reasoning builds on the idea that larger cities provide platforms with most resources, spillovers and market experience to expand quickly as a born global. Furthermore, liberal market institutions at the national level are also expected to foster internationalization in general. However, labour platforms exercising strong algorithmic control may match poorly with labour market institutions in Europe, and thus require adjustments by both platforms and social partners (Thelen, 2018). Hence, for those platforms (esp. in taxi and food delivery), we may expect platforms founded in coordinated market economies to show higher rates of internationalization than platforms founded in liberal market economies. Taking the number of countries in which a platform is active in 2021 as the dependent variable, we analyze the effect of city and country of origin on internationalization, by differentiating between platforms with high and low algorithmic control.

This paper examines the possibility of ‘purchasing’ labour-related services in crypto between the user that purchases a labour-related service from a platform worker and consequently transferring and allocating the renumeration benefits for the fulfillment of the work-related tasks and activities between the platform and the platform worker using blockchain technology. The content of the paper is divided into two sections. The first section addresses the technical aspects of blockchain technology with particular aspect of transferring the amount for purchasing a labour-related service between the user of the platform and the platform worker that is being engaged to fulfill the task, as well as following upon the transfer of the renumeration income from the digital labour platform to the platform worker, thus setting up an option for the platform workers to choose and enable whether they would like to use cryptocurrencies as a payment method for their work-related activities that they perform on a particular digital labour platform. The second section of the paper is addressing the legal and regulatory aspects of enabling the payment of online-based and location-based labour services via cryptocurrencies.

Regarding the technical aspects of using the blockchain technology in the world of work, the paper examines the general clarification of the blockchaing technology, explaining cryptocurrencies and the concept of volatility, which might extend or short the income of the platform worker due to volatility, if the renumerating benefit is transferred and held on a particular CEX or e-wallet. Furthermore, the subject of analysis of the paper is scoped around the explanation of fiat-backed, i.e. paper money pegged stablecoins (like USDT and USDC) and exploring the stablecoins that are using an algorhytm to maintain a consistent value, known as an algorhytmic stablecoins. Having in mind the high level risk that currently arises from the experimental character of the algorhytmic stablecoins, such as the TerraUSD bleak incident that lead to essentially the capitulation of the algorhythmic-backed stablecoins that, if chosen as an option of payment, might essentially annuls the platform workers’ value of the income, following a scenario of a possible de-pegging, additionally exponentiate the need of addressing the legal clarity of the cryptocurrencies on the EU level, in order to preserve the security, predictability and value of the platform worker’s income if they choose to be payed in crypto.

Furthermore, the first section of the paper explores the ways of setting up the infrastructure for paying the salary with an offered or chosen cryptocurrency, which essentially means selection of an adequate blockchain system to transfer the salary amount (choosing between volatile coin or stablecoin); addressing the question of avoidance of network congestion and selection of the transferring layer on the blockchain system, if the blockchain is multilayered in order to use in-between layers and bridges built on the blockchain; creation of E-wallet for the employer (to transfer funds) and the employee (to receive funds); Generating the sender and receiver address on the blockchain; Transferring the amount from the sender (the platform) to the receiver (the platform worker) and inspection and supervision by the national authorities of the funds that are transferred in relation to the ordering of a certain work-related services by the users of the platform.

Moreover, the paper outlines the intermediary aspect of doing a crypto transfer, thus exploring the options of selection and accession of a Centralized Exchange (CEX) to do the transfer from one payment address to another receiver address, or transferring directly from the sender of the funds to the receiver of the funds for the completion of a work-related activity via decentralized applications (dAPPS) that are built on a particular blockchain system.

Having the above-mentioned in mind, the second section and scope of this paper is targeting the need of coverage and the necessity of establishing legal and regulatory clarity for using crypto payments as a form of paying a certain amount of financial renumeration in the work-related context between the three parties that are interdepended in the platform economy: the platforms, the platform worker and the user/client of the service that is ordering a particular labour-engagement task via digital labour platform.

Regarding the legal contextualization of the usability and implementation of the blockchain technology in the non-standard forms of employment in relation to compounding and overlaying the technology and the law in the context of hiring and engaging labour via platforms, when there is a connection between the non-standard forms of employment with the usability of digital tools to access a potential working engagement, the possibility of ‘paying in crypto’ for the services being provided, can create a certain salary payment decentralization pattern, which indicates the importance of staying vigilant in tracking transactions on a particular blockchain by the national public authorities. This component is notably visible from the lenses of the public authorities that deal with revenue and income taxation, which indicates that national laws could be subject to amendment in order to enable payment in crypto and appropriate taxation of the income. Regarding the component of prevention of money laundering in the platform economy, the paper examines the EU legislation and the set-up regarding the compliance of AML (anti-money laundering) and KYC (know your customer) policies that are implemented for the digital labour platform, to appropriate monitor the workers on the platform into detection and prevention of money laundering in a particular context where the ‘paying in crypto’ option is enabled on the main platform, or while using a subsidiary tipping platform that aims to help the support of workers.

Having in mind that, as of now, in the current EU legal context, there are not any regulatory legislation in force that forbid or ban the use of blockchain technology and cryptocurrencies as a form of payment the labour services that are purchased from the digital labour platforms and are being performed by a platform worker, the last segment of the paper is dedicated toward the notion of composing an adequate combination and mixture between the new technology and digital transformation with its’ adequate usability in relation to the non-standard forms of employment, expressed through the working engagements that occur on digital labour platforms, for the need of pointing out further solutions how to increase the scope of the payment options for the working engagements and labour services that are being provided throughout the digital labour platforms, and consequently shaping up the platform economy in Europe, while laying the way to comprehensive legal solutions that regulate this non-standard form of employment and addresses the crypto payment as adoption of a digital innovation solution as a form of compensation for the work that is being performed by the platform workers.

Much research on digital labor platforms distinguishes between web-based platforms and location- based platforms. The first ones are platforms that outsourced work by direct call up to a geographically dispersed population that (such as ATM or Upwork) when the second are based on a dedicated application that allocates work to individuals in a specific geographical area (Berg, Furrer, Harmon, Rani, Silberman, 2018). However, a comprehensive literature review reveals a second difference in the way these platforms operate. The web-based platforms respond to a “just in time” dynamic (De Stephano, 2016), when the location-based platforms to a “just in place” dynamic (Wells, Attoh, Cullen, 2020). Thus, according to labor jurist De Stephano (2016, ibid), the expression “just in time”, applied to platforms, refers to an employment offer that combines easy access to a wide variety of (micro)tasks with a flexible model of work organization. These can connect an available workforce at any time, free of fixed time constraints and paid on a per-service basis. For Wells et alii (2020, ibid), the perspective is somewhat different. The notion of “just in place” emphasize the power of these platforms – such as Uber – to position drivers the best possible way in urban space to meet customers’ demand at any time. In both cases, the platforms form an ecosystem that places the worker in a new space and a singular temporality, suggesting a major transformation of the nature and practices of work.

Adopting an economics of convention (EC) approach, the presentation investigates one of these transformations by analyzing the decision-making space that is devolved to drivers in the convention framed by the Uber company. EC is a heterodox approach which focuses on the study of economic conventions as mutual expectation system (Postel, 1998). They are socio-cultural frames through which economic actors must coordinate, interpret, and evaluate their actions (Ponte, 2022) to build a coordination solution. Thus, in an increasingly digital world, the platform- type intermediary organization can be considered as the coordinating agent in capacity to frame a particular type of conventions. These conventions are called in standard economics a “two-sided” or “multi-sided” markets and they are understood as a particular modality for bringing together supply and demand, based on the double coincidence of needs (Rochet, Tirole, 2003, 2006; Evans, 2003; Caillaud, Jullien, 2003; Armstrong, 2006). These types of markets are founded on cross- network externalities. These can be defined as effects caused when the interest of one side to participate in the platform is conditioned by the number of users of the platform belonging to the other side (Rochet, Tirole, Ibid.; Caillaud, Jullien, Ibid.; Rysman, 2009). From an EC perspective they are the platform core value proposition. Platform organizations, especially marketplaces, derive their market power from their ability to apprehend the users’ utility of their service, on each side of the market, and then offer a single coordination solution so both of their need can be fulfilled.

The presentation will be divided into three parts. The first part constitutes a short but formal step towards deepening the theory to understand labor platform organizations from a conventionalist approach. The second part introduces our methodology. The communication is based on data collected from a French drivers’ exchanges gathered in a forum (about 100 000+ comments between 2015 and 2021). The data is processed through natural language processing (NLP) methods. Last, the third par present our results. Using NLP, specifically topic modelling tools, with special attention to constraint and autonomy we have intended to analyze and measure the decision-making space that is devolved to drivers in the convention framed by the Uber company, in the different time and space of the convention (including: working hours, rides “quality”, prices, location).

Web3 is a group of technologies that encompasses blockchain protocols, fungible (cryptocurrencies) and non-fungible tokens (NFTs), decentralized finance (DeFi), decentralized autonomous organizations (DAOs), decentralized applications (DApps), and the metaverse (Momtaz, 2022; Murray, Kim & Combs, 2022). It follows a natural evolution of the internet as Web1 (read) allows users to navigate static pages, Web2 (read-write) empowers an era of user- generated content, and Web3 (read-write-own) affords digital ownership to creators and users. The transition towards Web3 is made possible through decentralized blockchain protocols that rely on peer-to-peer networks (Korpal & Scott, 2022) and the programmability of smart contracts (Murray, Kuban, Josefy & Anderson, 2021). Furthermore, scholars are beginning to recognize the impact Web3 will have on society, such as supporting the achievement of the sustainable development goals (Voshmgir, Wildenberg, Rammel, & Novakovic, 2019), grass-roots movements (Ducrée et al., 2020), and education (Ferdig, Cohen, Ling & Hartshorne, 2022). Web3 can be viewed through a complex socio-technical digital innovation ecosystem (Wang, 2019), allowing actors to coordinate action among interdependent, loosely coupled and co-evolving value networks (Pagani, 2013). In a Web3 ecosystem, there exist four main governance mechanisms: (a) access is granted by actors who hold a wallet containing fungible and non-fungible tokens; (b) algorithmic governance is conducted through DApps, which are created for storage, voting, and financial purposes; (c) protocols communicate with the blockchain architecture which includes L1 and L2 chains, bridges and oracles; and (d) social governance is conducted through DAOs who provide access, discussion and compensation—see Figure 1.

Figure 1
A Conceptual Model of Web3 Ecosystem Governance

The emergence of Web3 complicates our understanding, as existing management theories in the strategy literature—e.g., transaction cost economics (TCE) (Williamson, 1975, 1985) and agency theory (Fama, 1980; Jensen & Meckling, 1976)—fail to explain Web3 governance structures as they are rooted in traditional forms of organizing. For example, TCE has provided a foundational governance basis to study the multiproduct firm (Hill & Hoskisson, (1987), the multidivisional form of an organization (Hoskisson, 1987), and equity joint ventures (Hennart, 1988). Similarly, agency theory enabled studies on corporate governance and boards of directors (Zahra and Pearce, 1989), executive compensation (Tosi and Gomez- Mejia, 1989), and the market for corporate control (Kosnik, 1990). Recent scholars have attempted to apply both TCE and agency theory to sub-components of Web3, primarily focusing on the protocol layer (i.e., blockchain architecture) (Sun, Garimella, Han, Chang & Shaw, 2020; Murray et al., 2021) and algorithmic governance (i.e., DApps) (Bellavitis, Fisch & Momtaz, 2022). However, strategy scholars have not examined Web3 as a novel form of organizing (Puranam, 2018), thereby failing to address the remaining sub-components which make up Web3 ecosystem governance, namely, access layer (i.e., actors) and social governance (i.e., DAOs). Failing to view Web3 as a novel form of organizing is problematic as it limits the understanding of the strategic management field towards a decentralized web with creator and user ownership at its foundation. Consequently, the current shortcoming of the TCE and agency theory may have far-reaching implications as the inferences drawn from previous studies may lose their relevance for actors that access Web3 and coordinate action through social and algorithmic governance mechanisms built on top of open – source protocols.

As a response, this study addresses these shortcomings by complementing TCE and agency theory with a microfoundational insight of organizations—see Table 1; offering a mechanistic explanation of actions and interactions conducted by a coordinated set of actors in tokenized value networks (Foss & Linder, 2019; Voshmgir, 2020). Specifically, based on netnography, this research will offer empirical support for treating Web3 ecosystem governance as a novel form of organizing to solve the universal problems of organizing along the dimensions of task division, task allocation, reward distribution, information flows and exception management (Puranam, Alexy & Reitzig, 2014). This research draws on 454 Web3 projects (Access: Web3 Database) from 2017 to 2022 active in consumer, fashion, film & entertainment, finance, gaming, infrastructure, metaverse, music, NFTs, social, talent, and ticketing categories.

This research intends to make three key contributions. First, this study recognizes a new empirical phenomenon—Web3—and elaborates on how traditional management theories fail to explain its governance mechanisms. TCE (Williamson, 1975, 1985) and Agency theory (Fama, 1980; Jensen & Meckling, 1976) risk losing explanatory power if they fail to integrate with a microstructural approach that adopts novel forms of organizing at its core. Second, focusing specifically on Web3 ecosystem governance and microstructures, this research aims to provide empirical support on how Web3 ecosystem governance responds to the universal problems of organizing along the dimensions of task division, task allocation, reward distribution, information flows and exception management (Puranam, Alexy & Reitzig, 2014; Puranam, 2018). Third, this study broadens the discussion beyond Web3 ecosystem governance as a novel form of organizing and opens the possibility of a rich research agenda related to Web3. Information technology and organizational design scholars are particularly well equipped to address future research questions. For example, what Web3 bestowed behavioural processes will affect existing forms of strategic organizational decision-making? How will agentic technologies (e.g., NFTs and Blockchains) be bundled to form an autonomous governance mechanism (e.g., DAOs and DApps) for platform- based industries? How will Web3 alter the dynamics of value creation (open platforms) and value capture (tokenized assets)? How will traditional business models built on leveraging customer data respond to Web3’s decentralized movements towards data ownership?

The world of work is changing at an unprecedented pace. With the emergence of the platform economy, the capitalist production process has undergone a rapid restructuring. The labor process at the workplace is also increasingly being digitally mediated. Drawing upon the Marxist approach in the study of relations of production in industrial capitalism, and premised on the theory of alienation, this study explores the changing nature of virtual and remote work beyond the traditional notion of the workplace due to the digitalization of the labor process. The theory of alienation is employed to shed light on specific practices such as anonymity at work, reduced interaction with the employers, consumers, and co-workers, and absence of supervision by the managerial class because of the introduction of algorithmic management of labor. While the critique of alienation is alive in the digital era, the logic of alienation replicates itself across the societal landscape, and the realm of digital media is far from exempt. Thus, this study helps us unveil the exploitative social relations behind the flexibility and autonomy offered to the platform labor. First, it briefly reviews the different theoretical frameworks of alienation in the digital economy. Subsequently, it elucidates how is alienation experienced and articulated in the platform economy. According to Marx, the alienation of labor occurs when the worker is alienated from the product of labor, from the production process, from fellow workers, and from self. The platform economy, built at the interface of technology and networking intensifies the feeling of alienation. The work setup, as a result of algorithmic management techniques enabled by platforms, has come a long way from being an assembly line industrial setup or electronic workshop. However, the platform economy relies heavily on the algorithmic management of labor. Algorithmic management is referred to as ‘giving the responsibility of assigning tasks and making decisions to an algorithmic system of control, with limited human involvement (ILO, 2021, p.33). In other words, the platforms have the ability to use algorithms to allocate, monitor, evaluate, reward, and manage the working conditions of labor (Graham and Woodcock, 2018; Gandini, 2018;

Rosenblat, 2018; Wood et al., 2019). For instance, deciding payouts and incentives, receiving orders and boosts based on geo-location, complying with the rating system, or facing penalties and unprecedented dismissals, platform labor has an interesting, yet unexplored relationship with the algorithms. First, while the platform users consider themselves autonomous, the surveillance and control of user data for marketable purposes makes the users lose control over their data. It is like an alien power exercised by those who have captured the data and thus brings estrangement, or alienation of their own product of labor. Next, the notion of alienation among platform labor has also been explored from Spatio-temporal axes. The blurring of work and home boundaries creates alienation of space and time for workers, as they lose control over their personal spaces and leisure time. Similarly, despite the belief that they exert control over their working time, the on-demand nature of their work leads to intensified work conditions and time-squeeze. This leads to unsocial and irregular working hours in order to meet client demands. Finally, reduced horizontal and vertical interaction among workers and employers due to the introduction of remote work and the algorithmic management of the labor process has led to the alienation of platform labor from fellow workers. While algorithmic control provides platform workers with formal control over where they work, workers may have little real choice but to work from home, and this can lead to intensification of a feeling of social isolation and lack of social integration. Hence, this study suggests that the alienating tendencies of digital technology and platform economy seem to be intensifying, given the lack of autonomy of and prevalence of dataveillance over the digital labor, blurred boundaries between work and home, and the lack of co-located workplaces in the digital era. However, why is it important? It is important because this is directly related to the quality of jobs emerging in the platform economy and has broader policy implications, especially in the post-pandemic scenario where unprecedented digital transformations have permanently fused technology with our everyday lives and work environment.

An increasing number of qualitative studies have documented the precarious conditions of platform workers (de Krijger, 2019; Huang, 2022; Cant, 2019; Wood et al., 2019). However, insufficient attention has been paid to explaining the causal mechanisms of such precarity. In the absence of a particular causal model, legal scholars have dominated the debate emphasizing employment security as the primary determinant (De Stefano and Aloisi, 2018; Eurofound, 2018a; Cherry & Aloisi, 2020). In their perspective, formal employment status with fixed time and wage contracts leads to higher levels of job quality. In contrast, self-employed status, characterized by a high uncertainty in working hours and income, leads to precarious working conditions (Hauber et al., 2020).

This legal perspective has significantly influenced recent policy decisions and attempts to regulate platform work. At the national level, parliaments have proposed measures to prevent the misclassification of platform workers as independent contractors (Hießl, 2021). At the regional level, the European Commission has recently presented a Directive Proposal on Platform Work, which enunciates that “in-work poverty and precariousness would […] decrease as a result of reclassification and the resulting improved access to social protection” (COM, 2021). In both cases, the intention is to prevent work precarity by reclassifying platform workers as self-employed workers.

In this regard, my research examines the presumption that employment security would effectively lead to higher job quality (i.e., less precariousness). This study explores the possibility that, although providing access to social protection, higher employment security may lead to adverse effects on other job quality dimensions. If that is the case, then reclassifying platform workers as employees could be counterproductive if the power imbalance between platforms and workers remains untouched. With the absence of collective power, the new employee status may lead to higher work intensity and lower income and autonomy, as platforms would seek higher productivity to compensate for the increments in labor cost.

Based on this general hypothesis, this research employs a mixed-method approach (Fig.1). The first phase involves unstructured interviews with food delivery couriers operating in Rotterdam. Through these interviews, I compare specific dimensions of job quality (e.g., income, work intensity, working times, and risk propensity) among employees and self-employed couriers. For the quantitative phase, I am conducting internet questionnaires based on the Employment Precariousness Scale (Julià et al., 2017). I have collected 128 questionnaires from platform workers in Rotterdam, and I continue in the data collection process. My conceptual framework (Fig 2) is built upon the multidimensional perspective of job quality, mainly developed in the work precariousness literature (Benach et al., 2014; Julià et al., 2017), complemented by the incentive perspective of Contract Theory (Grossman & Hart, 1986; Hart & Moore, 1990; Holmstrom & Milgrom, 1994).

In fact, the three hypotheses of this study are based on Contract Theory’s premises. The first one is that the firm’s decision to hire workers as employees or outsource their labor as independent contractors will determine the allocation of control rights (Hagiu & Wright, 2019). Control rights yield decision capacity over the everyday aspects of work, which are commonly unspecified in contracts given their incomplete nature (Grossman & Hart, 1986). In this regard, I hypothesize that H1: the more control rights workers possess, the higher their degree of perceived autonomy. In other words, despite being controlled by algorithmic management techniques (Huang, 2022), I expect couriers operating as freelancers to perceive a higher degree of work autonomy than couriers operating as employees.

The second hypothesis indicates that H2: possession of control rights is negatively correlated to income. In this sense, since self-employed couriers retain most control rights, the firm will have to offer them higher-powered incentives (i.e., dynamic rates in pick hours) to incentivize more deliveries. In contrast, when the platform possesses most control rights (i.e., when it hires workers as employees), it will compensate couriers with low-powered incentives (i.e., wage and social benefits). I hypothesize that this difference in payment scheme will affect the total amount of income that couriers make per hour. In this sense, I expect to confirm the result of other exploratory studies (FNV, 2020), showing that self-employed couriers make a higher income per hour than employed ones.

The third and most relevant hypothesis is that H3. couriers operating as employees will be subject to higher physical work intensity than self-employed. I hypothesize that the labor costs taken by the platform (e.g., payroll taxes, pension contribution, holidays, social benefits) push the firm to make use of workers’ subordination as employees to its highest. Considering the labor market characteristics of the food-delivery industry: (1) labor-intensive industry, (2) low-skill labor force, and (3) high turnover rates, I hypothesize that the only possible way for these platforms to increase productivity is by increasing the number of deliveries per time and reducing break times, driving workers to physical exhaustion and health problems. The expected findings regarding this hypothesis might be extended to other low-skilled in-site types of platform work, such as cleaning services or car driving, where the firm can only increase productivity by increasing output.

For practitioners, this research could bring relevant implications. The unexpected consequences of turning self-employed platform workers into employees would demand a revision of the EU Directive Proposal on platform work. If the hypotheses hold, effective legislation should aim not only to change employment status but also to shift the power imbalance between workers and platforms. This might be achieved, for example, by providing workers collective representation rights regardless of whether they are considered self-employed or employees.

As ever more organisations, workers, governments, and researchers turn to online outsourcing, it becomes important to obtain a comprehensive understanding of the incrementally growing body of research on this topic. Starting from the early 2000s, online freelance marketplaces like Upwork, Freelancer, and Fiverr, have been facilitating the online outsourcing of small-size and high-skilled service work to online freelance workers around the globe. Since then, numerous(inter)national policy documents have emphasized the promising labour opportunities provided by thistype of digital marketplaces(Kuek et al. 2015; ILO 2019a; World Bank 2019). As the size of the industry continuous to grow and the supply of workers becomes increasingly dispersed across countries (Stephany et al. 2021), these workers’ casual and flexible work arrangements and the global competition they face have become a well- established example of the changing world of work (ILO 2019a; World Bank 2019). Research on online outsourcing in general and the labour practices of online freelance workers in particular, could provide a lens through which practitioners, governments, and researchers look at and shape jobs in the future. However, since myriad terminologies have been used to articulate (labour in) the industry (Heeks 2017) and considering that research on online outsourcing and online freelance labour is dispersed across a range of academic disciplines, a substantive overview is lacking. This study fills this gap by learning the breadth of empirical, academic research on online outsourcing and online freelance labour and synthesising the findings to identify trends and prevalent themes that have been discussed. It also highlights knowledge gaps, based on which a future research agenda is being constructed.

Through a systematic literature review, this study identifies 92 peer-reviewed and empirically grounded articles on online outsourcing. The analysis of these articles consists of two steps. First, based on the author-selected keywords, a keyword co-occurrence map is constructed which groups research themes in four clusters: (1) the functioning of online freelance marketplaces, (2) workplace control and algorithmic management, (3) online freelance labour practices, and (4) experiences of online freelance workers. An additional map visualizes in which time period research on certain themes was trending. It shows that in the early years of research on online outsourcing, researchers were mainly interested in the technical aspects of online freelance marketplaces. They studied the workings and implications of reverse auction mechanisms and the inferences of reputation systems and other platform-introduced measures to improve triadic trust between marketplaces, clients, and workers. Overtime, the social aspects of these marketplaces gained ground with research centring around online freelance labour practices and online freelance workers, i.e., clusters three and four.

The second part of the analysis focusses on these social aspects by analysing the 56 articles grouped in these clusters in more detail. A qualitative, thematic analysis of these papers gives a comprehensive understanding of the socio-economic opportunities that online outsourcing offers to online freelance workers and the systemic struggles they face. It also provides an exhaustive overview of how they exercise their agency to overcome these struggles and attempt to build a sustainable career on online freelance marketplaces. In studying these aspects, previous research has paid specific attention to online freelance workers from developing countries.

As online freelance marketplaces keep altering their platform policies, it is important that future research continues to study how this impacts the opportunities and struggles experienced by online freelance workers and how they exercise their agency to deal with these changes. With respect to the latter, this study highlights a remarkable gap in the literature regarding the regularly drawn linkage between the future of work and lifelong learning (ILO 2019a; ILO 2019b). So far, little remains known about how online freelance workers shape their career on online freelance marketplaces by committing to formal and informal lifelong learning and future research should aim to fill this gap. This study finishes with implications and (policy) suggestions for practitioners and governments.

In this paper, we examine the possibility for online platforms on taxi and domestic services, to be part of the future of hybrid work. Combining multiple jobs to make ends meet is becoming a reality for more people every day. We know that hybrid employment is especially frequent in the Dutch platform economy, in which a majority of workers combines platform work with other forms of work (Pesole et al., 2018). Although platform work is often studied in isolation (Ilsøe, Larsen, & Bach 2021), our research question is: ‘Hybrid employment in the platform economy: Who is juggling multiple jobs and how?’

As part of the multi-method approach in our study, we have conducted 30 narrative interviews with hybrid platform workers. We sought to compare experiences of people doing hybrid work via household platforms such as Helpling or Werksters with experiences of people providing taxi services via platforms such as Uber or Bolt. Drawing on the results, we distinguished different modes of multiple jobholding in the platform economy as forms of reshaping work practices. Although platform work offers challenges for workers regarding for example income, stability, and security, we find that it also offers solutions for people who want to top up their income. Factors such as entrepreneurialism, flexibility, and independence, ensure that people can combine platform work with other types of work and that they feel they are in charge of their own work week. One respondent for example works half of the year for Uber, and the other half as a bridge worker. Another respondent works 20 hours per week as a marketing specialist, and 5 to 10 hours per week via Care.com. Furthermore, we interviewed people who run their own company in a different industry next do doing platform work, and students who juggle multiple jobs, among which platform work, next to completing their education.

Following from their narratives, parttime platform work has become a way for workers to reshape their work life balance. It offers a – sometimes temporary – solution for unexpected situations such as a pandemic, or for flexible work contracts at other companies. Furthermore, gamification within platforms causes workers to view it as their hobby instead of their parttime job. Specifically, it seems to offer solutions for migrant workers in the Netherlands, who experience a distance to the labor market. Because of cultural and bureaucratic barriers in the regular Dutch labor market, they find more success in entering the platform economy (Van Doorn 2021; Rözer et al 2021; Schor 2020; Huws et al. 2019). Some migrant workers in our study googled for ‘cleaning jobs’, as they felt that this was the stereotypical job for immigrants with a language delay. They did not have much success applying for regular jobs, which is why they purposively searched for cleaning platforms. Furthermore, not having any colleagues to work with, and the blurring of photos and names online, made migrant workers feel more protected against discrimination in their workplace. Via platforms they felt simply more in charge of their own work and less pushed to comply with a Dutch work culture.

As labor shortages are rising while platform work is expanding, we aim to contribute to a more sustainable work future. By focusing on perspectives of people combining platform work with other types of work, who are in this way reshaping their work life balance, we argue there might be possibilities for platforms to cooperate with offline employers and to offer new ways of working. As platform work currently offers many risks for workers, we discuss what hybrid platform workers would need from the platform in order to make their work more secure, safe and sustainable for the future. By describing bottom-up challenges and support factors within the platform economy, we aim to contribute to a fairer use of workers’ labor potential while acknowledging what platforms have to offer for this group.

Gig workers connect to short-term ‘gigs’ – via an app that is controlled by an algorithm (Kuhn & Maleki, 2017). When gig workers start working in the gig economy, they need to learn how to interact with the app. That is, more and more workers work in the gig economy via an app instead of being onboarded by ‘real’ people in an organization. As a consequence, at the beginning when gig workers are new in the gig economy, gig workers have to figure out on their own the skills and knowledge to understand their role.

However, the current understanding in the gig work literature is limited to understanding the interaction between algorithm and human from the perspective of the human as a follower of the algorithm as leader or manager. From the algorithm side, algorithmic management defines management performed through self-learning algorithms that are in control of decisions regarding labor priorly performed by HR professionals or supervisors (Meijerink & Bondarouk, 2021). The algorithm does not only match work but also has control over the work process (Duggan et al., 2020; Möhlmann, et al., 2021). For example, Rosenblat and Stark (2016) elucidate how the algorithm decides who gets which gig and how much the gig workers gets paid. Thereby, algorithmic management is characterized as a self-learning machine. That is, the algorithm self-learns how to optimize its formula to control its environment. Thereby, in the algorithmic management literature, the workers’ perceptions toward seeing and dealing with the algorithm are assumed to be static.

By contrast, from the human side, a limited number of studies have started to explore the role of technology on the changing nature of work (Kellogg et al., 2020), focusing on perceptions of algorithmic management decision making (Lee, 2018) and user experience of algorithm artifacts (Shin, et al., 2020).

As a consequence of this perspective on the interaction between algorithm and human, the algorithm is understood to dynamically adapt to the human, but the ways in which the human side can be dynamic is overlooked. To address that gap, this study focuses on a key process in employment relations; socialization. The socialization literature, describes a process in which the worker transforms from being an outsider of an organization into becoming an insider of it. Socialization is defined as “a process by which an individual acquires the social knowledge and skills necessary to assume an organizational role” (Van Maanen and Schein 1979, p. 211). The socialization process plays a key role for influencing a worker’s organizational identification (Ge et al., 2010) and commitment (Solinger et al., 2013). Yet, we do not understand how workers ‘socialize’ when working ‘for’ an app in the gig economy.

Drawing on an ethnographic study consisting of 75 semi-structured formal interviews, 14 conversational interviews in the field and 10 days of field work resulting in 80 hours of shadowing, we generate insights on gig workers’ work experiences with an app. Through participant descriptions and shadowing, we developed a process that consists of three sequences describing the socialization to an app: 1) descriptive instruction stage, 2) treasure hunt stage, 3) final socialized stage 3a) working for an algorithm or 3b) working with an app. Based on the distinct sequences within the socialization process, gig workers have different interpretations of what is going on with the app which in turn causes different interactions with the app. Linked to the different interactions distinct commitments are formed toward the app.

This study contributes to gig work and socialization literature in two ways. First, we add to gig work literature (cf. Duggan et al., 2020; Meijerink & Keegan, 2019) by developing a dynamic perspective on gig workers’ perceptions of algorithmic management. We elaborate how gig workers socialize to the technology they are working with in the gig economy. We connect through this socialization perspective new organizational forms such as platform work with the peoples’ work experiences working on it. Second, we add to socialization literature (Solinger et al., 2013; Van Maanen & Schein, 1979) by theorizing socialization with an app. We examine socialization in new era digital workplaces consisting of working via an app.

This research has important practical implications for platform organizations. For example, including gamifications in an app only triggers and influences the attitudes and work behaviors of gig workers in sequence 2 or 3a. Platforms need to adopt different engagement practices to reach all workers on their platform.

The arrival in Brussels Capital Region of Uber and Airbnb, two major players in the sharing platform economy, has not remained unnoticed. Especially governments and competitors have paid close attention to the way both companies have made inroads in the market, witness the broad media coverage and scholarly interest. While both companies are known for their contribution to digital disruption, they also have had a major part in reshaping the market and institutional playing field, in the hopes of increasing market share and gaining acceptance among governments, users and other actors of the platform economy ecosystem.

Responses by the Brussels government and direct competitors such as taxi industry and hotel industry, have been strikingly divergent, depending on how Uber’s and Airbnb’s actions tended to interfere with the agendas of those industries. Uber’s arrival has come with more tensions than Airbnb’s arrival. The fact that both companies use technology and infrastructure in different partly explains this difference. As a result, issues of competition, legality, employment, and taxation, and the unavoidable disputes that revolve around these issues, have different meanings and outcomes in ridesharing and home sharing.

With our contribution, we aim to bring together insights from our ongoing and completed research on market and non-market strategies in the sharing platform economy in Brussels Capital Region. We initiated the research to meet a need for exploring sharing platform businesses in the Brussels context and to better understand how these app-based technology companies interact with other actors in that context. By mapping this context, we may identify the main challenges and opportunities for governments, incumbents, and for sharing platform companies seeking their way in the market.

The main purpose of our contribution is 1) to compare market and non-market strategies of Uber and Airbnb, as they expanded into the Brussels market; 2) to identify the major institutional benefits and challenges that emerged from their market expansion; and 3) to identify the institutional and strategic responses that the Brussels government and the incumbent taxi and hotel industries addressed to the sharing platform companies.

The findings in brief show that Uber and Airbnb employed market and non-market strategies in tandem to broaden their scope of operations and increase legitimacy in the Brussels market. Yet, a closer look at the results reveals a more differentiated picture. More specifically, Uber’s arrival was more conflictive than Airbnb’s arrival. Uber had to face protests, strikes, and legal complaints by the taxi industry over issues of unfair competition, licensing, legality and employment. Airbnb mainly had to confront disputes over issues of tourist taxation, data sharing, and the negative impact of the home sharing business on the quality of local community life and on real estate prices. Also, the establishment of regulations towards a level playing field made more progress in the home sharing than in the ridesharing segment.

The Brussels government, as evidenced, has alternately accommodated either newcomers or incumbents when it came to regulatory and legislative matters, although this alternating support was more pertinent in ridesharing than in home sharing. The different governance levels –European, national, regional, and city level– were another challenge for Uber and Airbnb, having to cope with a complexity of rules of the game in Brussels. Also noteworthy is that right-wing politicians were more favorable to platform entrepreneurs and legislative reforms than left-wing politicians, further challenging the decision making.

We recommend policy makers to be more proactive in creating a level playing field that reconciles the interests of both platform businesses and traditional incumbents. Innovative business models will continue to mark the future, whereas old and new sectors will eventually transform and, as it seems, converge towards new market outcomes. The spillovers that we observed between ridesharing and ridehailing, but also between homes haring and hotel, industry indicate that old and new businesses tend to mimic one another’s strategies and imitate technologies to find their way towards a common playing field. It is the government’s role to not lag behind in how this digital ecosystem is getting shape, while it is platform companies’ role to develop strategies that help to liaise with governments, rather than fixing institutional issues afterwards. Both Uber and Airbnb hold the potential to promote the urban sustainability agenda, to foster small business entrepreneurship and to create employment. The government should work towards a good balance between fair and sufficiently strict rules, on the one hand, and a minimum of bureaucratic burden on hosts and drivers, on the other hand, to ensure that also smaller players,such as home sharing hosts, can remain in the game.

The predominant focus on platform work as paid work risks to overshadow the amount of unremunerated and volunteer work both being reshaped by and reshaping the platform economy. This contribution aims to find parallels between the ‘gig work’/flexible work debate on the one hand, and the gigification/flexibilization of volunteer work on the other. Somewhat different from existing notions of ‘free labor’ and ‘hope labor’ in the digital economy, which often address online cultural production and social media use (Terranova 2000, Kuehn & Corrigan 2013), this contribution addresses the re-organization of ‘civic’ volunteering, such as the assistance of disadvantaged populations and community work.

First, I pay attention to the appearance and rise of ‘volunteer’ platforms in both the Netherlands and Germany, which appear in numerous forms and uphold slightly different interpretations of the platform notion. Nevertheless, as a whole these platforms do form part of a bigger industry known as the ‘volunteer market’ (Bussell and Forbes 2002) or the ‘solidarity and care economy’ (Travlou and Bernát 2022). This industry and the implementation of platforms within it carry a logic different from ‘regular’ platform industries, where not venture capital but public-private partnerships and project-based funding play a decisive role.

Second, I show how volunteer work both shapes and is shaped by these platforms. To illustrate the first process – the shaping of platforms – I address the large amount of volunteer labor carried out behind the scenes of the Berlin-based platform GoVolunteer. Specifically the work of interns – which counts over 500 since its foundation – is essential for the platform to function in the first place, with the interns carrying out relational work with the platform’s partners and deploying communication and marketing strategies.

With regard to the other side of the coin – how platforms shape volunteering – I show how these platforms rephrase ‘large’ social issues such as homelessness, poverty and migrant integration into small projects and gigs on the platform. By promising the users to ‘make a difference in one hour’, social needs are repacked into the smallest possible units, which can then be traded and exchanged on the platform. This packaging of social needs is necessary for the platform to uphold its promise of flexibilization and gigification of volunteer work.

Advocates of so-called ‘platform cooperativism’ not only claim that working conditions in the platform economy could be improved by combining digital platforms with the ownership and governance structure of cooperatives, but also that practices of workplace democracy would be more equitable when facilitated by information technology (Davis, 2016; Scholz & Schneider, 2016). In this study, I critically assess the latter claim using survey data from a network of four Italian worker cooperatives that use information technology for all collective decision-making.

The increasing interest and arguments for workplace democracy are often met with concerns over sustaining it in the face of efficiency pressures. Captured in notions like the ‘iron law of oligarchy’ (Diefenbach, 2019) or ‘degeneration thesis’ (Langmead, 2016) is the expectation that worker cooperatives eventually fail or have all power concentrated at the top. Because participation in decision-making involves democratic transaction costs (Pozzobon et al., 2012), a typical pattern in worker cooperatives is that of a participation elite and inactivity among other members (Lees & Volkers, 1996). Crucially, such inequality in member participation is opposed to the principles and values of (platform) cooperativism. Previous literature has suggested various ways of resisting degeneration (Langmead, 2016), including representative selection via sortition instead of elections, greater direct democracy and cooperative education. Neither of these remedies have proven to solve the problems of degeneration completely.

Information technology lowers the transaction costs in collective decision-making (Davis, 2016), such as by increasing information availability on decision options, simplifying the process of voting, and  broadening the opportunities for members to monitor each other and their leaders. This raises the question whether inequalities in member participation persist, or that digital mediation successfully facilitates collective decision-making on larger scales, for less committed members, in more loose-knit groups, and among members of diverse skill levels.

The research theorizes member participation in worker cooperatives in the form of a social dilemma: since individual members are unlikely to have a pivotal vote, all members are incentivized to remain inactive in decision-making (Wippler, 1986). However, the more typical observation in worker cooperatives is of a participation elite and inactivity among other members – reflecting broader trends in political participation (Geys, 2006). To hypothesize these inequalities in member participation, I elaborate on the social dilemma of member participation in large worker cooperatives (i.e. cooperative size) and then extend my model with pro-social preferences (i.e. affective commitment), group influences (i.e. social capital), and information levels (i.e. human capital).

To test whether variations in member participation persist in the decision-making of digitally organized worker cooperatives, I employed survey data (n = 418) from members of a network of four Italian worker cooperatives in the cultural, education and IT sectors which uses a digital platform as their primary infrastructure. The results show how digital mediation facilitates workplace democracy on larger scales and among members with lower human capital, yet fails to level participation rates for members with lower affective commitment and less social capital. Hence, this study provides a more nuanced understanding of the potential for workplace democracy of going digital. It thereby contributes to platform economy scholarship that called for cooperatives to democratize platform work (Scholz & Schneider, 2016; Schor, 2020) and to literature on resisting degeneration tendencies in worker cooperatives (Langmead, 2016).

Researchers increasingly pay attention to the role of trust in the platform ecosystem. Introducing trust into a platform ecosystem confronts the platform provider’s dilemma: trust leads to value creation, but too much of it can lead to disintermediation between customers and complementors. Therefore, excessive trust can damage complementors’ online value creation and potentially the value capture for a platform provider. We study how different trust bases derived from platform certification (calculative trust) and repurchase (relational trust) influence complementors’ online value creation. Analysis of a panel dataset––comprising of 35594 sampled complementors who provide customization services on 1688.com, the biggest business-to-business platform operated by Alibaba in China––supports our hypotheses: additional certification weakens the positive effect of platform certification on the complementors’ online value creation. There is an inverted U-shaped relationship between the repurchase rate of customers and complementors’ online value creation. The results of our analysis also provide evidence for relational trust that is more important in promoting complementors’ online value creation when the purchase value is high, while calculative trust seems to be less efficient.

Andreas Ihl, PhD

Danube University Krems, Krems, Austria
Track: Platform Economy

With the rise of platform work, researchers and practitioners alike are concerned with the question how we can make platform work more motivating and engaging for workers (Bush & Balven, 2021; Kost et al., 2020). Scholars ask how platform work can move beyond being a (side) hustle and move towards a fulfilling work setting (Ashford et al., 2018; Caza et al., 2021; Jabagi et al., 2019; Spreitzer et al., 2017). This study addresses these concerns. It investigates how the platform design – particularly the extend of its usefulness (utilitarian value), its
enjoyment while using it (hedonic value) and its possibilities to interconnect with other workers (relational value) perceived by workers – can add to platform workers’ thriving. The concept of workers’ thriving is a key element of positive work experiences in demanding work settings. Thriving is a positive psychological state among individuals in which they perceive a sense of vitality (a positive state of arousal) and learning (a sense of development) associated with their work activities (Spreitzer et al., 2005). Its enhancement is particularly
critical in platform work since such work settings often come with central challenges such as isolation, uncertainty, and stress (Ashford et al., 2018; Caza et al., 2021; Petriglieri et al., 2019). Still, our knowledge on how to improve workers’ thriving in platform work is limited. Due to the high relevance in relation to e.g., physical, and psychological health, productivity (Goh et al., 2022) and for dealing with challenges in platform working, I raise the question, what platforms can do to support workers’ thriving. As studies from organizational contexts show, carefully designed work environments in terms of e.g., supportive social structures can offer an important contribution to individuals’ thriving at work (Goh et al., 2022; Spreitzer et al., 2005). However, platforms now offer a different way of working and confront workers with missing traditional organizational structures replacing them, with a highly remote, virtual, technology-depended and platform-enabled work environment (Kuhn & Maleki, 2017). Thus, several of the found antecedents that push thriving in organizational work are not or less present in the context in platform work (Ashford et al., 2018), while other factors, especially the technological environment in form of the platform’s designed processes and properties, become dominant (Schroeder et al., 2021). Due to the omnipresence and high relevance of the platform design for work experiences and behaviour and to fill the mentioned the research gap on what platforms can do to support workers’ thriving, I argue that the design of platforms in which workers operate centrally influences their thriving in platform
work.

To answer this question, I shed light on the evaluations of the platform design by workers in the central dimensions of the design’s usefulness (utilitarian value), enjoyment in use (hedonic value) and social components in interconnecting workers (relational value) (Gerow et al., 2013; Köse et al., 2019; Waldkirch et al., 2021; Wood et al., 2018). I analyse how these aspects affect workers’ thriving in its emotional dimension of vitality and its cognitive dimension of workplace learning. My empirical analysis builds on a two-wave survey including 652 workers and a PLS-SEM approach. While this is ongoing research, the preliminary findings show that the perceptions of the platform’s usefulness, the enjoyment in using it as well as its social components are central factors that foster workers’ thriving in both dimensions of learning and vitality.
These insights offer several additions to research and practice about the nature of platform work and beyond. First, I extend the current body of knowledge by investigating external factors influencing thriving in platform work and the crucial role of platform design. I thereby show the
merit platform design offers for workers and their thriving. Second, by evaluation three different types of aspects of platform design’s values (utilitarian, hedonic and relational), I offer a nuanced perspective on crucial ways of enhancing platform designs enabling thriving and move platforms beyond their passive intermediating role and offer new pathway for developing platforms. Third, by assessing the two dimensions of thriving separately, I add to the understanding for the emotional and cognitive processes in place, depending in the type of perceived aspect of platform design. Last, beyond platform work: Workplaces become increasingly technology-reliant, remote, and virtual. I show how a user-centred design of information systems can help to deal with challenges in the new world of work and how their careful design is critical for motivating and engaging a remote workforce.

Although the “gig economy” is a popular concept in the literature and has been studied theoretically in terms of the dynamics of its pure form, platform organization (Meijerink and Keegan, 2019), little is known about what it actually means for organizations and existing processes once the decision is made to adopt platform-based work.  Researchers and practitioners agree that gig work poses new challenges for HR, procurement, and other organizational stakeholders by redefining the boundaries between employers and employees (Kroon and Paauwe, 2021). Intra-organizational responsibilities are shifting between departments, specific insights however, of how these challenges affect the role of HR and other organizational stakeholders are not yet provided.

For HR research a key question is how gig work affects our thinking of strategic HRM and how the role of HR is impacted once organizations decide to introduce gig work. This complex question is not easy to answer, as the emergence of gig work appear to (1) impact existing HR processes), (2) affect HR responsibilities and (3) introduce the use of algorithms which seem to fundamentally redefine the way we work from an organizational perspective.

To understand how gig work and algorithms impact the role of HR and other organizational stakeholders, we argue it is important to understand how HR processes and practices are changing in the context of gig work. By reviewing traditional HR practices within gig work we  get an overview of how responsibilities are shifting between different internal and external parties.

Our analysis shows that the introduction of gig work brings fundamental changes to the people management process, as the contents of responsibilities are divided between existing and new actors such as the platform provider and the end user. We identify the tensions that may arise in decision making related to human resource management processes due to shifting responsibilities between the actors involved.

The existing literature has already highlighted that the introduction of platforms and systems lead to fundamental challenges for organizational actors of different departments. With the introduction of algorithms, a new level of complexity is added as certain steps of the decision-making process are affected, involving different actors. We know that digital technologies are impacting existing HR processes in many ways, as technology offers new opportunities for candidate matching and selection, speeds up processes, and creates new access to data sources that facilitate decision making. However, specific insights of how decision-making from an HR perspective is affected when introducing algorithms is clearly missing. There is a lack of concrete insights about which tasks of original HR processes are taken over by technology via algorithms or other organizational actors and which remain with HR or other organizational stakeholders. The first part of our article therefore deals with concrete changes in content and responsibilities for organizational processes in terms of managing gig workers. Since gig work can be introduced in different forms through the use of different platforms, this article explores different types of platforms and the specific implications for organizational actors.

Our findings foreshadow that the current role of HR remains limited since decisions are made on an operational level and operations do not see the need to involve other different stakeholders for managing gig workers.

However, we argue that it is imperative for HR to be involved when it comes to gig work to ensure comprehensive talent management. The role of HR and procurement needs to be redefined, as the clear distinction between different types of workers does not comply total talent management of the 21st century. The focus of a line manager shifted from having person-related fit to having a job/task related fit so, that the requested job is successfully accomplished. We therefore strongly recommend organizations to redefine the boundaries of HR and Procurement and create a contingent workforce team handling all types of employees and focusing for a best fit to the job rather than on a best fit of the person. Such a team, composed of HR and procurement professionals, is then responsible for establishing policies that provide clear guidance on when, how, and what type of platform to use, highlighting the differences between corporate-owned and public platforms. This team should actively evaluate data provided by artificial intelligence or reports on the platform and bases future decisions on it for strategic workforce adjustments and performance management. Given the increasing volume of data provided by public or proprietary platforms, we see the need to introduce an internal organizational team, a Contingent Workforce Team, responsible for performing (A) predictive analytics from a market perspective (the responsibility of Procurement) to deviate supplier strategies and (B) from a talent perspective (the responsibility of HR) to proactively lead the fight for talent.