Digitalisation and working time

'Working time' refers to any period during which workers are working, are at their employer’s disposal and are carrying out their activities or duties, in accordance with national laws and/or practice. Working hours vary for workers in different occupations or at different life stages, and gender is particularly important in determining these differences. The digitalisation of work has changed the nature and the characteristics of working time, which creates both opportunities and risks for workers.

Overview

Full screen

Share

Show pages

Digitalisation

• Long and non-standard working hours
• Constant availability
• Lack of data privacy and excessive control
• Monitoring and surveillance of working time: implications for privacy and control
  

• Improved work–life balance
• Increased autonomy
• Enhanced organisation of working time and productivity

Risks

Opportunities

Introduction

The digital revolution can be defined as a general acceleration in the pace of technological change in the economy, driven by a massive expansion of our capacity to store, process and communicate information using electronic devices. These capacities of digital technologies enable communication and production for work purposes to be carried out from anywhere and at any time, with the exception of biological limitations related to biorhythms. Having access to the internet and cloud computing from anywhere using digital devices greatly reduces the temporal and spatial limits on work activities.

In the industrial era, the place of work and working time were constrained by the factory-based production system, in which tasks had to be performed at a specific workplace. These activities needed some temporal (and physical) coordination and therefore they normally followed standard working times involving morning and afternoon shifts.

In the post-industrial world of work, however, digital technologies enable more flexible production systems and service provision. These technologies provide access to information at all levels and points of the economic process, providing the potential for a very high level of flexibility in work organisation. Workers can potentially work at any time and from anywhere. A typical example is working remotely with computers, laptops and smartphones. However, all vectors of digitalisation (automation, digitisation and platform work) provide more opportunities to work remotely than non-digitalised work environments.

Apart from flexibility, other general characteristics of the organisation of working time in the digital world of work are increased autonomy for workers and the expectation of constant connectivity. Additional elements to consider in understanding working time in the digitalised work environment are the use of algorithms, which can influence or determine the organisation of working time, and the capacities of the technologies to carry out monitoring and surveillance of employees’ performance and working time.

Some of these characteristics are not only a consequence of digitalisation but are also related to general economic, social and labour market trends. These are not discussed in this document; however, digital technologies clearly contribute to their development.

Autonomy versus surveillance

Workers’ autonomy in the digitally enabled work arrangement of telework and ICT-based mobile work (TICTM) has been well documented. TICTM is a work arrangement characterised by working from more than one place, enabled by ICT. Generally, telework refers to working from home, whereas ICT-based mobile work includes working from places other than the employer’s premises, with various degrees of mobility. There are two characteristics of this work arrangement: first, the separation between the place of work and the employer’s premises, where traditionally higher levels of control are in place, and second, the profile of the workers involved. Work carried out remotely is mainly related to professional activities conducted by highly qualified white-collar workers. However, autonomy is not limited to this context and these occupations.

In the context of platform work, autonomy involves the freedom of workers to choose which tasks they do, their working time, and how to organise and perform their work. However, this is not the case for all types of platform work, and some platforms (with regard to on-location platform-determined work) impose more rigid limits on the organisation of working time based on the application of specific algorithms that determine when and where work has to be carried out.

There are no clear implications for further autonomy in relation to automation unless this automation implies the development of project-based work by IT companies.

Finally, in all digital-based work, both the use of algorithms and the use of monitoring and surveillance methods can pose a risk for working time autonomy and the data privacy of workers.

Flexibility

Workers and companies have made use of digital technologies/ICT to enhance flexibility in the organisation of working time. This development is also linked to other societal trends, which are not the focus of this document (for example, the globalisation and feminisation of the labour market).

ICT has facilitated new ways of organising work by giving workers more flexibility regarding when and where work can be performed. These forms of work organisation rely less on regular rhythms and, instead, work tasks are allocated more flexibly. This shift has been accompanied by a more general trend towards work that is project based and fragmented, carried out on demand and paid according to performance.

The European Working Conditions Survey (EWCS) 2015 found a high level of correlation between having a flexible work schedule and working with ICT.

Figure: Shares of workers reporting a flexitime schedule by level of ICT use, EU27 and the UK, 2015 (%)

Flexibility is also a key topic in the context of platform work. Many platform workers indicate that having increased autonomy and flexibility in their work is a significant motivation for undertaking platform activities. This is important not only as a large share of these workers have another (main) job but also because it helps in meeting caring and other personal obligations. As in the case of autonomy, this flexibility seems to be accessible only in some types of platform work, particularly those in which the skills requirements are higher and in which the platform has less influence on work organisation.

Unpredictability, availability and the role of algorithms

One characteristic of the digitalisation of work is the ability to access work tools and the actual organisation (for example, company, colleagues, clients, subcontractors) instantly at any time and from anywhere. This represents a substantial difference from more traditional types of work organisation, where work is performed in specific places and at specific times. The consequence is that employees working in a digitalised environment are more likely to be contacted outside standard working hours and with high levels of unpredictability. The automation of work can also contribute to unpredictable working time patterns because of the potential fragmentation of work (that is, some tasks are automated and some are not) and the use of algorithms designed to respond to demands without considering any specific organisation of working time. This may have implications for the organisation of working time (for example, prolonging the working day to make up for the interruptions) and work–life balance. Moreover, it can have an effect on work processes because of the number of work-related interruptions employees may be subjected to.

Another implication of digitalisation for working time relates to the use of algorithms to assign tasks, and therefore when and where work has to be carried out. As of mid-2021, this was most evident in some types of platform work. Platform algorithms can push employees to work when they do not expect to work, which can have a negative effect on their work–life balance. Moreover, several platform work activities require workers to be on standby or on call most of the time.

Blurring of working time and rest periods

The unpredictability, constant availability and use of algorithms inherent to remote work and platform work and, to a lesser extent, automation can potentially result in the blurring of boundaries between work and private life. The challenge is to clearly determine what is working time and what are rest periods. In some contexts, this leads to a high degree of porosity between the two, a situation that not all workers are able to manage well and that might have consequences for work–life balance and the psychosocial well-being of workers.

Opportunities

• Improved work–life balance: increased autonomy and flexibility in the organisation of working time provide clear opportunities for adapting working time to the needs of workers and employers. Therefore, if working time is well organised, it can have a positive impact on work–life balance and workers’ well-being.
• Less need for commuting: as digital technologies facilitate working from different places, there is less need for commuting, with positive implications for work–life balance and more efficient use of time.
• More efficient organisation of working time: increased flexibility in the organisation of working time provides opportunities to use human resources at different times and therefore allows a better fit between the different elements of production processes (for example, reducing wasted time between tasks).

Risks

• Long and non-standard working hours: in certain work contexts with 24/7 accessibility to work tools, relatively high workload levels and levels of autonomy, and a work culture that promotes high levels of availability and working long hours, workers who work remotely (for example, from home) end up working long and non-standard hours.
• Constant availability (ill-defined on-standby or on-call situations): the ability to contact workers at any time and anywhere, either formally or informally, may make more common situations where workers voluntarily or involuntarily carry out work at the request of their employer both during working time and during rest periods.
• Excessive monitoring and surveillance: employers are obliged to record working time; however, in the context of digital-based work, some of the methods used can create risks with regard to data protection and the privacy of workers. Moreover, they may jeopardise working time autonomy, which is inherent to the various ways of working remotely.

Concluding commentary

Digitalisation has an impact on how working time is organised. This relates to both technological aspects and organisational aspects. From a technological perspective, the two main drivers of change are the increase in ways of organising working time because of reduced temporal and spatial limits and the use of algorithms to determine working time patterns. For example, employees may be virtually on call all the time, even when they are not formally working, or working time may be more irregular and blurred. Digitalisation also reduces the time needed to perform tasks, and therefore it has the potential to shorten working time.

In principle, in the context of a competitive labour market, there is strong pressure to be constantly available. At the same time, technology enables workers and employers to organise their work in a way that is more suitable for both, providing a satisfactory work–life balance and meeting production requirements.

In the future, with the expected increase in the digitalisation of the world of work, the opportunities for improving work–life balance could be fostered while monitoring risks such as those related to the increased surveillance of working time. Social partners and governments should take an active role in addressing these challenges to help shape working time in the future world of work.

Automation is one of the ‘vectors of change’ identified as part of the broader notion of ‘digitalisation’ in Eurofound’s conceptual framework. It is the replacement of human input, in full or in part, by machine or software input. Advanced robotics, both for services and for manufacturing, is grouped with autonomous vehicles under the automation vector, since the ultimate aim of their application is to substitute machine for human input.

Introduction

Automation has an impact on the types of tasks carried out and the order and way in which they are performed. The effects of automation on the organisation of working time are less obvious and may depend on the specific occupation and tasks to be performed. For example, workers in maintenance roles may be called on at any time to fix non-functioning automated processes or take control of an algorithm if something is not occurring as expected in the production process. Another general effect is that automation ‘liberates’ working time and therefore has the potential to reduce working time. However, this effect depends on both the automated tasks in a specific occupation and working time and employment policies at company level.

Opportunities

Advanced robotics can improve the quality of working conditions (such as work–life balance) because of the potential to work remotely, and automation in general can result in shorter working hours because of reduced workloads and improved quality of working time in terms of more consistent work schedules, regular working hours and reduced commuting time (for example, in the case of autonomous vehicles).

Risks

The impact of advanced robotics may be negative in the case of 24/7 production processes if workers are called on at unsocial hours to check or fix failures in automated processes.

Moreover, working time flexibility and autonomy can be reduced in the case of tasks that are more dependent on automated and algorithmically managed processes. This aspect is relevant for artificial intelligence (AI), for example. AI is a wide-ranging branch of computer science concerned with building smart machines capable of performing tasks that typically require human intelligence, and its application may therefore leave less room for human flexibility and autonomy.

Concluding commentary

Automation can lead to reduced working hours as long as company policies favour this option. However, in some cases, tasks may be reorganised, and thus the advantages of reduced working time may not be realised. Improvements in working time can also result from working remotely and the associated flexibility and autonomy. The role of algorithms is also very important in determining or influencing the organisation of working time.

Public policies, companies and social partners can contribute to shaping the way in which automation affects working time, for example in relation to the configuration and management of algorithms or the extent to which an interest in reducing working time is put into practice.

Digitisation is one of the vectors of change forming part of digitalisation in Eurofound’s conceptual framework. It refers to the process through which aspects of the physical world are rendered into data and virtual models, and vice versa. Three main technologies fall under this vector of change, namely 3D printing, augmented and virtual reality (AR/VR), and the internet of things (IoT).

ICT has an impact on the world of work by providing the ability to work at any time and from anywhere (remote work). This has implications for working and employment conditions, including working time.

Introduction

Research on digitisation, and more specifically the use of ICT at work and the organisation of working time, has been more prominent than other areas related to digitalisation. Eurofound research shows that there has been a real erosion of the working time standards of the industrial era.

In TICTM, it is more difficult to distinguish between working time and non-working time, which some authors have referred to as ‘increasing time permeability’. The internet, smartphones, tablets and laptops have become part of everyday business and life, increasing the likelihood of workers being reachable by job-related communications such as telephone calls, emails and instant messages outside normal working time. The development of cloud-based technology has also contributed to this trend, providing access from anywhere and at any time to documents and resources, which, prior to the spread of digitisation, were available only in the workplace.

The adoption of ICT for working remotely has grown, as demonstrated by the increasing number of workers working from home and the digitisation of businesses. With the COVID-19 crisis forcing many companies to switch to telework, there has been a spike in the number of teleworkers in Europe. In 2020, an average of one in five employed people worked occasionally or usually from home.

The use of AR/VR can result in reductions in workload, as the technology complements human activities and decreases task complexity. As in the case of other technologies, AR/VR can shorten the time needed to perform certain tasks. In relation to 3D printing and IoT, employees and companies will also benefit from increased opportunities for remote work, which may have implications for the organisation of working time, as discussed above.

Opportunities

Shortening of the time needed to perform tasks and further opportunities for working remotely are common traits of digitisation at work. This can result in improved organisation of working time and a better work–life balance.

Specifically, the main advantages of TICTM are as follows.

• Teleworking is seen by many workers as a way to increase their autonomy and flexibility at work, with positive effects on work–life balance.
• Positive effects usually also include reductions in commuting time.
• These aspects can have a positive impact on productivity, partly because of more efficient use of time.

Risks

The main disadvantages of TICTM, and to some extent the other technologies in the digitalised workplace, relate to working remotely.

• It is more difficult to distinguish between working time and non-working time. Where this is not well managed (for example, in cases of constant availability and highly demanding jobs), it can pose difficulties in maintaining a good balance between work and private and family activities.
• The (perceived) need to work during one’s free time may, in some contexts, be aggravated by an organisational culture that accepts long working hours and heavy workloads as normal, leading to an intensification of work.
• Those working from home, and especially those working with ICT with high levels of mobility, are significantly more likely to report working long hours (in excess of the 48 hours per week stipulated in EU legislation) and having rest periods of fewer than 11 hours between working days.
• Presenteeism – working when one is sick – also seems to be indirectly affected by ICT use. Workers in digitalised environments using ICT devices are more likely to report virtual presenteeism – working from home when one is sick, although not to an extent that precludes working – which can impair performance, impede recovery and affect long-term health.
• These risks have been demonstrated during the expansion of working from home as a result of the COVID-19 pandemic. In some countries, there is evidence that the school closures and strict lockdowns implemented in response to the pandemic have contributed to people working in their free time and working longer hours.

Concluding commentary

TICTM arrangements (and to some extent other digitisation technologies) are in many ways advantageous for workers compared with the traditional arrangement of being based in a single workplace outside the home. However, there can be disadvantages. They can lead to longer working hours, the blurring of boundaries between work and private life, and increased intensity of work. Workers generally experience better working conditions and outcomes when TICTM is performed occasionally. In TICTM arrangements, workers are also more likely to engage in a new phenomenon enabled by digitalisation: virtual presenteeism.

Depending on how it is implemented, TICTM can improve or undermine the work–life balance of workers. The implementation of regulations to improve work–life balance in the context of flexible working using ICT varies across Europe. Only a few countries have adopted legislation that addresses the right to disconnect.

Some of the policy-related implications are described below.

• There are differences in how TICTM is performed in practice, which should be taken into consideration in policymaking. Intensive TICTM should be limited, for instance, because it has a negative impact on workers.
• Systems of monitoring and control should be designed to give TICTM workers real autonomy, to ensure that data are used appropriately and to prevent working time patterns from damaging workers’ health and well-being.
• TICTM should be promoted as a way to improve work–life balance (for example, through the incorporation into law of the Work–Life Balance Directive).
• There is also a need to assess whether the present Working Time Directive and the European framework agreement on telework are sufficient in terms of the implementation of provisions for these workers (and their protection), including measures to record, monitor and control their working time.
• The regulation of TICTM – for example, through legislation on the right to disconnect – may be the only way to curb the trend towards a work culture characterised by self-imposed work intensity, project-based work, work paid according to performance, and constant availability. Collective bargaining and social dialogue should play a role in the design and implementation of such initiatives.

Platform work is a form of employment in which organisations or individuals use an online platform to access other organisations or individuals to solve problems or provide services in exchange for payment.

Introduction

The organisation of working time in some types of platform work, notably online platform work, has similarities with the organisation of working time in TICTM, in relation to both advantages (flexibility) and disadvantages (potential blurring of boundaries between work and home life).

In general, online platform work activities can be carried out at any time and from anywhere. Platform workers can therefore potentially enjoy a high degree of working time flexibility, but this depends on the type of platform and activity; some platform workers may be requested to work at short notice. Moreover, the configuration of algorithms can predetermine the organisation of working time, leaving little room for autonomy. This holds particularly true for low-skilled, small-scale tasks mediated through platforms.

One key characteristic that distinguishes the working time duration of platform workers is whether or not this work represents their main paid job. For the majority of platform workers, the activities carried out through platforms represent additional income and not their main paid job, and therefore the time spent working through platforms is normally less than the time spent working by employees in standard full-time employment. Surveys suggest that the majority of active workers perform platform work for fewer than 30 hours a week. The Joint Research Centre’s Collaborative Economy (COLLEEM) survey reported that less than 6% of platform workers spend a significant amount of time (at least 10 hours per week) on platform work.

In any case, because of the diversity of activities carried out on platforms, it is difficult to generalise about their impacts on the duration and organisation of working time.

Opportunities

In general, platform work provides high levels of flexibility and autonomy, which are ‘partly enabled by the Tayloristic breakdown of what were once occupations into their smallest possible components’ (Schmidt, 2017a), as jobs are turned into projects, tasks or even microtasks.

One of the main reasons cited by workers for engaging in platform work is the flexibility it provides, in terms of the selection of tasks, working time, workplace, amount of work and work organisation. Some authors suggest that the main motivation for joining a platform, besides the monetary compensation, is the combination of flexibility and personal control. This includes ‘the possibility to set one’s own schedule, select jobs and negotiate rates’ (De Groen and Maselli, 2016).

Risks

However, these higher levels of flexibility and autonomy are not always available. In some types of platform work, workers may be required to work at very short notice and to tight deadlines (for example, Uber drivers in surge areas). Some platforms also enforce a rapid pace of work without breaks. Even when platform workers set themselves the goal of limiting their working hours, they are sometimes unable to stop. This is because of the overabundance or shortage of work on the one hand, and the fear of being penalised by the platform or clients for being unavailable on the other. Some studies also report platform workers being on call at all times, interrupting social engagements and personal time to work, or waking up in the middle of the night to bid on projects or ‘grab tasks’ when they are posted in different time zones.

The levels of flexibility and autonomy are also affected by the dependency of workers on the platforms. For workers whose main or only source of income is their platform work, scheduling and time management are less under their control and much more a function of the availability of tasks and the need to earn an income.

These situations have implications for work–life balance. Several authors refer to the blurring of boundaries between private and professional life, especially because of the need or perceived need to be permanently available. This can have an impact on work–life balance, as workers may have difficulties in focusing on their work, for example because of family-related interruptions, or on their free time, as there is pressure to be on standby in order not to miss any upcoming tasks.

Moreover, work on many platforms is dependent on algorithms, which can exert a lot of control over workers’ schedules and work processes. For example, in the personal transport sector, when it comes to deciding what tasks to accept or decline, passenger assignment through algorithmic systems inhibits the autonomy of workers. In this way, platform work limits the scope for working time autonomy and flexibility. Even when the decision to work or not work is formally up to the drivers, algorithms, ratings and rewards systems are aimed at exerting control over them and the quality of services provided, and therefore their organisation of working time.

In relation to working time duration, there are risks of platform workers working longer hours if the client underestimates the workload involved in specific work requests, and of being unpaid for working time spent searching for tasks and taking part in contests for work.

Finally, some platform workers are subjected to high levels of unpredictability. Interviews with UK platform workers showed that some had experienced psychosocial health hazards connected to long working hours, ‘including long and unpredictable waiting periods’ (Huws et al, 2017). The unpredictability is more generally related to work opportunities and earnings. Consequently, working time can be unpredictable and workers do not always have control over it; some may be required to work unsocial hours and to be available at short notice.

Concluding commentary

Overall, research finds that platform workers work a rather limited number of hours – with the exception of those who perform platform work as their main job (which tends to result in long working hours). Some authors have highlighted that platform work involves antisocial working hours and unpaid working time (waiting periods and time spent searching and bidding for tasks). This influences workers’ work–life balance, as does their perception that they need to be always available.

Most of the information available indicates that platform work has good potential regarding flexibility, autonomy and control, as it is – at least in theory – up to workers to decide which tasks to bid for and, if successful, when, where and how to perform these tasks. In practice, the research finds that platform work has some limitations, imposed by the availability of work and the platforms’ terms and conditions (including ratings and algorithmic task assignment).

In conclusion, it is debatable if platform work can provide opportunities for greater working time flexibility and autonomy (and therefore for a better work–life balance). The opportunities and risks associated with platform work are related to the type of activity carried out, the need to be permanently available and the pressure to work longer hours because of low hourly wages, and the organisation of working time, which is determined by the algorithms used.

Related material

References

Eurofound sources

Eurofound (2015), New forms of employment , Publications Office of the European Union, Luxembourg.

Eurofound (2017), Automation of work: Literature review , Eurofound working paper, Dublin.

Eurofound (2017), Coordination by platforms: Literature review , Eurofound working paper, Dublin.

Eurofound (2017), Digitisation of processes: Literature review , Eurofound working paper, Dublin.

Eurofound (2018), Automation, digitisation and platforms: Implications for work and employment , Publications Office of the European Union, Luxembourg.

Eurofound (2018), Platform work: Types and implications for work and employment – Literature review , Eurofound working paper, Dublin.

Eurofound (2019), Further exploring the working conditions of ICT-based mobile workers and home-based teleworkers , Eurofound working paper, Dublin.

Eurofound (2019), Mapping the contours of the platform economy , Eurofound working paper, Dublin.

Eurofound (2020), Game-changing technologies: Transforming production and employment in Europ e , Publications Office of the European Union, Luxembourg.

Eurofound (2020), Right to disconnect in the 27 EU Member States , Eurofound working paper, Dublin.

Eurofound (2020), Telework and ICT-based mobile work: Flexible working in the digital age , Publications Office of the European Union, Luxembourg.

Eurofound (2021), Working time , web page, accessed 17 September 2021.

Eurofound and the International Labour Office (2017), Working anytime, anywhere: The effects on the world of work , Publications Office of the European Union, Luxembourg, and the International Labour Office, Geneva.

Other sources

Berg, J. (2016), Income security in the on-demand economy: Findings and policy lessons from a survey of crowdworkers , Conditions of Work and Employment Series No. 74, International Labour Organization, Geneva.

Brawley, A. and Pury, C. (2016), ‘Work experiences on MTurk: Job satisfaction, turnover, and information sharing’, Computers in Human Behavior, Vol. 54, pp. 531–546.

De Groen, W. P. and Maselli, I. (2016), The impact of the collaborative economy on the labour market , Centre for European Policy Studies, Brussels.

De Standaard (2017), ‘ Deliveroo past verloningssysteem aan maar ontkent dat fietskoeriers verzekering verliezen ’ [‘Deliveroo applies compensation system but denies that cyclists lose insurance’], 25 October.

Directorate-General for Communications Networks, Content and Technology (2019), Digital Economy and Society Index , web page, accessed 24 August 2021.

Eurostat (2019), European Union Labour Force Survey (EU LFS) , web page, accessed 24 August 2021.

Huws, U. (2016), A review on the future of work: Online labour exchanges or crowdsourcing , web page, accessed 24 August 2021.

Huws, U., Spencer, N. H., Syrdal, D. S. and Holts, K. (2017), Work in the European gig economy: Research results from the UK, Sweden, Germany, Austria, the Netherlands, Switzerland and Italy , Foundation for European Progressive Studies and UNI Europa, Brussels, and University of Hertfordshire, Hatfield, UK.

Jiang, L., Wagner, C. and Nardi, B. (2015), ‘Not just in it for the money: A qualitative investigation of workers’ perceived benefits of micro-task crowdsourcing’, Proceedings of the Annual Hawaii International Conference on System Sciences , pp. 773–782.

Lee, M. K., Kusbit, D., Metsky, E. and Dabbish, L. (2015), ‘Working with machines: The impact of algorithmic and data-driven management on human workers’, Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems , pp. 1603–1612.

Lehdonvirta, V. (2018), ‘Flexibility in the gig economy: Managing time on three online piecework platforms’, New Technology, Work and Employment, Vol. 33, No. 1, pp. 13–29.

Martin, D., O’Neill, J., Gupta, N. and Hanrahan, B. (2016), ‘Turking in a global labour market’, Computer Supported Cooperative Work, Vol. 25, No. 1, pp. 39–77.

Pesole, A., Urzí Brancati, M. C., Fernández-Macías, E., Biagi, F. and González Vázquez, I. (2018), Platform workers in Europe: Evidence from the COLLEEM survey , Publications Office of the European Union, Luxembourg.

Rosenblat, A. and Stark, L. (2015), ‘Uber’s drivers: Information asymmetries and control in dynamic work’, SSRN Electronic Journal, doi:10.2139/ssrn.2686227.

Rosenblat, A. and Stark, L. (2016), ‘ Algorithmic labor and information asymmetries: A case study of Uber’s drivers ’, International Journal of Communication, Vol. 10, pp. 3758–3784.

Schmidt, F. A. (2017a), Crowd design: From tools for empowerment to platform capitalism, Birkhäuser, Berlin and Boston.

Schmidt, F. A. (2017b), Arbeitsmärkte in der Plattform-ökonomie – Zur Funktionsweise und den Herausforderungen von Crowdwork und Gigwork , Friedrich Ebert Stiftung, Bonn.

Smith, A. (2016), Shared, collaborative and on demand: The new digital economy , Pew Research Center, Washington, DC.

Valenduc, G. and Vendramin, P. (2016), Work in the digital economy: Sorting the old from the new , working paper, European Trade Union Institute, Brussels.

Image © igorstevanovi/Shutterstock