BEU, Alm.del - 2021-22 - Bilag 139: Orientering om ny NFA-artikel om unges arbejdsmiljø på digitale platforme, fra beskæftigelsesministeren

Beskæftigelsesudvalget 2021-22
BEU Alm.del Bilag 139
Offentligt

Safety Science 149 (2022) 105674

Contents lists available at

ScienceDirect

Safety Science

journal homepage:

www.elsevier.com/locate/safety

Who takes care of safety and health among young workers?

Responsibilization of OSH in the platform economy

Mette Lykke Nielsen

a, *

, Cæcilie Sloth Laursen

, Johnny Dyreborg

Aalborg University, A.C Meyers Vænge 15, 2450 København SV, Copenhagen, Denmark

IT University of Copenhagen, Rued Langgaards Vej 7, 2300 København, Denmark

The National Research Centre for the Working Environment, Lersø Parkall

105, 2100 København

Ø,

Denmark

A R T I C L E I N F O

Keywords:

Employment relations

Digital platforms

Gig economy

Prevention

Protective gaps

Migration toward accidents

A B S T R A C T

This study explores how young workers experience employment relations and responsibility for Occupational

Safety and Health (OSH) in the platform economy. The study is based on 29 qualitative interviews with young

Nordic workers (age 18–30) who find work through digital labour platforms and social media platforms. The

European Agency for Safety and Health at Work asserts that the placement of responsibilities for OSH in the

platform economy is challenged by the unclear categorisations of employers, employees and self-employed, and

that existing labour law and OSH legislation might be inapplicable. Even though most platforms position workers

as self-employed, the study shows that the young workers rarely experience themselves as being self-employed

and assume that the platforms take care of OSH. When operating in this grey zone, the young workers risk being

left without protection and societal resources for improving their OSH. Rasmussen’s model ‘migration towards

the boundary of unacceptable safety performance’ is used to discuss how work activities in platform work is

driven by strong cost and effort gradients, while, the counter gradient, in terms of OSH systems, at the same time

is very weak or completely absent.

1. Introduction

Digital technologies, such as artificial intelligence (AI), advanced

robotics, exoskeletons, mobile devices and online platforms are devel-

oping at a fast pace of change. Such new technologies are paving the way

for new modes of organizing and controlling the nature of work (Coyle,

2017),

which adds complexity to the work processes, and thus posing

new challenges for the management of risks at the workplace (Ras-

mussen, 1997).

Moreover, these digital technologies have enabled the

emergence of various digital platforms, in what is termed the platform

economy or the gig economy

. Digital labour platforms are commercial

online platforms (such as Handyhand and Freelancer), which mediate

services and tasks that can be delivered either locally or remotely (Huws,

2015).

In addition, social media platforms (such as YouTube and

Twitch) have enabled new types of work (Nielsen

et al. 2019; Abidin et

al 2020).

These developments in technology and new ways of organising

work are transforming employment relations and the possibilities for

prevention of occupational safety and health (OSH) problems among

workers, particularly young workers, who are the focus of this paper,

since they as a group is more prevalent on the digital platforms (Ilsøe

Madsen, 2017; Popescu et al., 2018).

Research in this field is still on an

early stage, and knowledge about the connection between platform

work and OSH is not very well covered, and empirical studies in

particular are lacking.

Current understandings of work and employment relations are

challenged by the continuous emergence of new forms of work and work

arrangement, which are often deemed precarious (Antonucci

et al.,

2014; Casas-Cort

s, 2014; Kalleberg, 2009; MacDonald

Giazitzoglu,

2019). Scheuer (2017)

proposes that these new forms of work:

[have] given rise to a somewhat blurred distinction between employee and

the self-employed, which may have led to an increase in the number of

people in this ‘grey zone’ (…). Employers may, in some cases, benefit from

* Corresponding author.

E-mail address:

[email protected]

(M.L. Nielsen).

The term ‘platform economy’ is often preferred by labour unions and governmental institutions, while the term ‘gig economy’ is commonly used in research

articles. We use the term ‘platform economy’ because the term underscores the centrality of digital platforms in this new economy. In addition, we find that the term

‘gig economy’ excludes other types of new work arrangements which cannot be defined as gig work, such as making a living by being an e-sports gamer.

https://doi.org/10.1016/j.ssci.2022.105674

Received 21 December 2020; Received in revised form 15 December 2021; Accepted 6 January 2022

BEU, Alm.del - 2021-22 - Bilag 139: Orientering om ny NFA-artikel om unges arbejdsmiljø på digitale platforme, fra beskæftigelsesministeren

M.L. Nielsen et al.

Safety Science 149 (2022) 105674

applying this terms of employment (…) as it significantly reduces

employer obligations’’ (Scheuer,

2017, p. 85,

translated by the authors).

Workers in the platform economy can be said to comprise a subgroup

within this ‘grey zone’ between being an employee and a self-employed

person (Huws

2015; Ilsøe 2017; Jesnes et al., 2016; Manyika et al.,

2016)

because categories such as employer, employee, and self-

employed are not easily applicable in the platform economy (Garben,

2017; Standing, 2016; Weber, 2018).

Digital labour platforms often refer

to platform workers as entrepreneurs and independent contractors

(Prassl,

2018; Todolí-Signes, 2017; Zwick, 2018),

which leaves them

without employee protection, because the responsibilities for OSH are

relegated to the workers on the platforms.

Scheuer (2017)

points out

that this ‘in-between’ group is often left without protection because they

are not categorised as employees. Also, as noted by

Nielsen et al. 2019,

these types of work are not in the scope of the labour inspection au-

thorities, and are not receiving the attention and resources from occu-

pational safety and health professionals. Thus, these new employment

relationships create ‘protective gaps’ for this group of workers, including

gaps in employment rights, gaps in social protection, gaps in represen-

tation, and gaps in enforcement of rights and OSH (Grimshaw

et al.,

2016; Nielsen et al. 2019).

Several European Member States acknowledge the existence of a

third group on the labour market that operates in the ‘grey zone’ be-

tween employee and self-employed, and the terms solo self-employed,

bogus self-employed, and ‘economically dependent self-employed

worker’ are some of the terms used to describe them (Narvaiza,

2011).

This phenomenon is not restricted to the platform economy, but is also

seen in the construction sector (Arnholtz

et al., 2018; Vershinina et al.,

2018)

and in the German meat industry, where subcontracting has been

introduced in slaughterhouses (Wagner

Refslund, 2016).

One of the most well-known labour platforms is the app-based taxi

service Uber. Uber was one of the first foreign labour platforms to come

to Denmark in 2014. Shortly after their arrival, the Danish Transport

Authority reported Uber to the police. Together with the Danish Taxi

Council, they claimed that Uber was an illegal taxi service because the

company violated the Danish Passenger Transport Act by driving in non-

approved cars and using non-certified drivers. Later that year, prose-

cutors filed charges against Uber. In 2017, Uber announced that they

closed their Danish branch due to a new political agreement on the Taxi

Act (Kristiansen

Andersen, 2017).

This did not prevent other digital

labour platforms from establishing their businesses in Denmark. Since

2014, a number of labour platforms have emerged (Ilsøe

Madsen,

2021).

The issue of digital management and control of work in the digital

labor market is currently under debate, because these changes lead to a

lack of clarity as to who is responsible for the working environment in

the digital labor markets.

It is typically young people who are engaged in platform work

(Balaram

et al., 2017; Garben, 2017; Ilsøe

Madsen, 2017; Popescu

et al., 2018),

and the young workers are known to have higher risks of

work injuries compared to their older colleges (Dyreborg

et al. 2019;

Nielsen et al. 2018; Nielsen et al. 2013; Nielsen et al. 2017, European

Agency for Safety and Health at Work, 2007; Garben, 2017).

For this

reason, it is paramount to examine young workers’ perceptions of their

employment relations

when they work via digital platforms and what

happens with the responsibility for OSH. This paper thus examines how

young workers on digital platforms experience their employment re-

lationships and how responsibility for OSH is affected. We also provide a

discussion of the possible consequences for health and safety at work.

2. Health and safety on digital platforms

There is still limited research that specifically focuses on OSH among

persons working via digital labour platforms and social media platforms.

Work via online digital platforms resembles temporary and ‘nonstan-

dard’ employment forms, which are characterized by a poor working

environment, increased risks, and weakened social conditions (Garben,

2017; Nielsen et al., 2019).

Offline work, mediated through digital la-

bour platforms, often encounters the same kinds of OSH risks as seen

when the same work tasks are performed within standard employment

forms, e.g., lack of safety training, lack of safety equipment and lack of

breaks (Jesnes

et al. 2016; Huws 2015).

Previous research also suggests

that the risks of strain injuries and work accidents are very much

dependent on whether work is done online or offline (Huws,

2015;

Jesnes et al., 2016; Nielsen et al., 2020).

However, digitalization adds a number of new mechanisms, which

can reinforce some of these traditional OSH risks (Huws,

2015; Prassl,

2018).

In continuation of this, previous literature has noted that digi-

talisation can intensify work by increasing the pace, surveillance, and

working hours, and that the technology contributes to a boundaryless

working life (Gregg,

2011; Nielsen et al., 2019; Prassl, 2018; Samant,

2019; Wood et al., 2018).

A number of studies suggest that risks in

digitally mediated work are linked to the platforms’ use of ‘algorithmic

management’ (Lee

et al., 2015; Mohlmann

Zalmanson, 2017; Nielsen

et al., 2020).

Multiple studies of ride-hailing platforms (Chen,

2018; Lee et al.,

2015;

2017;

Mohlmann

Zalmanso, 2017; Rosenblat

Stark, 2016)

find that the non-transparency of the algorithms used on the platforms

causes drivers to feel a sense of unfairness and a loss of autonomy.

Ratings also serve as a control mechanism on the platforms (Gandini,

2019; Raval

Dourish, 2016; Wood et al., 2018)

and several researchers

assert that evaluation through rating systems can cause stress among

workers (Garben,

2017; Huws, 2015; Jesnes et al., 2016),

and have

negative psychosocial impacts (Lee

et al., 2015).

Additionally, workers

on digital labour platforms may have difficulty navigating or rejecting

transgressive behaviour from customers because they fear poor ratings,

and thereby their opportunity to maintain an income (Chen,

2018;

Moore, 2018; Wood et al., 2018).

Finally, digitized management

methods can lead to detailed monitoring and ‘cyberbullying’ (Moore,

2018).

Online workers, such as influencers on social media platforms are

particularly vulnerable to being targets and victims of cyberbullying

(Abidin,

2019).

Introduction of safety learning routines and instruction is a classical

approach to prevent occupational accidents (Hale,

1984; Kjell

n, 2000),

including work injuries among young employees (Danish

Working

Environment Authority, 2015).

However, since most workers in the

platform economy are considered self-employed, no one, besides the

workers themselves, has the formal responsibility for safety introduction

and learning. Therefore, these workers receive very little or no intro-

duction to work and safety training (Nielsen

et al., 2019; Nielsen et al.,

2020; Nielsen

Laursen, 2020).

However, the risk of work-related ac-

cidents and injuries might be more prevalent in platform work due to the

temporary and task defined nature of the work, which is known to

correlate with the risk of injuries (Garben,

2017; Nielsen et al., 2018).

Additionally, the responsibilization of the platform workers to take care

of the OSH problems themselves in combination with the absence of

authorities and resources for the prevention of OSH problems among

these young workers, can further increase their risks at work.

3. Method

3.1. Data collection

Rather than using the term employment relationship as a reference to a

relationship of dependence between an employer and employee, we use the

term more broadly to refer to people in employment (opposed to unemployed)

which encompasses employees, self-employed, solo self-employed and so forth.

The paper builds on 29 qualitative interviews with young workers

between 18 and 30 years old in the Nordic countries who find work in

the digital labour market. The workers on social media platforms are

BEU, Alm.del - 2021-22 - Bilag 139: Orientering om ny NFA-artikel om unges arbejdsmiljø på digitale platforme, fra beskæftigelsesministeren

M.L. Nielsen et al.

Safety Science 149 (2022) 105674

Table 1

Overview of participants and experienced OSH risks.

Type of platform

Online digital labour platforms (WorkSome,

’Din

tekst

forfatter’, Freelancer)

Interviews

Age and sex

2 women

4 men

21–28 years of

age

3 women

5 men

19–25 years of

age

Experiences of OSH risks

•

Lack of colleagues, training and support

•

Borders between work and private life are highly blurred (availability 24/7)

Dependence on non-transparent algorithms and ratings

•

Reported neck and back pains, stress-related illness

•

Lack of colleagues, training and support

Borders between work and private life are highly blurred (availability 24/7)

Insecurity and- or physical hazards related to working in foreign homes

Strains derived from the affective work- and relationship with customers.

Dependence on non-transparent algorithms and ratings

Risks of accidents depended on the work performed

Long working hours

Borders between work and private life highly blurred

Missing colleges (isolation)

Stressful to maintain affective relationships with followers and viewers

Hard tone, hatred, or even online threats

Stress and sleeping difficulties

Neck and back pain

Location based digital labour platforms (Wolt, HandyHand,

Chabber, Care, Dogley, Hilfr)

Social media platorms (YouTube, Facebook, Instagram and

personal blogs)

5 woman

6 men

20–32 years of

age

Professional gamers (livestreaming on Twitch)

1 woman

3 men

21–27 years of

age

Long working hours

Borders between work and private life are highly blurred.

Missing colleges (isolation)

Harsh tongue or hate speech

Stressful continuously to maintain affective relationships with followers and

viewers

•

Irregular, evening and night work (adjusting to time zones for tournaments and

when subscribers are active)

•

Demanding physically and psychosocial work tasks with high intensity.

•

Neck and shoulder pain, sleeping disorders and stress-related symptoms and

illnesses.

Total

11 women 18

men

19–32 years of

age

gamers, YouTubers, and Influencers, whereas workers on digital labour

platforms are cleaners, dog walkers, couriers, waiters, babysitters and

the like (Table

1).

The data were collected via two research projects

during 2016–2019. In the first place, the platform workers were

recruited through open invitations to participate in the project via the

digital labour platforms’ own newsletters and Facebook groups, but this

approach was unsuccessful. We therefore contacted the young platform

workers via other channels: Directly through their profile on the digital

platforms; via their contact information on the platform; via Facebook

Messenger; via the researchers’ network; or contacted if we saw them in

the street with a logo or company equipment indicating affiliation with a

platform. Fourteen out of these 35 young people agreed to participate.

Young workers on social media platforms were recruited using the

young people’s own publicly available online contact information,

through contact with influencer agencies, and through the researchers’

networks. Fifteen out of these 22 young people agreed to participate.

The interviews were conducted at the time and place that suited the

worker and lasted 1–1.5 h. The young workers were asked about their

life situation, work routines, work organisation, pay, social and eco-

nomic risks, and OSH risks, including whom they would contact if

injured. Quotes are translated from Danish. The participants’ personal

information is handled in accordance with GDPR, and thus original

names and personal identifiers are pseudonymised.

In addition to the interviews, we collected the Terms and Conditions

from the different digital labour platforms where the young workers

were signed up. The Terms and Conditions can be considered official

documents from the platforms (Bryman,

2016),

and they provided us

with insights into how they define the users of the platform, their re-

strictions, and the division of responsibilities. Finally, the project has

been conducted as a collaborative process with inspiration from the

Canadian Knowledge-Transfer-Exchange Model (KTE model) (Van

Eerd,

2017).

Thus, all phases of the project have been developed in close

dialogue with stakeholders. This was done during six workshops, where

representatives from Unions, representatives for platform owners,

Influencer agencies, and the OSH authorities participated and contrib-

uted to the discussion and interpretation of the study. This provided the

project with unique insights into the work of the various actors in the

platform economy.

3.2. Analytical approach

All interviews were transcribed verbatim and thematically coded

(Braun

Clarke, 2006)

in NVivo 12. The thematic codes chosen for

closer inquiry for this particular paper were ‘physical and psychosocial

work environment’, ‘work conditions’, and ‘help and support in the

work’. We read the coded excerpts from all 29 interviews, and within

these codes, we looked for the young workers’ narratives of the

employment relationships in their work on digital platforms. The seven

cases represented below were chosen because they represent three

different forms of digital platforms and because the interview quotations

provided a rich and detailed insight into the workers own experiences of

their employment relations. The Terms and Conditions from the

different digital platforms included in the analysis were read with a

specific focus on the platforms’ positioning of themselves, the workers,

and the customers.

Rather than examining OSH risks in the Danish platform economy,

this study seeks to unfold the ‘grey zone’ of employment relations and to

outline the various risks and responsibilities for OSH on digital plat-

forms. The risks found in this material are reported in

Table 1

and

constitutes a follow-up on analyses presented elsewhere (Nielsen

et al.,

2019; Nielsen et al., 2020).

BEU, Alm.del - 2021-22 - Bilag 139: Orientering om ny NFA-artikel om unges arbejdsmiljø på digitale platforme, fra beskæftigelsesministeren

M.L. Nielsen et al.

Safety Science 149 (2022) 105674

3.3. Theoretical approach to risk and safety at work

The organisation of work via online platforms adds complexity and

increases the opaqueness of the responsibility for occupational safety

and health in platform work. In this way, platform work represents a

business model that entails fragmentation of workplaces and disruption

of OSH responsibilities, as

Weil (2019)

refers to as ‘fissured workplaces’,

which involves companies outsourcing what is not their central business

activities, while they still maintain in tight control of the outcomes of

those subsidiary activities, by introducing incentive systems supported

by increasingly sophisticated software (Weil,

2019),

such as algorithmic

management systems (Laursen,

Nielsen

Dyreborg, 2021).

Rasmussen’s (1997)

dynamic safety model, i.e., migration toward

the boundary of unacceptable safety performance, addresses these dy-

namic aspects of safety in a complex work system (Waterson

et al.,

2017).

The model is based on the principle of functional abstraction,

which makes it suitable for many types of work systems, including

platform work, and the situational mechanisms that may lead to injuries

(Cook

Rasmussen, 2005; Le Coze, 2015).

Rasmussen’s ‘functional

abstraction’ perspective to model processes involving safety and risk,

implicates attention to how workers are influenced by economic cost

reduction and workload pressures that drive work activities towards the

boundary of unacceptable safety performance. In their media coverage

analysis of the platform economy,

Nilsen et al. (2020)

applied Ras-

mussen’s model in their discussion of factors that could potentially

affect the safety and working conditions of workers in the platform

economy. Inspired by

Nilsen’s et al. (2020)

use of Rasmussen’s model,

we examined the basic mechanisms underlying the safety and health

risks related to the business models shaping the work activities on digital

platforms.

4. Analysis and results

First, we offer a categorisation of different digital platforms. Second,

we outline the employment relations on digital labour platforms and

social media platforms. Finally, we explore the young workers’

narratives of employment relations and OSH responsibility on the two

forms of digital platforms.

4.1. Categorisation of digital platforms

Based on the literature and empirical material, we suggest a cate-

gorisation of the platform economy, as presented in

Fig. 1

below. As can

be seen from the figure, the different platforms in the platform economy

mediate different types of work. This article focuses on the two main

groups, ‘digital work platforms’ and ‘social media platforms’, including

their subcategories.

4.2. Transformed employment relations on digital labour platforms

Digital labour platforms mediate the supply and demand of services

Fig. 2.

Digital labour platforms: A three-sided employment relationship.

Fig. 1.

Different categories of digital platforms in the platform economy.

BEU, Alm.del - 2021-22 - Bilag 139: Orientering om ny NFA-artikel om unges arbejdsmiljø på digitale platforme, fra beskæftigelsesministeren

M.L. Nielsen et al.

Safety Science 149 (2022) 105674

Fig. 3.

Social media platforms: A five-sided employment relationship.

and tasks between workers and customers. Thus, instead of a two-sided

(dyadic) relationship between an employer and an employee, there is a

three-sided (triadic) relationship between a digital platform, those who

work, and the customer who receives the service (Jesnes

et al., 2016;

Kristiansen

Andersen, 2017),

as illustrated in

Fig. 2.

In this way, the

digital labour platforms constitute a new actor that mediates, and by

that also controls, the transaction between worker and customer (Prassl,

2018).

Due to this three-sided relationship, the traditional categorisations of

employer, employee, and self-employed are not easily applied in the

platform economy (Garben,

2017; Standing, 2016).

The European Agency for Safety and Health at Work asserts that the

current legislation is not aligned to the conditions of the platform

economy and suggests that the vague categorisations of employers and

employees on platforms complicate the placement of OHS re-

sponsibilities. As a consequence, the awareness and care for health and

safety regarding work activities fall between these various actors.

Furthermore, the legislation varies across national borders; therefore

workers fall into different categories across member states of the Euro-

pean Union (Garben,

2017).

4.3. Transformed employment relations on social media platforms

Another kind of paid activity that has emerged with the digital rev-

olution is carried out by persons/influencers/gamers earning money by

producing, posting, and streaming content on digital social media

platforms (Abidin,

2019; Abidin et al., 2020).

This includes, for instance,

uploading videos onto YouTube or streaming a Counter-Strike game on

Twitch (Johnson

Woodcock, 2017).

Then there are the people who

view the content, whom we label ‘content consumers’, rather than

customers. Finally, there is the platform itself where the content is being

published. Besides this three-sided relationship, the influencer/gamer

can be affiliated with an influencer agency or club, and the influencer/

gamer can engage in commercial collaborations with companies who,

for instance, pay the influencer/gamer to advertise a product (Abidin

et al., 2020).

In this way, the companies can be perceived as the influ-

encer/gamers’ customers. As illustrated in

Fig. 3,

paid activities on so-

cial media platforms can involve a five-sided relationship.

While the types of work carried out across the two types of platforms

- social media platforms and digital labour platforms - are arguably very

different, they also share common features. The most important simi-

larity is that work activities across the two types of platforms are both

enabled and conditioned by the platforms, which challenges the tradi-

tional labour market legislation and the division of workers into em-

ployees or self-employed (Garben,

2017; Standing, 2016).

Additionally,

while employment relations can be debated, most workers on the two

types of platforms are currently classified as self-employed, even though

the platforms to a very large extent control and manage the task per-

formance. For this reason, we see it as an analytical opportunity to

explore workers’ experiences of their employment relations and re-

sponsibilities for OSH across different types of digital platforms.

4.4. Employment relations and OSH responsibilization on digital labour

platforms

In the following section, three different digital labour platforms are

described. The three platforms were chosen because each of them ex-

emplifies different employment models. The presented example quotes

from the interviews

serve to illustrate the young workers’ different

narratives of how transformed employment relations and responsibili-

zation for OSH are experienced. The first platform, Dogley, is a loca-

tionbased work platform offering dog walking and dog sitting. This

platform positions workers as self-employed. The second platform, Hilfr,

is also a locationbased labour platform offering cleaning services. On

this platform workers can both be freelancers and employed directly by

the platform. Finally, the digital labour platform Chabber functions as a

digital agency for temporary workers and employs all workers as hourly

waged temps. At Chabber, restaurants, events, and private customers

can book freelance waiters, bartenders and kitchen help. Despite the

platforms’ different employment models, the young workers’ narratives

of their employment relationships on the three different platforms have

some similarities.

4.4.1. Dogley: ‘Strange to have an employer who is so invisible’

Ditte is a 24 year old student with a profile on the platform Dogley.

Here she is booked by private customers to walk their dogs. Ditte ex-

plains that to be accepted by the platform, she had to have a job inter-

view by phone.

I received a call from a woman from Dogley. Now that I had applied for

the job, she wanted to know why I wanted to work at Dogley, and she

wanted to hear about my relationship with animals. It was a very

informal, pleasant conversation where she just wanted to know a bit about

me and my motivation. (…) She said that it sounded fine, and a short

while after my profile was created in the system.

Based on the quote, one would assume that Dogley employs Ditte,

since Dogley hosted a job interview. According to Ditte, Dogley is her

employer even though she is aware that it is an untraditional employ-

ment relationship; ‘it is strange to have an employer who is so invisible’

We have not had access to the young influencers’/gamers’ contracts, and for

this reason we do not know the specifics of the contracts.

BEU, Alm.del - 2021-22 - Bilag 139: Orientering om ny NFA-artikel om unges arbejdsmiljø på digitale platforme, fra beskæftigelsesministeren

M.L. Nielsen et al.

Safety Science 149 (2022) 105674

she says. Dogley, on the other hand, makes it explicitly clear in their

terms and conditions that they are not employers and that they have no

part in the agreement between workers and dog owners (Dogley,

2018).

They describe themselves as a platform where dogwalkers can post ad-

verts for services (Dogley,

2018).

However, they take a service fee of

18% from the dogwalkers’ pay for managing the payment transactions

and insurance for the dog. It is important to note that the insurance is for

the dog and not the dog walker. Ditte, however, is of another conviction:

Dogley has an insurance if something happens. So, you can call them,

which I would do. I haven’t thought that much about it though, because I

have never considered that something might happen.

If Ditte is bitten during work, she says she would contact Dogley in

addition to the dog owners because she knows Dogley has insurance. She

is not aware that the insurance does not cover the worker, only the dog.

She perceives her work as a dog walker to be part of an employ-

er–employee relationship.

4.4.2. Hilfr: ‘Being far away if you need help’

Hans is 22 years old, and besides his studies he works via the plat-

form Hilfr, which offers cleaning services. Here Hans has a profile, and

customers can book him for cleaning jobs by selecting a day and time.

Hans can then accept, reject, or suggest another day for the job. Hilfr

takes a 10% commission for each booking. Workers on the platform are

called ‘Hilfrs’. There are two types of Hilfrs, Freelance Hilfrs and Super

Hilfrs (Hilfr,

n.d.).

If a worker has had 100 working hours via the plat-

form, they can (if they so wish) become Super Hilfrs, who are then hired

by Hilfr and covered by a collective agreement between the United

Federation of Workers in Denmark [3F] and Hilfr (3F

Hilfr, 2018).

Hans is a Freelance Hilfr; he is covered by insurance from the platform,

but is considered self-employed by the platform. Hans, however, does

not consider himself self-employed:

Interviewer:

Do you consider yourself self-employed?

Hans:

No, not really. I don’t really know… No, I actually look at it as if I

have a job at Hilfr.

Hans acknowledges that the work is different from a traditional job

because there is no manager giving him instructions, but he says that the

distance and having an online platform as a mediator of the work

sometimes leaves him feeling quite alone in the work. If a problem

occurred while Hans was at a cleaning gig, he says he would contact

Hilfr-support, either through the chatfunction on the platform’s website

or their Facebook page. He has received no training or advice on correct

work postures (ergonomics) or other safety issues related to the work.

This becomes his individual responsibility, which he is aware of: ‘There

is no supervisor telling you

“try

to do this then you will last a couple of

hours longer”’.

If we recall

Fig. 2

and the three-sited employment relationship, we

see that for Super Hilfrs the platform takes employer responsibility. This

is not the case for Freelance Hilfrs like Hans. However, the platform does

insure the workers, and workers can contact them if problems occur.

Another question concerns the customers’ responsibilities in this rela-

tionship. The customers provide the cleaning equipment and detergents,

and Hans has experienced that sometimes the cleaning equipment is

suboptimal; for instance, the mop has been too short for him, which led

to him experiencing back pain. However, he is in doubt about what to do

in this situation.

Hans:

I don’t know if I can allow myself to say,

“You

have some really

shitty equipment”.

Interviewer:

What is holding you back from doing something?

Hans:

I guess it is the fear that then people will just find somebody else (to

do the job).

Even though it is the customers’ responsibility to provide the

equipment, Hans is hesitant to point it out if the equipment could be

better to improve his work environment. He fears that the customers will

choose another Hilfr to do the cleaning next time. Hans is dependent on

the customers being happy with his cleaning services: the customers

provide ratings and reviews after a cleaning gig, and this influences the

likelihood of getting booked for new cleaning jobs. This is an example of

how individualistic conditions in the platform economy (Huws,

2015;

Jesnes et al., 2016; Prassl, 2018)

can cause the worker to neglect OSH in

order to maintain an income.

4.4.3. Chabber: ‘What are our rights?’

Charlotte is 24 years old and is currently having a sabbatical year.

She takes on several jobs, and one of them is as a freelance waitress on

the platform Chabber. Chabber employs workers as temporary workers

paid by the hour, but it is the customers/companies hiring the workers

for a gig who are labelled as ‘employers’ in Chabber’s terms and con-

ditions. The customers/companies are required to ‘ensure that the em-

ployees are offered the same economic conditions, including salary,

retirement and holiday pay, and the same facilities and benefits as the

permanent employees.’ (Chabber,

2019).

The customer determines the

duration of a job and the hourly wage but, as stated, the customer must

comply with existing collective agreements. Chabber pays workers their

salary monthly and deducts taxes, taking a fee of 4.81 Euro for each shift

a worker has (Chabber,

n.d.-a).

Thus, in the three-sided employment

relationship model, both the platform and the customer seem to take on

some responsibility in the employment arrangement; paying the worker

income after tax, reporting tax to the tax authorities, and protecting the

worker in accordance with collective agreements and existing

legislation.

To work via the Chabber platform, an individual creates a profile and

applies for posted jobs, and the customer then selects which workers

they want. The customer and the worker then engage in an Employment

Agreement. After completion of a job, the worker and customer rate

each other. Charlotte tells how during one shift the company altered the

agreement:

Charlotte:

At one time, we were 24 waiters. It was a shift from

17:00–00:30, and because we had been so quick to clean up after the

event, we were already done by 22:00. [The company] had just sometime

before said that we would get paid for the duration we were booked for,

even if we were done a bit earlier. Then [the company] said to us:

“You

will get pay till 22:30, but then you can just go home”. It was strange

because they had just said we would get paid for two hours more. People

stood completely quiet and didn’t really know what to say. Because can

they do that? We don’t know. What are our rights? (…) That was a

moment where you as a Chabber were in doubt of your rights.

Charlotte feels cheated by the company, but is also in doubt of her

BEU, Alm.del - 2021-22 - Bilag 139: Orientering om ny NFA-artikel om unges arbejdsmiljø på digitale platforme, fra beskæftigelsesministeren

M.L. Nielsen et al.

Safety Science 149 (2022) 105674

rights. According to Chabber’s FAQ, Chabber seemingly takes the side of

the customer as they allow them to alter the duration of the shifts

(Chabber,

n.d.-b).

Charlotte perceives Chabber as her employer, but

within the three-sided employment relationship, her rights and bargai-

ning power seem restricted by the platform. Chabber caters to the cus-

tomers and allows them to alter the conditions of the employment

agreements, such as cancelling or shortening a shift. Thus, Chabber

negotiates and determines the agreement for Charlotte’s work, which

shows that Chabber takes on obligations and responsibilities usually

taken on by an employer, which confuses the relationship between the

customers employing Charlotte and Chabber. Charlotte is often unsure

about her rights; she is, for instance, also in doubt whether she is insured

by Chabber. When asked what she would do if she was injured while

working, she says:

I don’t know. I don’t know if you are particularly insured in that matter. I

guess if I broke my leg on a Chabber shift I would investigate what I could

do. But I am not exactly the most insured person.

Charlotte does not know what she would do if she was injured, and

she does not have her own accident insurance. However, in this case,

Charlotte is covered by a Chabber occupational injury insurance, in case

of an accident at the workplace.

In all three cases, the workers perceive the platform as their

employer, even when the platforms take no employer responsibility. The

placement of OSH responsibility is experienced as unclear and distrib-

uted between all three actors in the employment relationships. This

makes it a grey zone, where the workers are left to navigate on their

own, which poses a risk.

4.5. Employment relations and responsibilization for OSH on social media

platforms

The social media platforms represent an even more complex

employment relationship. In this section, we see how an influencer,

YouTuber and gamer experience their employment relations and how

they perceive the responsibility for OSH in their work.

4.5.1. Influencer: ‘In a way, I feel as if I am employed’

Annette is 26 years old and works full-time with social media. She

has her own blog, and she also posts content on Snapchat, Facebook, and

Instagram. She is a so-called ‘influencer’ and engages in commercial

partnerships with companies who pay her to be an ambassador for the

company and to promote them, or their products, on her different social

media platforms. Annette characterizes herself as a self-employed.

However, she has a contract with an influencer agency, which she

almost perceives as her employer:

I perceive myself in a way as being self-employed, even if I have a con-

tract. I have a contract at the influencer agency and feel in a way as if I am

employed because I have a desk there. I have colleagues, and I eat my

lunch there. But I do invoice them. And I have other sources of income as

well. And I am not restricted to be there from 08 to 16. Therefore, I

perceive myself as… therefore I am my own boss.

She explains that the agency is her workplace. Four other influencers

are affiliated with the agency, and Annette regards them as her col-

leagues. A telecommunication company has hired Annette through the

agency, and they have entered into a partnership. The

telecommunication company is the agency’s customer. Therefore,

Annette invoices the influencer agency for the work she does for the

company.

Looking at Annette’s employment relationships, we see that they

involve many different actors, as illustrated in

Fig. 3.

Annette produces

content about her life, which she posts on four different platforms where

followers can view her content. Besides this, she has a contract with an

influencer agency that brokers a contract for a commercial partnership

with a company. In this complex employment relationship, she both

positions herself as self-employed and at the same time feels as if she is

employed by the agency.

In a way, I have two managers whom I can ask anything. They call and

nag me about things, for example deadlines, or if they wonder where they

can find my posts.

While the agency feels like an employer for Annette, and at times

behaves like an employer (by checking up on her, as described in the

quote), Annette’s contract is as self-employed:

If I should become sick, then I cannot get sick leave. If I do not do the job, I

do not get paid.

Annette will not get money during sickness, and she has to establish

vacation and pension arrangements for herself. Annette is aware of the

difference between an ordinary employment relationship and her own

employment relation as an independent.

4.5.2. YouTuber:

’I

create YouTube-videos and earn money on ads,

sponsorships, and collaborations’

Jeppe is 20 years old, and for two years, he has worked fulltime as a

YouTuber. He makes YouTube videos and posts them on his own

channel. There are ads in the videos, and in this way, Jeppe earns money

when people view his videos. YouTube takes a cut of the ad money, but

for every 1000 views, Jeppe earns around 2,40–3,10 EUR. Whether

videos are displayed on the front page of YouTube or in suggested videos

is influenced by the algorithms on YouTube (van

Dijck, 2013).

Besides his income from YouTube ads, Jeppe also has sponsorships

and engages in commercial partnerships with companies where he posts

videos on his channels as a part of a campaign. Jeppe is a part of a

YouTuber agency that procures many of the campaigns for him. His

income from YouTube and from the campaigns go through the agency,

and they take a cut of 20% from his income. Jeppe thinks this is a fair

deal:

They are the ones who procure almost 95% of all the campaigns I have

made, so I think it is fair enough that they are paid for this service.

The agency pays Jeppe his income monthly and reports to the tax

authorities how much he has earned. Jeppe himself, however, has to pay

his taxes. Similar to Annett, Jeppe has a five-sided working relationship.

When asked if he can contact the agency about his working envi-

ronment, he says

‘Yes, one hundred per cent’.

But it is clear from the

interview that OSH is not something he has given much thought to,

because it has not been relevant for him.

4.5.3. Gamer: ‘I am hired as a live-streamer’

Rasmus is 23 years old, and since his college graduation, he has

earned money as an e-sports gamer. At one point, he played profes-

sionally in tournaments on a Counter-Strike team, but now he earns his

BEU, Alm.del - 2021-22 - Bilag 139: Orientering om ny NFA-artikel om unges arbejdsmiljø på digitale platforme, fra beskæftigelsesministeren

M.L. Nielsen et al.

Safety Science 149 (2022) 105674

money as a live-streamer and as an expert commentator for tourna-

ments. Rasmus has a contract with an American e-sports club as a

Counter-Strike streamer. The club has different e-sport gamers for

different games affiliated with them. They also have professional teams.

Rasmus explains that his role is as a sort of entertainment figure, which

means when Rasmus live-streams, he represents the club, and he uses

the gaming equipment of the club’s sponsors. Rasmus explains the

relationship in the following way:

The affiliation is not particularly profound or extensive but more like a…

You could say I am self-employed. But I have this contract on the side

which gives me a supplementary income. But you could also say it is a sort

of security and an employment relationship.

Rasmus gets a fixed income each month from the club. If his stream is

doing well, he can get a bonus. Besides this, Rasmus has his own income

from his stream by playing ads; additionally, the viewers can pay a

monthly subscription to be a part of his channel on the platform Twitch.

Viewers can also donate money directly to him. Twitch takes 30% of the

income Rasmus makes via Twitch. The e-sport club has negotiated this

deal for him; the division for smaller players is usually 50%.

In accordance with

Fig. 3,

Rasmus uses the platform Twitch to live-

stream Counter-Strike games. Similar to YouTube, Rasmus gets ad

money when people view his stream. In addition, he gets money when

people subscribe to his channel, and followers can donate money

directly to him. He is affiliated with a club, which pays him monthly to

sponsor the club’s customers by using their gaming equipment. This

constitutes a five-sided relationship. While Rasmus views himself as

independent, the contract with the club is described as an employment

relationship. However, unlike the two previous cases, he has no strong

affiliation with the club, and he would not contact them about issues

related to the working environment.

Rasmus live-streams three to six hours at a time, five to six days a

week. It is often mentally exhausting to live-stream because besides

playing Counter-Strike and being energetic on camera, there is also a

live chat where he corresponds with his followers.

It is stress. But [he sighs]… I feel it is improving. I feel

…I

can control it. I

feel that I have the freedom to take a week off away from the stream. If I

need to do that, I can”.

He tells about periods of stress, but places the responsibility to deal

with it on himself. Additionally, if he takes leave or vacation, it is un-

paid. This demonstrates the individualised conditions of the work and

shows that Rasmus is aware of this personal responsibility.

4.6. Summarising the findings

The analysis shows how some digital labour platforms take no

employer responsibilities, while others offer employment for some

workers. In the third example, the platform Chabber functions as an

agency for hourly-waged temporary workers, and all workers are

employed by the platform; however, the responsibilities are relegated to

the customers who hire the workers for jobs. The different employment

relationships are highlighted in

Table 2

under ‘Digital labour platforms’

illustrate the disparities between different digital platforms and how

they operate, which complicates a clear categorization of roles such as

employer, employee, and self-employed.

The young workers who use social media platforms to earn money

are more explicit about their position as independent workers, and the

influence of the platforms is more subtle (Table

2).

While embedded

algorithms on YouTube and Twitch influence the visibility of the You-

Tubers and Gamers’ channels and videos, they themselves control how

they work, what content they produce, and which collaborations they

accept. Instead it is the influencers’/gamers’ affiliation with an agency

or club that can be experienced as a traditional employment relationship

for the workers.

The analysis shows a difference between social media platforms and

digital labour platforms in relation to how the platforms influence and

control the work. The influence and control of the platform on the work

are more prominent with digital labour platforms: many of the digital

labour platforms establish various conditions for the work e.g., when

using algorithmic management systems (Laursen,

Nielsen

Dyreborg

2021).

5. Discussion

A crucial characteristic of the platform economy is that the different

work arrangements are mediated through digital platforms. This study

found that young workers on digital platforms seem to have a number of

OHS problems, and that the responsibilities for their OSH are unclear,

and at the same time, these young workers have limited resources to deal

with the risk they encounter in their work on digital platforms.

5.1. Protective gaps for young workers on digital platforms

This study suggested a categorization of digital platform work

(Fig.

1),

and the study focused on the two main groups, ‘digital labour

platforms’ and ‘social media platforms’, including their subcategories.

We found that digital labour platforms and social media platforms have

a three-sided and a five-sided employment relationship, respectively.

Thus the platform economy challenges the traditional dyadic relation-

ship between employers and employees (Figs.

2 and 3).

Table 2

Young workers’ Employment Relations in the Platform Economy and the responsibility for OSH.

Type of platform Name of platform/ type Employment relations and responsibility:

of work

reported by platform/agency

Digital work

platforms

Dogley

- Dog walking/sitting

Hilfr

- Cleaning

Chabber

- Waiter

- Cook

- Bartender

Influencer /YouTuber

Gamer

Self-employed

- No insurance

Super Hilfr (employed)

Freelance Hilfr (self-employed)

- Insurance

- Not covered in case of sickness absence

Employed as temporary worker

- Responsibility delegated to customer

Platforms (YouTube): self-employed

Agency: unknown

Twitch: self-employed

Agency: unknown

Employment relations: experienced by young

worker

Platform as employer (job interview)

Platform as employer

OSH responsibility: experienced by

worker

Grey zone: Both individual,

customer and Dogley

Grey zone: Both individual,

costumer, and Hilfr

Grey zone: Unclear

Platform as employer

Social media

platforms

Self-employed, but with an influencer agency who

resembles an employer

Self-employed, but with a professional sports club

who resembles an employer

Grey zone: Individual and agency

responsibility

Individual responsibility

We have not had access to the young influencers’/gamers’ contracts, and for this reason we do not know the specifics of the contracts.

BEU, Alm.del - 2021-22 - Bilag 139: Orientering om ny NFA-artikel om unges arbejdsmiljø på digitale platforme, fra beskæftigelsesministeren

M.L. Nielsen et al.

Safety Science 149 (2022) 105674

Fig. 4.

Migration towards the boundary of unacceptable OSH performance in digital platform work.cost and efforts gradiens push young workers activities towards

the boundary of unaccectable OSH performance Adjust from

Rasmussen’s (1997)

drift to danger’-model.

The study shows that for the young workers on digital platforms, it

remains unclear to them who is taking responsibility for the OHS in their

work, which entails that preventive measures are very weak or entirely

absent. At the same time, these young workers seem to be exposed to

risks at work that resembles temporary and ‘non-standard’ employment

forms, which are characterized by an increased risk of accidents and

other OHS problems, and weakened social conditions (Garben,

2017;

Nielsen et al., 2019).

Consequently, young people working on digital platforms experience

a ‘protective gap’ when it comes to OSH. These results support previous

research in this field (Grimshaw

et al., 2016; Nielsen

Laursen 2020;

Rubery et al., 2018).

5.2. Understanding the basic mechanisms that challenge OSH on digital

platforms

In his seminal paper on risk management in a dynamic society,

Rasmussen (1997)

proposed a descriptive dynamic safety model on

migration towards the boundary of unacceptable safety performance

(accidents).

Rasmussen (1997)

suggests that analyses are not focused on

human errors and violations, but on the mechanisms generating

behaviour in the actual, dynamic work context. Rasmussen’s dynamic

safety model (Fig.

describes a

safe space

safety envelope

for work

activities within three boundaries, i.e., the boundary of economic fail-

ure, boundary of work overload and boundary of unacceptable safety

performance that forms an envelope (Cook

Rasmussen, 2005; Ras-

mussen, 1997).

Working within this envelope is considered a ‘safe

space’, which means the risk of accidents or other OSH problems is

within acceptable limits. However, the work activities are influenced by

gradients that drive task performance away from the boundaries of work

overload and economic failure and towards the boundary of unaccept-

able safety performance (accident or other OSH risks).

While Rasmussen’s model was developed within the realms of high-

risk industries and in a traditional organizational context (Waterson

et al., 2017),

we suggest, in the same vein as

Nilsen et al. (2020),

that the

fundamental drivers or gradients that Rasmussen suggested, would

apply to digital platform work as well.

We have operationalized each of these three gradients by using the

results from this study. Young workers on the digital platforms

experience the cost gradient in terms of the algorithmic management

system, the payment-per-task system and customer ratings, such as

views/likes from followers, that is set up to encourage workers towards

effective delivery of the service task, which may incentivize risky be-

haviours, and thereby increase the risk of accidents. These young

workers are evaluated before, under and after they have finished a task,

which means that there is a strong incentive structure putting pressure

on task completion, and thus provides some basic mechanisms that in-

crease the risk of the platform workers. These evaluation criteria are

often unclear and ambiguous, and thus also create uncertainty in terms

of when they have completed their work satisfactorily. The young

workers in the present study experience the least efforts gradient, e.g., in

terms of workloads related to speed, suboptimal equipment in cleaning

jobs, long working hours, bullying, affective work in relation to cus-

tomers, which leads to an increased risk in task completion. The same is

seen for food deliveries, where these basic incentive mechanisms lead to

particularly risky work situations, where young platform workers cycle

quickly through traffic without a helmet with one hand on the handle-

bars and their phone in the other (Nielsen

et al., 2020).

Thus, the young workers in the platform economy are pushed by the

very same gradients as proposed by

Rasmussen (1997).

When a strong

cost gradient and effort gradients drive the work activities,

Rasmussen

(1997)

suggests that the result very likely will be a systematic migration

toward the boundary of functionally unacceptable safety performance,

which might compromise safety and accidents or other health outcomes

may occur (Cook

Rasmussen, 2005; Rasmussen, 1997).

This means,

that strong gradients or forces are put on the young workers and thus

driving them to the boundaries of unsafe OHS performance (Fig.

4).

This

is particularly problematic in digital platform work, as the counter

gradient, in terms of an adequate OSH system and suitable safety pre-

vention measures, are very weak or entirely absent.

The present study shows, that it is up to the young digital platform

worker to make the trade-off between the effort gradient and the cost

gradient set up by the platform, and then the counter gradient, i.e., the

safety precautions, which the majority of the platforms in our study has

transferred to the young workers. The platforms have also transferred

the cost of prevention to the workers, i.e., safety equipment and in-

structions for doing work safely in relation to location-based platform

work, as also noted by Nilsen (2020). As Hans (doing cleaning work via

BEU, Alm.del - 2021-22 - Bilag 139: Orientering om ny NFA-artikel om unges arbejdsmiljø på digitale platforme, fra beskæftigelsesministeren

M.L. Nielsen et al.

Safety Science 149 (2022) 105674

the platform ‘Hilfr’) reported, suboptimal cleaning equipment and no

training or instruction in correct work postures contributed to him

having back problems in connection with the cleaning work. Should it

happen, that a platform worker is injured at work, then only some of

them are covered by an occupational injury insurance. Ditte, who takes

jobs as a dog walker, was sure she was covered by an occupational injury

insurance, but only the dog was covered by the insurance.

Similarly, workers on social media platforms have to deal with

health and safety themselves, such as in cases of long working hours,

neck pain, hate speech, or even online threats (Table

1).

Rasmussen’s control theoretic approach emphasizes that efforts for

improvement must then be directed toward the control of performance

in interaction with the boundary of unacceptable safety performance,

not on the control of errors (Cook

Rasmussen, 2005; Waterson et al.,

2017).

However, young workers on digital platforms have little or no

influence on the control of the performance design, which to a very high

degree is determined by the algorithmic management system. When the

digital platform owners take little or no responsibility for the in-

teractions between the performance system and the boundary of unac-

ceptable safety performance, this is left to the young workers to avoid or

compensate for the risks created by the work system.

5.3. Who takes care of the young workers’ OSH on digital platforms?

When young workers fall outside the traditional dyadic employer-

employee relationship, they are at the same time left with a very weak

or entirely absent OSH system (the counter gradient in Rasmussen’s

model), and they end up in a ‘protective gap’ in enforcement of rights

and OSH (Grimshaw

et al., 2016).

Furthermore, there is a lack of

coherence between decision making regarding the content and the

incentive structures underlying work performance and then the control

of risk and safety related to the work.

This advances an important question: ‘Who takes care of the young

workers’ OSH at digital platforms, thus the boundary of unacceptable

performance? Is it the client, the platform, or the young worker

themselves?

Neither national nor European legislation are clear about this pro-

tective gap. In connection with the legal definition of who is responsible

for the platform workers’ working environment, it is crucial to legally

define to what extent a given platform has access to

’dispose’

over the

work, and to what extent

’instructions

are given for the execution of the

work’ (Christensen,

2017p.21; Prassl

Risak, 2016).

Is a digital labour

platform to be legally defined as a neutral mediating marketplace based

on technology, or should a digital labour platform rather be defined as a

digitally based way of disposing of, instructing and controlling how and

when work is performed? This remains an unresolved legal issue in

Denmark and in the European Union. In Denmark most platform

workers are considered and treated as self-employed (Garben,

2017),

and work on this part of the Danish labour market is characterised by

being highly individualised and deregulated (Nielsen

et al., 2020).

Nevertheless, in spring 2021 the OSH of the workers and drivers related

to the online retail delivery service Nemlig.com attracted widespread

media attention in Denmark (Pr

ochold, 2021).

This happened parallel to

the announcement of the EU Commission’s ‘first-stage consultation of

European social partners on how to improve the working conditions for

people working through digital labour platforms’ (EU

Commission,

2021).

This has led the Danish government to announce, that more

political attention should be directed towards the ‘grey area’ of ‘platform

work’ and the working conditions of the platform workers, by proposing

’presumption

rule’, which places the responsibility on employers to

prove that their independent contractors are not

“employees”

(3F,

2021).

Regardless of such legislative discussions, we recommend that

strategies be developed to reach and protect workers on digital plat-

forms. This is important, since non-standard employment, temporary

work, and lack of introduction and training (aspects that all characterise

the platform economy) increase the risk of OSH issues (Garben,

2017).

is especially important to recognise how young platform workers can be

protected, since young workers as a group, are often more exposed to

injuries (Dyreborg

et al. 2019; European Agency for Safety and Health at

Work, 2007; Garben, 2017).

The platform economy’s expected growth

(Balaram

et al., 2017; Manyika et al., 2016)

underscores the importance

of communicating knowledge about the complex employment relations

associated with platform work - and its challenges - to governments,

social partners and industry, and to young workers engaged in atypical

employment via digital platforms. Our hope is that this paper can inform

policymakers and actors within the occupational health and safety field

about the protective gap and lack of enforcement of OSH in these types

of work environment, along with the strong incentive structures young

workers meet at the digital platforms.

6. Limitations

Even though, this study was based on a number of interviews, terms

and conditions of the digital platforms, and exchanges of data and re-

sults with stakeholders in the field, the sample of young workers is to a

high degree a convenience sampling, and will for this reason be biased.

While the study contributes with new knowledge on the organisation of

work and mechanisms at play that put young workers at risk at the

digital platforms, the study cannot tell the prevalence of OHS problems

among young people at digital platforms, and to what extend these risk

are higher than among young people at workplaces with traditional

employer-employee relationships. This is a challenge to collect un-

biased data on this group of workers. Future research should focus on

how to establish representative data on the OHS problems in this group

of young workers, in order to better target those young workers at

highest risk.

7. Conclusion

This study adds to the relatively sparse literature on OSH risks among

young workers on digital platforms, and the significance of changed

employment relations for the responsibilization of young workers for

their OSH. For digital labour platforms, the employment relations are

three-sided, and for social media platforms, the relations can even be

five-sided. This study demonstrated that young workers are not aware of

the formal rights and obligations that are relegated to them. The digital

labour platforms often operate in a way that makes the young workers

see them as their employers, as they control payment, instructions,

provide support functions, provide clothes and equipment with the

company logo on, and for some of them, their work is managed through

algorithmic management systems, wherefrom detailed instructions and

control emanate. This contributes to the impression among the young

workers that the digital platforms also take care of OSH and cover them

in case of an injury. Unfortunately, many of these young workers are

working without any kind of insurance that would cover them in the

event of an occupational injury. They are themselves responsible for the

OSH related to their work via the platforms.

The responsibilization of OSH is about the transfer of responsibility

for OSH from the digital platform or from the customers that receive the

services, onto the young workers, who are then called on to resolve the

OHS problems they encounter in their work. This is an insurmountable

task to put on these young workers, since they have limited control over

the planning and execution of tasks, and at the same time, they are under

a strong control and incentive structure. When young platform workers

operate in this grey zone, with complex relationships between workers,

BEU, Alm.del - 2021-22 - Bilag 139: Orientering om ny NFA-artikel om unges arbejdsmiljø på digitale platforme, fra beskæftigelsesministeren

M.L. Nielsen et al.

Safety Science 149 (2022) 105674

Christensen, S.W., 2017. Platformsøkonomien i et arbejdsretligt perspektiv [The Platform

Economy from an Employment Law Perspective]. Københavns Universitet.

Specialeafhandling RETTID 12 (2017).

Cook, R.

J. Rasmussen (2005). Going solid: a model of system dynamics and

consequences for patient safety, BMJ Quality

Safety, 14 (2): 130-134. Cook, R.,

Coyle, D., 2017. Precarious and Productive Work in the Digital Economy. National

Institute Economic Review 240 (1), R5–R14.

https://doi.org/10.1177/

002795011724000110.

Danish Working Environment Authority (2015). At-intern instruks IN-1-2, 6 marts/2012,

senest redigeret: 1. januar 2015. [Internal instructions IN-1-2, 6 March / 2012, Last

edited: 1 January 2015]. Copenhagen. Arbejdstilsynet.

Dyreborg, J., Sundstrup, E., Jørgensen, A., Kristiansen, J., Hougaard, K. S.,

Johansen,

N. F. (2019). Unges arbejdsmiljø og helbred [Rapport til Arbejdsmiljørådet]. Det

Nationale Forskningscenter for Arbejdsmiljø (NFA).

Dogley. (2018, May 22). Vilkår for anvendelse af Dogley [Terms for using Dogley].

https://dogley.com/da/betingelser-og-vilkaar [Retrieved on 13.01.2020].

EU Commission (2021). Questions and Answers: First stage social partners consultation

on improving the working conditions in platform work. (24.02.21) https://ec.

europa.eu/commission/presscorner/detail/en/qanda_21_656 [Retrieved on

09.05.21].

European Agency for Safety and Health at Work. (2007). Young workers—Facts and

figures Exposure to risks and health effects. OSHA Europe.

Garben, S. (2017). Protecting Workers in the Online Platform Economy: An overview of

regulatory and policy developments in the EU. [European Risk Observatory.

Discussion paper]. European Agency for Safety and Health at Work. https://doi.org/

10.2802/918187.

Gandini, A., 2019. Labour process theory and the gig economy. Hum. Relat. 72 (6),

1039–1056.

https://doi.org/10.1177/0018726718790002.

Gregg, M., 2011. Part-time precarity. Work’s intimacy. Polity Press, Cambridge, UK.

Grimshaw, D., Johnson, M., Rubery, J., Keizer, A., 2016. Reducing Precarious Work -

Protective gaps and the role of social dialogue in Europe. European Work and

Employment Research Centre, University of Manchester, UK.

Hale, A.R., 1984. Is safety training worthwhile? Journal of Occupational Accidents 6 (1-

3), 17–33.

Hilfr. (n.d.). Hilfr. Your new platform for cleaning services! Hilfr.dk. https://hilfr.dk

[Retrieved on 13.012020].

Huws, U., 2015. Online labour exchanges, or ‘crowdsourcing’: Implications for

occupational safety and health: Review article on the future of work. EU-OSHA.

Ilsøe, A., 2017. The digitalisation of service work

–

social partner responses in Denmark,

Sweden and Germany. Transfer: European Review of Labour and Research 23 (3),

333–348.

https://doi.org/10.1177/1024258917702274.

Ilsøe, A., Madsen, L.W., 2017. Digitalisering af arbejdsmarkedet—Danskernes erfaring

med digital automatisering og digitale platforme [Digitalisation of the labour market

- Danes’ experiences with digital automation and digital platforms]. FAOS

Forskningsnotat 157, 59.

Ilsøe, A., Larsen, T., 2021. Platformsøkonomiens udvikling i Danmark- stabilitet og

forskelle. [The development of the platform economy in Denmark- stability and

differences]. Tema nr. Digitalisering- i verdens mest digitaliserede samfund.

Samfundsøkonomi. (1), 45–53.

https://doi.org/10.7146/samfundsokonomen.

v2021i1.125544.

Jesnes, K.,

Øistad,

B. S., Alsos, K.,

Nesheim, T. (2016). Aktører og arbeid i

delingsøkonomien [Actors and work in the sharing economy]. Delrapport, FAFO-

notat 2016:23. FAFO.

Johnson, M.R., Woodcock, J., 2019. ‘It’s like the gold rush’: The lives and careers of

professional video game streamers on twitch.tv. Information, Communication

Society 22 (3), 336–351.

Kalleberg, A.L., 2009. Precarious work, insecure workers: Employment relations in

transition. American Sociological Review 74 (1), 1–22.

Kjell

n, U., 2000. Prevention of Accidents through Experience Feedback. Taylor

Francis, London.

Kristiansen, A., Andersen, S.K., 2017. Digitale platforme og arbejdsmarkedet- Hvad er

digitale platforme, og hvilke spørgsmål rejser de for arbejdsmarkedet? [Digital

platforms and the labormarket - What is digital platforms and what quistions do they

rais according to the labor market?] Forskningsnotat 151. Københavns Universitet,

FAOS.

Laursen, C.S, Nielsen, M. L.,

Dyreborg, J. (2021). Young Workers on Digital Labor

Platforms: Uncovering the Double Autonomy Paradox. Nordic Journal of Working

Life Studies, 11(4). https://doi.org/10.18291/njwls.127867.

Le Coze, J.C., 2015. Reflecting on Jens Rasmussen’s legacy. A strong program for a hard

problem. Safety Science 71, 123–141.

Lee, M. K., Kusbit, D., Metsky, E.

Dabbish, L. (2015). Working with Machines: The

Impact of Algorithmic and Data-Driven Management on Human Workers. In:

Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing

Systems - CHI

’15,

Seoul, Republic of Korea, 2015, 1603–1612. ACM Press. DOI:

10.1145/2702123.2702548.

MacDonald, R., Giazitzoglu, A., 2019. Youth, enterprise and precarity: Or, what is, and

what is wrong with, the ‘gig economy’? Journal of Sociology 55 (4), 724–740.

https://doi.org/10.1177/1440783319837604.

Manyika, J., Lund, S., Bughin, J., Robinson, K., Mischke, J.,

Mahajan, D. (2016).

Independent work: Choice, necessity, and the gig economy. McKinsey Global

Institute. https://www.mckinsey.com/featured-insights/employment-and-growth/

independent-work-choice-necessity-and-the-gig-economy [Retrieved on 13.012020].

Moore, P. V. (2018). The Threat of Physical and Psychosocial Violence and Harassment

in Digitalized Work. International Labour Office, Geneva, 2018. https://www.ilo.

org/wcmsp5/groups/public/—ed_dialogue/—actrav/documents/publication/

wcms_617062.pdf [Retrieved on 13.08.2020].

employers and customers as well as blurred responsibilities for the

working environment, it poses a risk to the health and safety of these

young workers.

There are, therefore, reasons to question the transfer of re-

sponsibilities for OSH from the digital platforms to the young workers,

and in particular, there is a need for making explicit the responsibility

for OSH in these jobs. Here is clearly a protective gap that needs to be

closed.

Credit authorship contribution statement

Mette Lykke Nielsen:

Conceptualization, Methodology, Writing -

Original draft preparation, Funding acquisition.

Cæcilie S. Laursen:

Conceptualization, Data curation, Formal analysis, Review

Editing.

Johnny Dyreborg:

Conceptualization, Methodology, Theoretical

approach, Writing - Review

Editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial

interests or personal relationships that could have appeared to influence

the work reported in this paper.

Acknowledgements

This research received grants from the Danish Working Environment

Research Fund under Grant 46-2017-09 and The Nordic Council of

Ministers under Grant 16-01158 We also thank researcher Kari Anne

Holte, NORCE, at Stavanger University, Norway and associate professor,

Louise Yung Nielsen, Roskilde University, Denmark, for providing

interview material for this study.

References

Hilfr (2018). Hilfr overenskomst 2018 [Hilfr’s collective agreement 2018]. F [United

Federation of Workers in Denmark]. Privat Service, Hotel og Restauration. https://

www.f.dk/fagforening/fag/rengoeringsassistent-(privatansat)/overenskomsten-hilfr

[retrieved on 1.12.21].

3F (2021). Regeringen går til kamp mod falske selvstændige | Fagbladet F (The

government goes to war against false self-employed | United Federation of Workers

in Denmark) [retrieved on 29.09.2021].

Abidin, C., 2019. Victim, Rival, Bully: Influencers’ Narrative Cultures Around

Cyberbullying. In: Vandebosch, H., Green, L. (Eds.), Narratives in Research and

Interventions on Cyberbullying among Young People. Springer International

Publishing, Cham, pp. 199–212.

https://doi.org/10.1007/978-3-030-04960-7_13.

Abidin, C., Hansen, K.S., Hogsnes, M., Newlands, G., Nielsen, M.L., Yung Nielsen, L.,

Sihvonen, T., 2020. A Review of Formal and Informal Regulations in the Nordic

Influencer Industry. Nordic Journal of Media Studies 2 (1), 71–83.

Antonucci, L., Hamilton, M.,

Roberts, S. (Eds.). (2014). Young People and Social Policy

in Europe. Palgrave Macmillan UK. https://doi.org/10.1057/9781137370525.

Arnholtz, J., Meardi, G., Oldervoll, J., 2018. Collective wage bargaining under strain in

northern European construction. Resisting institutional drift? 24 (4), 341–356.

Balaram, B., Warden, J.,

Wallace-Stephens, F. (2017). Good Gigs: A fairer future for the

UK’s gig economy. RSA (Royal Society for the encouragement of Arts, Manufactures

and Commerce). https://nls.ldls.org.uk/welcome.html?ark:/81055/vdc_

100060614186.0x000001.

Braun, V., Clarke, V., 2006. Using thematic analysis in psychology. Qualitative Research

in Psychology. 3 (2), 77–101.

https://doi.org/10.1191/1478088706qp063oa.

Bryman, A., 2016. Documents as sources of data. In: Social research methods, Fifth

Edition. Oxford University Press, pp. 545–568.

Casas-Cort

s, M., 2014. A Genealogy of Precarity: A Toolbox for Rearticulating

Fragmented Social Realities in and out of the Workplace. Rethinking Marxism 26 (2),

206–226.

https://doi.org/10.1080/08935696.2014.888849.

Chabber (n.d.-a). Basic information about salary at Chabber. http://faq.chabber.com/

en/articles/2544961-basic-information-about-salary-at-chabber [Retrieved on

07.01.2020].

Chabber (n.d.-b). Hvad hvis en vagt var kortere eller længere end oprindeligt angivet?

[What if a shift was shorter or longer than originally stated?]. http://faq.chabber.

com/en/articles/2169942-hvad-hvis-en-vagt-var-kortere-eller-laengere-end-

oprindeligt-angivet [Retrieved on 13.01. 2020].

Chabber. (2019). Terms and Conditions. https://www.chabber.com/

TermsAndConditions.pdf [Retrieved on 13.01. 2020].

Chen, J.Y., 2018. Thrown under the bus and outrunning it! The logic of Didi and taxi

drivers’ labour and activism in the on-demand economy. New Media

Society 20

(8), 2691–2711.

https://doi.org/10.1177/1461444817729149.

BEU, Alm.del - 2021-22 - Bilag 139: Orientering om ny NFA-artikel om unges arbejdsmiljø på digitale platforme, fra beskæftigelsesministeren

M.L. Nielsen et al.

Mohlmann, M.,

Zalmanson, L. (2017). Hands on the wheel: Navigating algorithmic

management and Uber drivers’ autonomy. The International Conference on

Information Systems (ICIS 2017), 18.

Narvaiza, Z. (2011). Opinion of the European Economic and Social Committee on New

trends in self-employed work: The specific case of economically dependent self-

employed work (own-initiative opinion). Official Journal of the European Union, C

18(08), 44–52.

Nielsen, M. L., Nielsen, L. Y., Holte, K. A., Andersson, Å., Gudmundsson, G., Heijstra, T.

M.,

Dyreborg, J. (2019). New forms of work among young people. Nordic Council

of Ministers. https://doi.org/10.6027/Nord2019-025.

Nielsen, M.L., Laursen, C., 2020. Unge på digitale arbejdsplatforme: Prekært arbejde eller

unge entreprenører? [Young workers on digital work platforms: Precarious work or

young entrepreneurs?]. Nordisk Tidsskrift for Ungdomsforskning. 2.

https://www.

idunn.no/ntu/2020/02/unge_paa_digitale_arbejdsplatforme_prekaert_arbejde_elle

r_un.

Nielsen, M. L., Laursen, C., Dyreborg, J., Nielsen, L.Y. (2020). Risiko og arbejdsmiljø

blandt unge på digitale arbejdsmarkeder- et kollaborativt udviklingsprojekt. [Risk

and working environment among young people in digital labor markets - a

collaborative development project]. Afslutningsrapport til

Arbejdsmiljøforskningsfonden. Det Nationale Forskningscenter for Arbejdsmiljø

(NFA) og CeFU. https://amff.dk/media/17274/2021_radar_amff_final.pdf

[Retrieved on 09.05.21].

Nielsen, M.L., Dyreborg, J., Lipscomb, H.J., 2018. Precarious work among young Danish

employees a permanent or transitory condition? Journal of Youth Studies 22 2019

(1), 7–28.

https://doi.org/10.1080/13676261.2018.1469739.

Nielsen, M.L., Dyreborg, J., Kines, P., Nielsen, K., Rasmussen, K., 2013. Exploring and

expanding the category of young adult workers. Situating young workers’ risk

management in the retail industry. Nordic Journal of Working Life Studies 3 (3),

219–243.

Nielsen, M.L., Gorlich, A., Grytnes, R., Dyreborg, J., 2017. Without a Safety Net:

Precarization Among Young Danish Employees. Nordic Journal of Working Life

Studies 7 (3).

https://doi.org/10.18291/njwls.v7i3.97094.

Nilsen, M., Kongsvik, T., Antonsen, S., 2020. Working Conditions and Safety in the Gig

Economy - A Media Coverage Analysis. e-proceedings of the 30th European Safety

and Reliability Conference and 15th Probabilistic Safety Assessment and

Management Conference (ESREL2020 PSAM15).

Popescu, G.H., Petrescu, I.E., Sabie, O.M., 2018. Algorithmic Labor in the Platform

Economy: Digital Infrastructures, Job Quality, and Workplace Surveillance.

Economics, Management, and Financial Markets 13 (3), 74.

https://doi.org/10

.22381/EMFM13320184.

Prassl, J., 2018. Humans as a Service: The Promise and Perils of Work in the Gig

Economy. Oxford University Press.

https://doi.org/10.1093/oso/

9780198797012.001.0001.

Prassl, J., Risak, M., 2016. Uber, Taskrabbit, and Co.: Platforms as Employers?

Rethinking the Legal Analysis of Crowdwork. Comparative Labor Law

Policy

Journal 37 (2016).

Prochold, J. (2021). Lagermedarbejdere hos Nemlig fortæller om deres hverdag: Jeg kan

ikke engang drikke vand, for så skal jeg på toilettet, og det

ødelægger

mine

effektivitetstal. [Warehouse employees at Namely tell about their everyday life: I can

not even drink water, because then I have to go to the toilet, and it destroys my

efficiency figures]. Politiken 10.04.21 https://politiken.dk/forbrugogliv/forbrug/

art8157981/%C2%BBJeg-kan-ikke-engang-drikke-vand-for-s%C3%A5-skal-jeg-p%

C3%A5-toilettet-og-det-%C3%B8del%C3%A6gger-mine-effektivitetstal%C2%AB

[Retrieved on 09.05.2021].

Safety Science 149 (2022) 105674

Rasmussen, J., 1997. Risk management in a dynamic society: a modelling problem.

Safety Science 27 (2–3), 183–213.

Raval, N.

Dourish, P. (2016). Standing Out from the Crowd: Emotional Labor, Body

Labor, and Temporal Labor in Ridesharing. In: Proceedings of the 19th ACM

Conference on Computer-Supported Cooperative Work

Social Computing - CSCW

’16,

San Francisco, California, USA, 2016, 97–107. ACM Press. DOI: 10.1145/

2818048.2820026.

Rosenblat, A., Stark, L., 2016. Algorithmic Labor and Information Asymmetries: A Case

Study of Uber’s Drivers. International Journal of Communication 10, 27.

Rubery, J., Grimshaw, D., Keizer, A., Johnson, M., 2018. Challenges and Contradictions

in the ‘Normalising’ of Precarious Work. Work, Employment and Society. 32 (3),

509–527.

https://doi.org/10.1177/0950017017751790.

Samant, Y. (2019). The Promises and Perils of the Platform Economy: Occupational

Health and Safety Challenges, and the Opportunities for Labour Inspection

[Statement]. https://www.ilo.org/global/about-the-ilo/how-the-ilo-works/

departments-and-offices/governance/labadmin-osh/events/world-day-for-safety/

33thinkpieces/WCMS_681619/lang–en/index.htm [Retrieved on 19.06. 2019].

Scheuer, S., 2017. Atypisk beskæftigelse i Danmark: Om deltidsansattes, midlertidigt

ansattes og soloselvstændiges vilkår [Atypical employment in Denmark: About part-

time employed, temporary employed, and solo self-employed’s conditions]. (LO-

dokumentation Nr. 1, LO.

Standing, G., 2016. Labour brokers: The precariat bears the strain. The corruption of

capitalism: Why rentiers thrive and work does not pay. Biteback Publishing.

Todolí-Signes, A., 2017. The ‘gig economy’: Employee, self-employed or the need for a

special employment regulation? Transfer: European Review of Labour and Research

23 (2), 193–205.

https://doi.org/10.1177/1024258917701381.

van Dijck, J., 2013. YouTube: The Intimate Connection between Television and Video

Sharing. In: van Dijck, J. (Ed.), The Culture of Connectivity: A Critical History of

Social Media. Oxford University Press, pp. 110–131.

https://doi.org/10.1093/

acprof:oso/9780199970773.003.0006.

Van Eerd, D., Saunders, R., 2017. Integrated knowledge transfer and exchange: An

organizational approach for stakeholder engagement and communications. Scholarly

and Research Communication 8 (1).

Vershinina, N.A., Rodgers, P., Ram, M., Theodorakopoulos, N., Rodionova, Y., 2018.

False self-employment: the case of Ukrainian migrants in London’s construction

sector: False self-employment of Ukrainian migrants in London’s construction sector.

False self-employment: the case of Ukrainian migrants in London’s construction

sector 49 (1), 2–18.

Wagner, I., Refslund, B., 2016. Understanding the diverging trajectories of

slaughterhouse work in Denmark and Germany: A power resource approach.

European Journal of Industrial Relations 22 (4), 335–351.

Waterson, P., Le Coze, J.-C., Andersen, H.B., 2017. Recurring themes in the legacy of

Jens Rasmussen. Applied Ergonomics 59, 471–482.

Weber, S.S., 2018. Working life on Nordic labour platforms. In: Hvid, H., Falkum, E.

(Eds.), Work and Wellbeing in the Nordic Countries (1st Edition]. Routledge,

pp. 339–358.

Weil, D., 2019. Understanding the present and future of work in the fissured workplace

context. RSF: The Russell Sage Foundation. Journal of the Social Sciences 5 (5),

147–165.

Wood, A.J., Graham, M., Lehdonvirta, V., Hjorth, I., 2019. Good Gig, Bad Gig: Autonomy

and Algorithmic Control in the Global Gig Economy. Work, Employment and Society

33 (1), 56–75.

https://doi.org/10.1177/0950017018785616.

Zwick, A., 2018. Welcome to the Gig Economy: Neoliberal industrial relations and the

case of Uber. GeoJournal 83 (4), 679–691.

https://doi.org/10.1007/s10708-017-

9793-8.