BookPDF Available

Abstract and Figures

This working paper investigates a rare case of platform work performed in the realm of regulated employment and explores the attitudes of platform workers towards collective representation. The argument is developed that workers’ attitudes and their propensity to engage in collective action can be related to their demographic and socio-economic characteristics and to their experiences of job quality. The empirical analysis is based on a case study of Deliveroo, a place-based food delivery platform. In Belgium, Deliveroo used another labour market intermediary, SMart, to organise, manage and legalise workers’ employment status. Thus, effectively, Deliveroo riders worked as dependent employees but were hired by SMart, not Deliveroo. Moreover, this arrangement was unilaterally terminated by Deliveroo during the period under analysis, sparking discontent and active protests among its riders. The results show that platform workers, in our case predominately young students, are not essentially different to their peers in their views on trade unions and their inclination to unionise. They do not generally hold negative views towards unions and do not consider unions incompatible with platform work. Instead, the results point to a lack of union activity in reaching out to riders as a reason for their non-membership. Engaging with them may offer trade unions a window of opportunity to win trust and demonstrate the added value of union membership in their school-to-work transitions.
Content may be subject to copyright.
‘Algorithm breakers’ are not
a different ‘species’: attitudes
towards trade unions of
Deliveroo riders in Belgium
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
Working Paper 2019.06
‘Algorithm breakers’ are not
a different ‘species’: attitudes
towards trade unions of
Deliveroo riders in Belgium
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
Working Paper 2019.06
european trade union institute
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil are senior researchers at the
European Trade Union Institute. Contact:, and
We are thankful to SMartBe, and Sarah de Heusch in particular, for helping to organise
the questionnaire survey and for sharing administrative data on Deliveroo workers. The
authors are also very grateful for the documentation kindly provided by Aline Hoffmann,
and for the remarks and suggestions of Callum Cant, Kristin Jesnes, Vincenzo
Maccarrone, Timothy March, Kyle Michiels, Philippe Pochet, Arianna Tassinari and Martin
Willems on earlier versions of this working paper.
ETUI publications are published to elicit comment and to encourage debate. The views
expressed are those of the author(s) alone and do not necessarily represent the views of
the ETUI nor those of the members of its general assembly.
Brussels, 2019
©Publisher: ETUI aisbl, Brussels
All rights reserved
Print: ETUI Printshop, Brussels
ISSN: 1994-4446 (print version)
ISSN: 1994-4454 (electronic version)
The ETUI is financially supported by the European Union. The European Union is not responsible for
any use made of the information contained in this publication.
3WP 2019.06
Abstract ........................................................................................................................................................ 4
Introduction ................................................................................................................................................ 5
1. The end of the SMart arrangement ...................................................................................... 8
2. Rider mobilisation and the long-standing unions ....................................................... 10
3. Riders’ profile: male, young and student; but also migrant .................................... 14
4. A very low unionisation rate ................................................................................................. 17
5. Prevailing lack of trade union exposure .......................................................................... 19
6. Platform-based food delivery is complementary to unionisation ......................... 22
7. Individual ‘fit’ and intention to unionise ........................................................................ 24
8. Labour market ‘fit’ and intention to unionise ............................................................... 27
Conclusion ................................................................................................................................................. 30
References ................................................................................................................................................ 33
Appendix ................................................................................................................................................... 37
4WP 2019.06
This working paper investigates a rare case of platform work performed in the
realm of regulated employment and explores the attitudes of platform workers
towards collective representation. The argument is developed that workers’
attitudes and their propensity to engage in collective action can be related to
their demographic and socio-economic characteristics, and to their experiences
of job quality. The empirical analysis is based on a case study of Deliveroo, a
place-based food delivery platform. In Belgium, Deliveroo used another labour
market intermediary, SMart, to organise, manage and legalise workers’
employment status. Thus, effectively, Deliveroo riders worked as dependent
employees but were hired by SMart, not Deliveroo. This arrangement was
unilaterally terminated by Deliveroo during the period under analysis, sparking
discontent and active protests among its riders. The results show that platform
workers, in our case predominately young students, are not essentially different
to their peers in their views on trade unions and their inclination to unionise.
They do not generally hold negative views towards unions, and do not consider
unions incompatible with platform work. Instead, the results point to a lack of
union activity in reaching out to riders as a reason for their non-membership.
Engaging with them may offer trade unions a window of opportunity to win
trust and demonstrate the added value of union membership in their school-
to-work transitions.
Digital labour platforms are defined by market-making through the matching
of clients with service providers (Drahokoupil and Piasna 2017). However, their
role in the labour market goes well beyond the provision of automated task
allocation. It often also includes performance management and control over
remuneration. Platforms thus provide different forms of algorithmic
management’ (Aloisi 2016), shifting a range of managerial responsibilities from
humans to machines.1To gain an understanding of the opportunities and
incentives for work organisation, it is essential to appreciate that platforms are
diverse (Drahokoupil and Fabo 2016). A major distinction is typically made
between geographically-dispersed platforms that organise the provision of
digitally-delivered services and place-based platforms that facilitate physically-
delivered services. In contrast to cloud-based work on dispersed platforms,
work through place-based platforms requires local human input and
interaction (Finkin 2016). They often also deploy managerial staff locally to
manage relations with regulators, clients and workers who are all concentrated
locally. Place-based platforms are directly affected by local regulations and
institutions, with the latter also influencing the working conditions and pay of
workers contracted by such platforms.
Place-based platforms, with food delivery as a prominent example, thus
provide opportunities and incentives for workers to organise collectively. As
their business model is based on building consumer base and loyalty around
their brands, food delivery platforms need to exert a high degree of control
over the work process to ensure consistent quality of service associated with
the brand (Kalleberg and Dunn 2016). Their business model also requires
control over the fees charged to consumers and restaurants, as well as over the
remuneration of delivery riders, or couriers. The high degree of control over
working conditions and pay by the platforms creates incentives for riders to
target them with collective action in the pursuit of their interests and needs.
Mobilising and organising strategies by trade unions, based on combining
offline one-on-one recruitment with digital community-building, in the process
fostering group identification and gaining network effects, can be potentially
effective in such a context.
Indeed, examples of grassroots actions by place-based platform workers are
not lacking (Vandaele 2018a). This is particularly illustrated by Deliveroo, the
1. In this process, aspects of human resource management, like the evaluation of platform
workers, are also outsourced to clients via rating systems.
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06 5
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
app-based, food delivery service. Founded in 2013, Deliveroo is probably one
of the most visible and well-known place-based platforms in major cities across
western Europe. The first-ever strike over pay was initiated by its riders in
London in the summer of 2016. Later on, direct action by the riders also
targeted other food delivery platforms in the United Kingdom (UK), and spread
across borders to several other European cities (Animento et al. 2017; Cant
2017, 2018a; Degner and Kocher 2018; Tassinari and Maccarrone 2017, 2018;
Vandaele 2017; Zamponi 2018). Akin to Hobsbawm’s analysis of workers’
machine-wrecking and rioting in the nineteenth century (Hobsbawm 1952;
Wood 2015), protesting riders could be labelled ‘algorithm breakers’. It could
be claimed that their grassroots actions against the algorithmic management
of work organisation are neither desperately futile nor retrograde. Like the
Luddites, actions are eventually aimed at the state, especially at the local level,
and are focused on pressing for the better regulation of employment terms and
conditions. It is a struggle that continues today, not without success (see Table
6 in the Appendix), and, as demonstrated by a nationwide strike in October
2018, also includes an industrial alliance between platform-based food delivery
workers and fast food workers in the case of the UK.
This working paper builds upon the work of Drahokoupil and Piasna (2019).
It provides an explorative case study of Deliveroo riders in Belgium and is
based on a survey, conducted in co-operation with the labour market
intermediary Société Mutuelle pour artistes (SMart), between December 2017
and January 2018. A complete database of the email addresses of all riders
registered through SMart in the period from September 2016 to August 2017
(N=3,279, of which about 1,000 were active in that period) was used to
distribute a link to a self-completed online survey. Respondents could choose
between three language versions: Dutch; English; and French. After sending
out an invitation email and two follow-up reminders, a total of 544 responses
to the survey was obtained. For the purposes of the analysis, only those
respondents who answered questions about trade unions were selected,
resulting in a final sub-sample of 289 riders, yielding a response rate of 8.8
per cent. This is representative of the entire population of Deliveroo riders
active through SMart in the period in question in terms of gender (p=0.37,
FET) and type of employment (p=0.81, FET).
Focusing on a worker perspective, we thus introduce new empirical evidence
in examining the trade union attitudes of Deliveroo riders in Belgium, and their
propensity to unionise. We argue that platform workers, at least in this type
of place-based platform work in Belgium, are not essentially different to their
peers in their attitudes towards unions and in their likelihood to join a union.
Predominantly young students, Deliveroo riders in Belgium do not greatly
differ from what is known about the attitudes towards unions of young people
in general. This does not imply, however, that riders should be considered and
organised as an entirely homogenous group defined by age; we identify that
there are other important differences within this workforce. Moreover, their
subjective understandings of their job quality shape their attitudes towards
collective organisation and trade unions (Goods et al. 2019). Consequently,
such diversity and subjectivity entail different opportunity costs for unions and
6WP 2019.06
prompt the advancement of diverse, tailor-made union organising strategies
either embedded in young people’s school-to-work transitions (Vandaele
2018b) or beyond this life phase.
This working paper is organised as follows. For a better understanding of the
survey results, sections 1 and 2 provide contextual information about Deliveroo
riders in Belgium, i.e. the SMart arrangement for employing riders, and their
mobilisation against the termination of this arrangement, undertaken within
the Riders Collective and with the support of the long-standing unions. Section
3 provides information about the demographic and other characteristics of
Deliveroo riders. This is followed by data on the trade union density of the
riders in section 4. Their attitudes towards unions are analysed in section 5.
Section 6 focuses on the propensity of the riders to unionise. This propensity
is further analysed by introducing the subjective understandings of the riders
of their job quality in section 7. Section 8 puts their job quality into the broader
perspective, i.e. the context of platform-based food delivery, and how this
labour market context influences intentions to unionise.
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
1. The end of the SMart arrangement
Deliveroo entered the Belgian labour market in September 2015. It had a
strong incentive to provide work to students. According to local regulations,
employed students in formal education, irrespective of their nationality, could
work for up to 475 hours per year while paying only 2.71 per cent social
insurance contribution and no tax. Employers’ social insurance costs were
also substantially lower for students: thus, the overhead charge for a student
was only nine per cent of gross income in contrast to 55 per cent for other
workers. However, and despite being in full control of hiring and shift
allocation, Deliveroo did not opt to employ riders directly, following its
general policy of avoiding being classified and perceived of as an employer.
Instead, as of May 2016, Deliveroo riders could either bill their services
through SMart, a labour market intermediary providing support to artists and
other project-based workers seeking to organise discontinuous careers
(Xhauflair et al. 2018), or work on a self-employed basis and invoice the
platform directly. The riders, the majority of them being students, opted to
work through SMart and thus benefit from employment status and the tax
advantages for students (Drahokoupil and Piasna 2019). This put SMart into
a position to negotiate, with Deliveroo and another food delivery platform,
Take Eat Easy, a joint protocol that standardised pay structures and
introduced some worker protection. Thus, crossing the boundaries between
traditional labour market actors (Xhauflair et al. 2018), SMart took
responsibilities which were partly those of an employer and partly those of
trade unions by establishing an employment relationship with riders but also
voicing their concerns vis-à-vis the platform – in the latter case, it acted as a
quasi-union (Vandaele 2018a).
According to the agreement between SMart and Deliveroo, SMart provided
riders with employment status and, therefore, as a formal employer, had to
comply with the legal minimum standards required in Belgium. Thus, riders
employed through SMart had access to social security, were guaranteed a
minimum wage and received partial reimbursement for the use of their mobile
phones. They were also guaranteed minimum three-hour shifts, which were
paid in full even if a technical problem or accident prevented a rider from
finishing a shift. SMart employees also received safety training and were
covered by work-related accident insurance and third-party liability insurance
(financed from the 6.5 per cent fee invoiced to the platforms). As a legal
employer, SMart also administered a salary fund that provided insurance
against the bankruptcy of the platform or late payments. The fund was soon
tested as Take Eat Easy could not compete with Deliveroo and went bankrupt
in July 2016. SMart disbursed €400,000 from its salary fund to pay the
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
8WP 2019.06
affected Take Eat Easy riders, gaining considerable legitimacy from riders for
its model (Dufresne et al. 2018).
Claiming that riders would benefit from greater flexibility, Deliveroo
announced, in October 2017, its intention to change its work allocation
algorithms, alter its remuneration approach to a per-delivery pay system and
terminate its partnership with SMart. The transition towards a self-
employment model lasted until January 2018 and it coincided, remarkably,
with the expansion of policies promoting platform work in Belgium (the so-
called De Croo Law) (see Lenaerts et al. 2017; Vandaele 2017). This new
framework for platform work offered tax relief to self-employed platform
workers and, from Deliveroo’s perspective, was as financially attractive as
relying on workers with employed student status. Using self-employed labour
gave Deliveroo the flexibility to change its pay system and working conditions
without having to negotiate conditions with SMart, and without dealing with
the other ‘constraints’ of an employment relationship. Deliveroo thus avoided
the prospect of being covered by the collective agreement that was being
negotiated at the time.
The Deliveroo case sparked a political debate about the employment
categorisation of app-based platform workers in Belgium, but this did not stop
Deliveroo from terminating its co-operation with SMart. More importantly,
however, this unilateral move by Deliveroo caused resentment among riders,
giving new impetus to their self-organisation and creating opportunities for
the long-standing trade unions to support their protests.
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
2. Rider mobilisation and the long-
standing unions
Belgium has been no exception when it comes to protests by food delivery
riders in western Europe against bogus self-employment, precarious
employment and payment models based on piece rates. The long-standing
Belgian trade unions have, since 2017, been engaging with them via support
for the Riders Collective (Koerierscollectief/Collectif des coursier.e.s). This
Collective was initially a self-organised network of food delivery riders with no
particular focus on socio-economic grievances, being informally set up in 2015
(Dufresne et al. 2018). Protests against the employment practices deployed by
Take Eat Easy took place in 2016, but the Riders Collective rose to prominence
in particular for its defence of riders’ interests in the settlement of Take Eat
Easy’s bankruptcy. It is clear that the bottom-up morphology of the Collective,
shaped by and embedded in the broader framework of the platform economy,
is quite different from that of the long-standing unions. Differences between
the Riders Collective and the unions comprise, among others, organisational
form and ideological identity, the membership domain, and both its conception
of membership and its relationship with its members – see Table 1.
Regarding organisational form, the Riders Collective can be conceived of as a
self-organised, network-based or decentralised, online and offline occupational
community of ‘rider-activists’ facilitating the mobilisation and organisation of
other riders. The Collective could be perceived as being in its ‘organic phase’
given its recent establishment (Boxall 2008). Such an evolutionary perspective,
based on organisational ecology, would imply that, if its leadership and
organisation were successful, the Riders Collective would ‘mature’ into a union,
fairly similar to conventional organisational forms of workplace-based
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
10 WP 2019.06
Source: authors’ own typology.
Table 1 Comparing the long-standing unions and the Riders Collective in Belgium
Riders Collective
Logic of membership
Long-standing unions
Monthly membership fees
Reduced fees, or free membership for certain member
Logic of influence
‘Pillarisation’: catholic, liberal and socialist identities
Main features
Union organisation
Ideological identity
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
unionism. Alternatively, the Collective could be categorised as an ‘alt-union’
in the platform economy, i.e. its organisational form resembles to a great extent
that which is considered as alternative labour (‘alt-labour’) in the US context:
worker advocacy groups filling a void in industries where the traditional labour
movement is not present and which are ‘typically modest in size, with few staff
and limited financial resources’ (Milkman 2013: 656; Oswalt 2016). If the
relationship between the Riders Collective and the unions (further) strengthens
in the future, then the Collective could be seen as an ‘orchestrating’ shell of the
more bureaucratic long-standing unions in mobilising and organising this
social media-based network of riders (Heckscher and McCarthy 2014; Pasquier
and Wood 2018; Wood 2015). Another option would be that the Riders
Collective, in collaboration with unions or otherwise, sets up a rider- or union-
owned cooperative.
Whether the Collective will be a long-lasting form of unionism remains to be
seen, but it is certainly not completely idiosyncratic in its form: union
formation in platform-based food delivery in several other, although not all,
west European countries looks quite similar in terms of representation
structures – see Table 5 in the Appendix.
Membership of the Riders Collective is free, irrespective of riders’ employment
status, and it is sufficient to ‘like’ its Facebook page to be considered a
member.2In practice, its membership is occupational-based in contrast to the
industry-based membership of the long-standing trade unions in Belgium.
Although membership in the unions is also free for students, the concept of
formal union membership, involving member registration and administration,
is thus not applicable to the Collective. Its membership is self-evidently
complementary to membership of the longstanding unions such that
overlapping membership cannot be excluded. This could especially be expected
of those riders who have spells of unemployment, which provides incentives
to unionise since unions are involved in the provision of unemployment
benefits (Van Rie et al. 2011). Simultaneously, ‘liking’ the Facebook page of the
Riders Collective offers an opportunity to unions to engage with riders online.
The Collective is ideologically neutral whereas the unions are, in contrast,
historically rooted in the traditional ideological pillars’ of Belgian society,
although this rivalry has blurred to a more pragmatic stance today (Faniel
2010). This reminds of the early days of unionism in Belgium, when small craft
unions were equally free from any political ideological demarcation
(Strikwerda 1997). The ideological neutrality of the Riders Collective implies
that it is, on paper, open to cooperation and alliance-building with all unions
whether catholic, socialist or liberal. Indeed, while being an ‘alt-union’ in the
platform economy, this labelling does not imply that there is a default position
of resentment between the Riders Collective and the unions.
2. 1,771 people have ‘liked’ the page at the time of writing (25 March 2019).
Thus, the Collective could reckon upon the experience of the long-standing
trade unions in negotiating a collective agreement for improving the terms and
conditions of the riders employed by SMart. While concluding a company
agreement would have been a kind of derivation of unions’ traditional
bargaining focus at industry level, with the company level as additional, it
would still be embedded in their dominant logic of the pursuit of influence
(Vandaele 2018a). In anticipation of the regulatory employment classification
of riders, the unions considered SMart as a second-best option, or the ‘lesser
evil’ in this case.3Apart from the National Centre of Employees (Centrale
nationale des employés, CNE),4the Belgian Transport Union (Belgische
Transportarbeidersbond/Union belge du transport, BTB/UBT) and Horval,
organising in transport and in food and catering respectively, and both affiliated
to the socialist General Federation of Belgian Labour (Algemeen Belgisch
Vakverbond/Fédération générale du travail de Belgique, ABVV/FGTB), were
also involved in the negotiations on a collective agreement, which were
suddenly halted after the decision of Deliveroo to end the SMart arrangement.5
Prior to the termination of the arrangement with SMart, there had already been
a rapprochement between the Riders Collective and the unions. Thus, a first
‘symbolic action’, by about thirty riders, took place in Brussels in July 2017,
with the logistical support of CNE and the transport workers union CSC-
Transcom, both affiliated to the Confederation of Christian Unions (Algemeen
Christelijk Vakverbond/Confédération des syndicats chrétiens, ACV/CSC)
(Dufresne et al. 2018; Vandaele 2017).6The riders were protesting against their
working conditions and, especially, the offshoring to Madagascar of Deliveroo’s
call centre for its service for French-speaking customers, which resulted in lay-
offs. Moreover, about 200 union activists were also protesting at McDonald’s
in Brussels as part of the international solidarity campaign behind the ‘Fight
for $15’ movement in the fast food sector in September 2017. McDonald’s had
been selected as a campaign target as this fast food company had been
cooperating with Uber Eats, which stood accused of working with ‘independent
After Deliveroo’s unilateral decision to change contractual terms, the Riders
Collective organised an ‘altershift’ involving forty riders in Brussels on 25
November 2017. This proto-strike was supported by the Christian unions, in a
low-profile manner, but the Collective was also open to the involvement of
3. The relationship between SMart and the unions (but also the employers’ associations) is at
least stressful in the other industries in which SMart operates.
4. CNE works closely with CSC-Transcom as the latter is considered the most appropriate
union for organising food delivery riders. CNE is involved in the negotiations as the riders
have been on a ‘white-collar’ employment contract with SMart.
5. The socialist white-collar union (Bond van Bedienden, Technici en Kaderleden/Syndicat
des employés, techniciens et cadres, BBTK/Setca) and the youth section of the socialist
union confederation are watching platform-based food delivery with interest, too. The latter
also holds true for the youth section of the CSC.
6. Apart from having some members within Deliveroo’s call centre in Brussels, CNE also had
members within SMart, facilitating the joint protocol between SMart and the food delivery
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
12 WP 2019.06
other unions. Riders also saw the support of ‘Critical Mass Brussels’7, as both
self-organised groups share vulnerabilities as cyclists on Brussels streets. Strike
actions in Brussels, including the occupation of the Deliveroo building by 15
to 20 riders, and in other cities against the obligation to move to self-employed
status were organised throughout January 2018, for which a strike fund was
set up (Cant 2018b). Although the actions were able to disrupt food delivery,
despite riders’ interchangeability, Deliveroo did not alter its payment model
and did not made any other concessions.
To conclude, most actions by Deliveroo riders have thus been concentrated in
Belgium’s capital, where a critical mass of them are members of the Riders
Collective. While combining smartphones with street protest has been rather
ephemeral in Belgium, this has not been the case in in several European
countries where the protests continue.
7. This is a citizens’ initiative reclaiming the streets for cyclists via protest rides on the last
Friday of every month.
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
3. Riders’ profile: male, young and
student; but also migrant
Deliveroo riders were mainly young and male and were, mostly, students.8
Table 2 summarises some of the demographic and other characteristics of the
surveyed riders. The vast majority were in formal education and, hence, worked
on fiscally more favourable student contracts, while the rest are referred to in
this working paper as ‘salaried workers’. The age structure of food delivery riders
is very skewed: their median age was 22 years (n=273), although the median
age stood respectively at 21 and 28 for students (n=227) and salaried riders
(n=46). Compared to men, women had a higher level of educational attainment
(p=0.02, FET). More than one-third of the riders were migrant workers, with
11 per cent from outside the European Union. Migrant riders were older
(M=24.5, SD=5.1) than Belgian-born ones (M=22.2, SD=4.1) (t(271)=4.09,
p=0.00) and their educational attainment was also higher (x̅(2)=18.19, p=0.00).
Most riders still lived with their parent(s) (see Drahokoupil and Piasna 2019:
14), with differences in terms of age (x̅(3)=66.10, p=0.00) and educational
attainment (p=0.00, FET): unsurprisingly, especially young riders with a non-
tertiary level of education (or lower) still lived with their parent(s). Equally,
students were more likely to live with their parent(s) (p=0.00, FET) than were
salaried riders. Finally, there was quite a degree of geographical concentration
in the survey sample: more than one-half of riders were working in Belgium’s
capital. The major cities outside Brussels for Deliveroo riders were Ghent
(n=52), Antwerp (n=32) and Liège (n=31); followed at some distance by Leuven
(n=9), Bruges (n=8), Waterloo (n=5) and Mechelen (n=2).
At face value, working for Deliveroo fits the vision of the platform economy,
offering workers freedom and flexibility. Entry barriers are indeed very low
and working hours extremely flexible. However, riders expressed grievances
that the flexibility of platform work was largely one-way, at the cost of greater
precarity in their working lives (Drahokoupil and Piasna 2019). Riders often
found that the flexibility they wanted and expected, in the form of control over
where and when they worked, was in fact not the flexibility they got. As one
rider put it in the survey: ‘We did not get work when we wanted.’ This was
because Deliveroo maintained tight control over shift patterns, and unilaterally
adjusted these in response to rising and falling demand. Riders were often
unable to book their preferred shifts and sometimes got disconnected entirely
from the app after rejecting too many shifts. Riders also perceived their
position vis-à-vis the platform as relatively weak, with limited scope for voicing
their concerns: ‘Deliveroo does not consider the advice of "its" couriers.’
8. The gender imbalance is similar to the generally low share of women in transport-related
activities in Belgium (Drahokoupil and Piasna 2019: 13).
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
14 WP 2019.06
Their perceived vulnerability was aggravated by the decision of Deliveroo to
terminate the cooperation with SMart. One rider asserted: ‘This is the end of
workers’ rights.’ This also highlights a lack of awareness among the riders of
alternative ways of representing their interests:
‘This is a bad thing because SMart was the only organisation able to
protect our rights in a more or less acceptable way.’
‘I think it is a shame; whilst working with SMart we had a sense of security
and felt we were being defended by them. Now we are out in the open and
we never know how much we'll earn.'
‘I don't think workers will have much negotiating power from now onnot
that we had plenty of it beforehand.’
The very limited autonomy that we have acknowledged in riders’ ability to
choose their shifts (in practice, there was a lack of transparency in shift
allocation as a result of the algorithm) was not compensated by income
security, since work for Deliveroo was characterised by low and intermittent
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
Source: Survey results.
Table 2 Demographic and other characteristics of Deliveroo riders in Belgium
Age groups
<=24 years
>=25 years
Non-tertiary education or lower
Employment contract
Salaried worker
Household status
Lives with father or mother, or both
Lives with spouse or partner
Lives with house- or room-mates
Outside Brussels
pay, insufficient and variable hours, and short-term involvement: half the
respondents had been working for Deliveroo for seven months or less. Riders
worked 23 hours in a month, on average, with a median of 17 hours. Short
working hours translated into relatively low monthly incomes: average gross
monthly income was €249, with a median of €177.
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
16 WP 2019.06
4. A very low unionisation rate
Although trade union membership growth has recently halted, Belgium is still
a country where the social custom of unionisation is relatively strong with
union density of about 55 per cent (Vandaele 2017). No recent data on density
in transport or restaurants are available. Yet, it can be assumed that
unionisation is high in transport due to the dominance of blue-collar workers,
who display above-average unionisation rates (Faniel and Vandaele 2010), and
quite high in restaurants because of the involvement of the longstanding
unions in the provision of unemployment benefits, the so-called ‘Ghent system
(Van Rie et al. 2011). Even so, the majority of Deliveroo riders in our survey
(n=272) were not unionised and direct experience of trade unions was also very
limited: 3 per cent (n=8) of the riders reported that they had been a union
member in the past, while about 4 per cent (n=12) had attended a union
gathering or meeting. Although the cost of union membership could hardly be
an issue in joining, as membership is free for most riders (see Section 2), only
seventeen Deliveroo riders were unionised at the time of the survey,
corresponding to a union density of 6 per cent in the survey sample. Fourteen
out of the seventeen unionised riders were male; thirteen had Belgian
nationality. Nine of the unionised riders were working in Brussels and five of
them in Liège.
In general, however, offering free membership seems to convince only a
minority of young people to unionise in the Belgian context (Delespaul and
Doerflinger 2018). Nevertheless, the youth organisations of the long-standing
unions are successful in terms of membership, and the unionisation rate
among young people is close to that of adults (Vandaele 2018b). In Deliveroo,
it was salaried riders in particular who were union members (p=0.00, FET),
pointing thus to the presence of other motives among riders for union
membership in Belgium. It is likely that such riders are in a relatively more
precarious labour market position, associated with higher risks of
unemployment. They are, therefore, probably more likely to join a union
because of the ‘Ghent system’. Indicative of this logic is that the median age of
unionised riders stood at 27, and that eleven of the unionised riders had a non-
tertiary education or lower; while four of them had a bachelor’s degree albeit
none a master’s. This reasoning does not exclude, however, that the union
membership of riders could also simply reflect the existence of reasons for
membership outside the context of the platform economy.
No distinction was made in the survey questionnaire between the long-
standing trade unions, with free membership for students, and the Riders
Collective, with free membership for all riders irrespective of employment type.
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
Although it could not be known if the riders in the questionnaire survey made
such a distinction themselves, it is reasonable to assume that they approached
this question in terms of the long-standing unions given their manifest role in
Belgian society (Faniel 2010). In fact, it is an open question as regards the
extent to which the Collective was well-known by riders at the time of the
survey (Lenaerts 2018). Notwithstanding that the relationship between the
Collective and the unions is not entirely free from tension, it is far less
conflictual than in some other countries, for example Italy (Tassinari and
Maccarrone 2018), such that the non-specification of organisational form is
less problematic for the analysis here.
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
18 WP 2019.06
5. Prevailing lack of trade union
As far as worker characteristics are concerned, Deliveroo riders in Belgium
were predominantly male and young, typically students, and often came from
a migrant background (see Section 3). This profile is not necessarily a drawback
for trade unions: in general, they can still rely upon a relatively high level of
social legitimacy especially among social groups exposed to economic
vulnerability, such as young people and migrant workers (Frangi et al. 2017;
Gorodzeisky and Richards 2019). Indeed, research on the attitudes and beliefs
that young people possess towards unions has repeatedly demonstrated that
the disconnect is largely a matter of a lack of awareness of, and knowledge
about, unions – but not of anti-unionism, at least in western Europe and in
English-speaking countries in the Global North (Tapia and Turner 2018; for
an overview, see Vandaele 2018b). Moreover, the literature points to the
presence of (critical) support for unions, pointing to a frustrated or unmet
demand for unionisation among young people; only a small minority of whom
hold negative opinions about unions in principle.
International findings such as these also hold true in a Belgian context, where
positive or critical trade union support is still widespread, although weaker in
Flanders, while the traditional social custom of union membership has become
a less important motive for unionisation among younger age categories
(Swyngedouw et al. 2016). While, to our knowledge, no recent survey results
are available in the Belgian context similar to ours about the propensity to
unionise (see however, Vendramin 2007), Figure 1 depicts the percentage of
people having ‘(very) much’ trust in trade unions among three age groups in
Flanders as a means of contextualising and illustrating the argument developed
in this working paper.9The direction of trust develops fairly similarly among
the age groups, although there are some exceptions, but the main point is that
trust in unions among the ‘youngest’ age group has been higher than the two
older age groups since the mid-2000s. Unless Deliveroo riders in Belgium
genuinely differ from their peers, there is little reason to believe that their
attitudes towards collective representation would considerably diverge from
those of workers outside the platform-based food delivery sector. It is
acknowledged, however, that labour market institutions, country diversity in
9. Four caveats should be made. First, the survey question is about trust in trade unions and
not about propensity to join a union. Second, the survey relates only to Flanders yet, as
already mentioned, it can be assumed that trust in unions is even higher in French-speaking
Belgium. Third, no distinction is made between respondents’ different employment
statuses. Again, union trust will be higher among employees in employment than it will be
among, for instance, people in management positions, entrepreneurs or business leaders.
Finally, and importantly, the age groups are very broadly defined.
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
the organisational landscape and riders’ ideological leanings and subjective
understandings of their own (occupational) identity can all influence the
decision on choosing a particular collective organisation for representing their
needs and interests (Jansen 2017; Tassinari and Maccarrone 2018; Newlands
et al. 2018).
Attitudes towards trade unions of the riders in the survey may be gauged
indirectly in response to the question ‘What is the main reason why you are
not currently a trade union member?’10 Table 3 demonstrates that there are
three main individual motives, of which two are almost equal in their relevance:
riders had not chosen to become a member either because they did not know
much about unions; or because they had never felt the need to do so. Compared
to native-born riders, a higher percentage of migrants considered that they
lacked knowledge about unions, while fewer of them questioned their utility.
The third most important reason was that Deliveroo riders, whether Belgian-
born or otherwise, had simply not been asked to join. This is closely linked to
the motive for non-membership amongst riders that Deliveroo did not have
union representation. Outspokenly negative feelings about unions, including
people who judge them outdated or no longer relevant, were found only among
a small minority. Additionally, some riders preferred to talk directly to
management or had doubts about the performance of unions either because
of their weakness, or because of a perceived lack of understanding of the needs
and interests of riders.
10. Respondents could tick only one option, while the options were randomised for each
respondent. The answer options and wording were inspired by Huiskamp and Smulders
(2010) and Tailby and Pollert (2011).
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
20 WP 2019.06
Figure 1 ‘(Very) much’ trust in trade unions by age group, Flanders, 1996-2018
Note: No data for the years 2001, 2003, 2007 and 2009.
Source: Based upon publicly available data from the ‘Vlaamse survey Sociaal-Culturele Verschuivingen (SCV-survey)’.
18-39 years 40-59 years 60 years and older
Grouping these motives into three major groups, lack of union knowledge or
agency (i.e. ‘lack of union exposure’) was the main motive for not being a
member, followed by ignorance of unions and then ‘hostility’ to them. Students
significantly more often referred to ignorance of unions, whereas the union
attitudes of salaried riders rather reflected a lack of union knowledge or agency
and were also rather more marked by union ‘hostility’ (p=0.02, FET). Salaried
riders thus had a more ambivalent understanding of unions. We found no
significant differences between union attitudes in terms of gender, age,
educational attainment, nationality or household composition.11
11. Household composition is too rough a proxy to measure parental union socialisation as no
information is known of the unionised status of either father or mother. Nevertheless, there
is some anecdotical evidence that such a socialisation might have played a role in riders’
awareness about long-standing union, especially if they are a student, as parents may have
advised them to contact unions after Deliveroo’s decision to end the SMart arrangement.
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
Source: Survey results.
Table 3 What is the main reason why you are not currently a trade union
member? (n=264)
Lack of trade union exposure
‘I don’t know much about trade unions and what they do’
‘I haven’t been asked by a trade union to join’
‘There is no trade union active within Deliveroo’
Trade union ignorance
‘I never felt the need to join a trade union’
‘I prefer talking directly to management’
Trade union ‘hostility’
‘I don’t like trade unions in general’
‘I don’t think trade unions are relevant’
‘Trade unions are too weak to make a difference
‘I don’t feel trade unions understand my needs’
6. Platform-based food delivery is
complementary to unionisation
In response to riders becoming mobilised, Deliveroo commonly claimed that
they were not representative of all workers. Yet, while the number of mobilised
riders might indeed be small, this is not surprising from a historical-sociological
perspective since union formation starts only once a critical mass has been
gained. Moreover, Figure 2 shows that almost forty per cent of riders believed
it possible that they might become a member, indicating that their union atti-
tudes are likely to be rather malleable. Put differently, a large part of riders
was undecided on whether or not to unionise, in accordance with their main
motive for not joining a trade union, i.e. their lack of knowledge of unions.
The survey results also demonstrate that, in general, only a small percentage
of riders was strongly oriented against unionisation, while a similarly small
percentage was, on the other hand, strongly convinced that they would join.
Finally, about one-quarter thought that they would probably not unionise
should they have a problem in their food delivery job, while an almost compa-
rable percentage believed that they would become a member in such a case.
Based on the grouped union attitudes, then riders with a lack of trade union
exposure had a mean level of 3.1 (n=147; SD=1.0) on a sliding scale towards
unionisation ranging from one (lowest) to five (highest), while this level stood
at 2.6 for riders characterised by union ignorance (n=78; SD=0.9) or union
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
22 WP 2019.06
Figure 2 Would you join a trade union if you had a problem in your Deliveroo job?
Source: Survey results.
Probability to unionise
Definitely not
Probably not
Tot al (n=262) Brussels (n=134) Other cities (n=128)
‘hostility’ (n=37; SD=1.1). There were no significant differences in propensity
to unionise when it came to gender, age, educational attainment, nationality
or employment contract.
All in all, such findings stand in contrast to the powerful Deliveroo narrative
stressing workers’ control and flexibility; that riders as ‘independent
contractors’ are not at all interested in unionising.
If a distinction is made between Brussels and the other Belgian cities where
Deliveroo was operating at the time of the survey, then the percentages shift.12
About one in four riders in Brussels believed that they would probably or
definitely not unionise, whereas this percentage stood at 42 per cent in the
other cities. The percentage of riders that would possibly unionise is similar in
and outside Brussels. Consequently, the percentage of riders outside the capital
who would probably or definitely join a union stood at 19 per cent, with the
figure in Brussels being 14 percentage points higher than this. There were,
however, no significant differences in the grouped attitudes towards unions of
riders in and outside Brussels (p=0.42, FET). The geographical bifurcation in
the likelihood of unionisation (p=0.03, FET) might reflect the intensity of the
riders’ mobilisation in Brussels, where a critical mass of them are members of
the Riders Collective, while this mobilisation was rather in its infancy in the
other Belgian cities at the time of the survey.13
The importance of union agency is further buttressed in that those riders who
had been in direct contact with trade unions, either because of previous
membership or by having attended a union meeting, were more likely to report
that they would join (M=3.40, SD=0.88) compared to riders without such past
union contact (M=2.86, SD=0.99) (t(261)=-2.34, p=0.02). The survey results
also hint at the importance of existing, prior action networks, like the Riders
Collective, network effects and group identification among riders. Riders who
had worked for other delivery platforms, such as Take Eat Easy or Uber Eats,
had a higher probability of contact with unions (p=0.00, FET). Riders who had
experience of different platforms were older (M=25.7, SD=5.7) than Deliveroo-
only riders (M=22.4, SD=4.1) (t(271)=-4.83, p=0.00); and were also more
likely to be salaried (p=0.00, FET).
Individual and labour market perspectives on job quality will next be brought
into the equation in order to further understand riders’ propensity towards
12. Cities outside Brussels are collapsed in the analysis as observations for each individual city
are low in number.
13. Male riders were more present in Brussels than women compared to cities outside Brussels
(p=0.00, FET).
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
7. Individual ‘fit’ and intention
to unionise
Seen from the outside, Deliveroo riders work in very similar conditions and
the quality of their jobs is comparable, being influenced by the very same app-
based technologies shaping their work arrangements and by the riders’ own
bargaining power (Rubery and Grimshaw 2001). The level of job quality they
experience is also influenced by the ‘fit’ between the individual circumstances
of the riders, including their life stage and need for income, and their
expectations of work and platform-based food delivery (Goods et al. 2019).
Subjective understandings of job quality are multi-dimensional at the
individual level. Three particular dimensions have been identified ethno -
graphically in the context of the platform-based food delivery economy: (1)
economic security; (2) autonomy over the work (see also Ivanova et al. 2018);
and (3) enjoyment at work. These dimensions are marked by interactions and
tensions within and between them, especially between ‘workers’ subjective
enjoyment of riding, the need to make money and work-related risks’ (Goods
et al. 2019: 14). These three dimensions of job quality are operationalised for
our purposes as follows (for details, see Drahokoupil and Piasna 2019).
Economic security refers to pay, income variability and economic risk, with its
assessment based on the question of how riders would cope with an unexpected
expense of €300. Four potential answer scenarios were set: ‘I would find it very
difficult to find the money’; ‘I could cover it myself by cutting back on other
expenditure or via a loan’; ‘I could cover it with the help of my family/others’
and ‘I could cover it myself without difficulties’. One-third of the surveyed
riders (n=91) said they would find it very difficult to cover an unexpected
expense of €300, which puts them in a category of ‘low’ economic security.
One-quarter of the riders (n=69) would only be able to cover it with help from
family or others, while 16 per cent (n=44) could cover it by cutting back on
expenditure; these two scenarios are regarded as ‘moderate’ in terms of
economic security. A further one-quarter (n=70) said they would be able to
cover such an expense themselves without difficulty; their economic security
is considered ‘high’.
Autonomy over the work is considered on the basis of a Likert-type scale
(Cronbach’s Alpha: 0.67), constructed from the following items: ‘I have control
over the pace of my work’; ‘I have control over the scheduling of my work’; ‘I
work under time pressure’; ‘Work is stressful’; and ‘Work has a negative impact
on my health and/or safety’. Thus, the items mainly refer to riders’ subjective
understandings of their working time.
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
24 WP 2019.06
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
Enjoyment at work reflects the opportunities for social interaction with riders,
restaurants and customers; as well as cycling itself. In the context of this
working paper, this enjoyment dimension could be measured only indirectly
via perceptions of the degree of social justice behind Deliveroo’s algorithmic
management, including its surveillance and discipline pressures, with the
engagement of the following items (Cronbach’s Alpha: 0.67): ‘I have been able
to communicate in a timely and effective way with Deliveroo in order to solve
problems’; ‘Considering all my efforts and performance, I get paid
appropriately by Deliveroo’; ‘Deliveroo adequately contributes to the costs of
my equipment (bike, clothing, mobile phone)’; and ‘Deliveroo has a fair system
of evaluating couriers’.
The cut-off points are set at 33 per cent and 66 per cent in order to create three
equal groups for these autonomy and enjoyment dimensions.
Trade unions are seen by workers as providing a collective voice regarding the
improvement of job quality (Freeman and Medoff 1984; Hartley 1992), and so
it is hypothesised that negative perceptions among Deliveroo riders on the
three dimensions will be associated with a higher propensity to unionise.
Figure 3 depicts the relationship between the three dimensions of job quality,
each categorised in terms of ‘low’, ‘moderate’ or ‘high’ job quality, and the
means of the expressed likelihood of riders to unionise on the basis of our scale
from one to five.
This demonstrates that willingness to join a union is indeed visibly higher
where there is stronger discontent among riders, i.e. where there are lower
levels of experienced job quality. The differences are significant for economic
security (x̅(2)=8.03, p=0.02) and enjoyment at work (x̅(2)=5.78, p=0.06),
Figure 3 The multi-dimensionality of job quality and propensity to unionise
Source: Survey results.
Probability to unionise
Low (n=84; n=92; n=80) Moderate (n=84; n=78; n=80) High (n=84; n=89; n=88)
Economic security Autonomy of work Enjoyment at work
although the latter only at the 10 per cent level, whereas the subjective
understanding of autonomy over the work is not significant (x̅(2)=2.36,
p=0.31). There are no differences in terms of gender, age, educational
attainment, household composition and grouped union attitudes for the three
Three significant differences could, however, be observed. The proportion of
migrant riders with strong economic security was lower compared to Belgians
(p=0.02, FET); riders from outside Brussels were relatively more negative
about their perceived autonomy over the work (M=3.0, SD=0.6) compared to
riders in Brussels (M=3.2, SD=0.5) (t(279)=-3.19, p<0.01); and salaried riders
were more critical of the algorithmic management of Deliveroo (M=2.4,
SD=0.7) than students (M=2.8, SD=0.8) (t(268)=3.01, p<0.01).
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
26 WP 2019.06
8. Labour market ‘fit’ and intention
to unionise
It is not only an individual assessment that influences the job quality of
Deliveroo riders, but also the labour market context and the wider socio-
political context (Goods et al. 2019).14 Whether riders perceive platform-based
food delivery as a labour market ‘fit’ will depend on their attachment to the
platform economy, i.e. their relative labour market position and their perceived
labour market alternatives or options.
We use the following indicators for assessing labour market ‘fit’. Labour market
attachment is measured by two continuous variables: the number of hours per
week that the riders had worked for Deliveroo in the past month; and net
monthly earnings from Deliveroo in the past month. In addition, the answer
categories to the question ‘Have you looked for another paid job since you
started working for Deliveroo?’ measure whether riders have available labour
market alternatives, and consist of ‘no’, ‘yes, in addition to the Deliveroo job’
and ‘yes, to replace the Deliveroo job’. Finally, the intention to continue
working for Deliveroo after the termination of the SMart arrangement is based
on the answer categories ‘no’, ‘yes’, and ‘I don’t know’.
14. The research design did not allow for assessing the wider socio-political context, i.e. the
questions of how and to what extent the platform economy should be regulated, especially
regarding employment classification, which would indirectly influence the costs and
benefits of unionisation.
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
Table 4 Statistically significant associations between labour market ‘fit’ and riders’
Riders’ characteristics
Employment contract
Household status
End of SMart
Note: n.s. = not significant.
Source: Survey results.
Labour market attachment
Table 4 provides an overview of the statistical significance between the four
indicators of labour market ‘fit’ and the demographic and other characteristics
of our respondents. Looking at gender, female riders were less attached to the
platform-based food delivery sector (M=232.6, SD=425.6) than men
(M=399.4, SD=208.9) in terms of earnings (t(242)=2.14, p=0.03). In other
words, their net monthly income from Deliveroo in the past month, at the time
of the survey, was lower. Women were also less likely to continue with
Deliveroo after the end of the SMart arrangement (p=0.01, FET). There were
no gender-based differences regarding labour market attachment based on
hours and labour market alternatives.
Concerning age, older workers tended to have a stronger labour market
attachment in terms of hours (r(262)=0.27, p=0.00) and earnings
(r(245)=0.51, p=0.00). Compared to younger riders, older riders were more
likely to be looking for work to replace the Deliveroo job (x̅(2)=6.15, p<0.05),
and they were also more likely to stop after the end of the SMart arrangement
(x̅(2)=9.63, p=0.01). Regarding educational attainment, there were differences
in labour market attachment based on earnings (x̅(2)=6.15, p<0.05) since
riders with a master’s degree had a higher monthly income than those with a
bachelor’s degree. Also, riders with a master’s degree were more likely to be
searching for a job as an alternative to their Deliveroo job (p=0.03, FET).
There were no differences concerning hours and the SMart arrangement.
Migrants were more attached to platform-based food delivery (M=453.6,
SD=390.1) than Belgian riders (M=344.1, SD=414.0) (t(246)=2.01, p<0.05) in
terms of earnings, but there were no other differences between them.
Regarding the employment contract, salaried riders were working more hours
per week on average (M=29.7, SD=34.9) than students (M=11.7, SD=22.7)
(t(273)=-4.43, p=0.00). Equally, salaried riders also earned more (M=793.6,
SD=618.23) than students (M=288.1, SD=269.3) (t(256)=-8.7, p=0.00). Put
differently, salaried riders had a stronger labour market attachment than
students, although there was no significant difference between them when it
came to labour market alternatives. Students were also less likely to continue
working for Deliveroo after the termination of the SMart arrangement (p=0.00,
FET). Riders living with their father or mother, or both, were working less than
those who lived with a spouse or partner who, in turn, worked less than riders
living with house- or room-mates (x̅(2)=19.5, p=0.00). The same pattern holds
true for average monthly income (x̅(2)=25.05, p=0.00). Finally, riders from
Brussels were working more hours (M=19.7, SD=34.2) than riders in other
cities (M=9.5, SD=10.5) (t(273)=-3.3, p=0.01).
What does all this imply for riders’ propensity to unionise? Platform-based
food delivery is a highly transient sector: most riders, especially students, per-
form platform work for a short time and tend to see food delivery as a side ‘gig’
in their school-to-work transition (Drahokoupil and Piasna 2019; see also De
Groen et al. 2016; Jan 2018). Given their short labour market tenure and the
presence of fewer labour market alternatives, the latter being composed sub-
stantially of ‘exit’ possibilities, it is hypothesised that riders will choose to cope
temporarily with low job quality, i.e. that they will remain loyal to Deliveroo,
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
28 WP 2019.06
thus lowering their incentives to invest in collective representation, and reduc-
ing the likelihood that they will view organisations of collective representation
as providing them with a voice mechanism (Freeman and Medoff 1984).
There is indeed some indication that the stronger the labour market
attachment based on hours and earnings, the higher the propensity to unionise.
Yet, attachment in terms of hours worked (x̅(4)=8.42, p=0.08) and earnings
(x̅(4)=8.20, p=0.08) are only significant at the 10 per cent level. Concerning
the intentions of riders to seek labour market alternatives, there are no
significant differences in propensity to unionise between riders who were not
looking for a new job, those who wanted to replace their Deliveroo job and
those who were searching for a job additional to Deliveroo. There are also no
differences in intention to unionise between riders who wanted to stay with
Deliveroo after the end of the SMart arrangement, those who wanted to quit
and those who did not yet know. These results are somewhat encouraging from
a union perspective as they indicate that organising campaigns are not
necessarily doomed to fail due to high labour turnover.
Figure 4 depicts mean levels in the propensity to unionise for each combination
of labour market alternatives and riders’ attitudes to unions – although it should
be noted that the number of observations is very low for several groupings. Nev-
ertheless, it is again demonstrated that riders with a lack of union exposure had
a higher intention to unionise compared to riders with other attitudes towards
unions. Moreover, riders who were thinking about finding another job than De-
liveroo were also more willing to unionise, while this was less likely to be the case
for riders who were simply staying in their Deliveroo job. In other words, think-
ing about exit is not exclusive to collective voice, while loyalty to Deliveroo means
indeed that collective representation and voice are relatively less appealing.
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
Figure 4 Propensity to unionise by labour market alternatives and union attitudes
Source: Survey results.
Probability to unionise
Low (n=10; n=11; n=9) Moderate (n=17; n=27; n=25)
union 'hostility' union ignorance lack of union exposure
Labour market alternatives
High (n=44; n=43; n=49)
This working paper has explored issues of collective action and unionisation
in the platform economy. The aim was to investigate whether platform workers
are substantially different from the offline workforce in their attitudes towards
trade unions, and whether the possibilities for organising them would also be
fundamentally different. In doing this, Deliveroo, a place-based digital labour
platform operating in the Belgian food delivery sector was studied. We
conducted a survey among Deliveroo riders and analysed the Belgian context
in terms of the regulatory framework and the mobilisation actions undertaken
by riders within the Riders Collective, being similar to the ‘alt-unions’ in the
US-context, and with the support of the long-standing unions.
Our analysis shows that the low rate of unionisation in this group of platform
workers is largely a matter of a lack of knowledge about trade unions, and a
lack of contact with them as opposed to union ignorance or ‘hostility’ towards
unions. Put differently, the predominately young student workers working for
Deliveroo do not greatly differ from their peers in terms of their attitudes
towards unions. They are not essentially hostile towards unions and do not
perceive unions as ill-suited to represent them vis-à-vis the platform; and we
indeed observed several grassroots initiatives for collective action in the period
under investigation.
The results also confirm that subjective understandings of job quality play an
important role in influencing riders’ decisions about joining a union. In
particular, a perceived lack of economic security provides a strong incentive to
unionise among riders, whereas the impact on unionisation of autonomy over
the work and enjoyment at work is less clear-cut. We also find partial support
for the argument that platform workers, similar to workers in general, view
unions as a positive, collective voice for improving job quality. The influence
of labour market context on the propensity to unionise can thus be
downplayed, which is somewhat encouraging from a union perspective since
it indicates that developing tailor-made strategies for organising platform-
based food delivery workers is not necessarily ill-fated due to high labour
turnover. Trade unions can act as a stable actor offering continuity of
experience and knowledge in this fluid sector.
Our findings cannot be generalised to all types of platform work because there
are other barriers to collective mobilisation inherent in platform work, which
is not place-based (Newlands et al. 2018). Grassroots action is particularly
present in the delivery- and transport-based platform economy (Vandaele
2018a). Despite many similarities with other types of digital labour platforms,
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
30 WP 2019.06
such as workers’ atomisation, hyper-flexibility and algorithmic management,
workers’ demographic and other characteristics, such as their power resources
and capabilities, are different. Yet, Deliveroo riders are not a different ‘species’
in comparison with young people in general, which is likely to imply, ceteris
paribus, that workers in other types of digital labour platforms will not
substantially differ in their union attitudes from workers in an employment
relationship within the same sector.
In addition, a number of other factors not included in the analysis could also
influence the propensity to unionise, such as the occupational identity of riders
and their ideological or political beliefs. There might also be measurement
errors because the study coincided with a mobilisation amongst riders
following the termination of the SMart arrangement. This might, in part,
explain their generally ‘positive’ attitudes towards unions; yet, as already noted,
those attitudes are very much in line with those of young people in general.
Furthermore, the Belgian regulatory arrangements regarding the platform
economy and its trade union context evidently differ from other countries in
which the relationships between riders, their self-organised structures and
long-standing unions can be more contentious. The recently-founded
Transnational Federation of Couriers in October 2018, set up by riders’ ‘alt-
unions’, whether or not supported by the long-standing unions, and grassroots
unions, from eleven west European countries (Dufresne 2018), might offer a
forum in which to exchange experiences and views, and might stimulate
mutual learning.
The absence of union agency and tailored union organising strategies at the
time of the study points to the novelty of the platform economy in Belgium.
Although still relatively marginal in terms of employment (Lenaerts 2018), the
case of food delivery platform work might offer opportunities for Belgian
unions to rediscover a more systematic organising approach based on managed
activism, and to set up small-scale organising experiments, not bound to a
physical workplace and beyond the union’s traditional realm, as a means of
creatively engaging with this specific group of workers. Unions might support
riders with leadership training and education programmes, which would also
help to overcome the lack of union exposure, potentially turning them into
future union activists. Belgium’s long-standing unions are catching up in this
domain (Wartel 2018) by developing organising and other strategies towards
Finally, the riders in our study were predominately young and were students:
engaging with them can offer unions a window of opportunity to win trust
among young people and demonstrate unions’ relevance in their school-to-
work transition, particularly as those formative experiences could influence
future attitudes towards unionism (Vandaele 2018b). It remains to be seen,
however, if such an active engagement will result in unionisation once students
enter stable careers in other sectors. Moreover, platform-based food delivery
enables trade unions also to reach out to migrants and to more precarious
salaried workers with a stronger attachment to platform work. It could also
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
provide unions with new legitimacy in other sectors confronted with digital
labour platforms. Organising riders, building collective identity and fostering
solidarities between them will require imagination, tactical creativity and
tailored long-term strategies centred around, for instance, dimensions of job
quality, which take into account the individual and labour market ‘fit’ of these
machine-breakers of the twenty-first century and which, ultimately, advances
the cause of social justice in the platform economy.
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
32 WP 2019.06
Aloisi A. (2016) Commoditized workers: case study research on labour law issues arising
from a set of “on-demand/gig economy” platforms, Comparative Labor Law and
Policy Journal, 37 (3), 653-690.
Animento S., Di Cesare G. and Sica C. (2017) Total eclipse of work? Neue Protestformen
in der gig economy am Beispiel des Foodora Streiks in Turin, Prokla, 47 (187),
Boxall P. (2008) Trade union strategy, in Blyton P., Bacon N., Fiorito J. and Heery E. (eds.)
The Sage handbook of industrial relations, London, Sage, 209-224.
Cant C. (2017) Precarious couriers are leading the struggle against platform capitalism,
Political Critique, 3 August 2017.
Cant C. (2018a) The wave of worker resistance in European food platforms 2016-17,
No Politics Without Inquiry! (#1), 29 January 2018.
Cant C. (2018b) #Slaveroo: an interview with Belgian riders, The Transnational Courier
Federation (#4.2), 31 October 2018.
Degner A. and Kocher E. (2018) Arbeitskämpfe in der „Gig-Economy“? Die
Protestbewegungen der Foodora- und Deliveroo-„Riders“ und Rechtsfragen ihrer
kollektiven Selbstorganisation, Kritische Justiz, 51 (3), 247–265.
De Groen W.P., Maselli I. and Fabo B. (2016) The digital market for local services: a one-
night stand for workers? An example from the on-demand economy, CEPS Special
Report No. 133, Brussels, European Centre for Political Studies.
Delespaul S. and Doerflinger N. (2018) Een verlaagde prijs als motief voor
vakbondslidmaatschap voor jongeren? Inzichten uit België, Tijdschrift voor
Arbeidsvraagstukken, 34.
De Stefano V. (2017) Labour is not a technology: reasserting the Declaration of
Philadelphia in times of platform work and gig economy, IUSLabor, 2/2017, 1-16.
Drahokoupil J. and Fabo B. (2016) The platform economy and the disruption of the
employment relationship, Policy Brief 5/2016, Brussels, ETUI.
Drahokoupil J. and Piasna A. (2017) Work in the platform economy: beyond lower
transaction costs, Intereconomics: Review of European Economic Policy, 52 (6),
Drahokoupil J. and Piasna A. (2019) Work in the platform economy: Deliveroo riders in
Belgium and the SMart arrangement, Working Paper 2019.01, Brussels, ETUI.
Dufresne A., Leterme C. and Vandewattyne J. (2018) Les mobilisations du collectif des
coursier.e.s contre Deliveroo, in Gracos I. (ed.) Grèves et conflictualité en 2017,
Courrier hebdomadaire No. 2383-2384, Bruxelles, Centre de recherche et
d’information socio-politiques, 45–58.
Dufresne A. (2018) Coursiers de tous les pays, unissez-vous ! La naissance de la
Fédération Transnationale des Coursiers, Smart, 18 December 2018.
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
Faniel J. (2010) Caractéristiques et spécificités des syndicats belges, in Arcq E., Capron
M., Léonard E. and Reman P. (eds.) Dynamiques de la concertation sociale, Bruxelles,
Centre de recherche et d’information socio-politiques, 93–119.
Faniel J. and Vandaele K. (2010) Implantation syndicale et taux de syndicalisation
(2000-2010), Courrier hebdomadaire No. 2146-2147, Bruxelles, Centre de recherche
et d’information socio-politiques.
Finkin M. (2016) Beclouded work, beclouded workers in historical perspective,
Comparative Labor Law and Policy Journal, 37 (3), 603–618.
Frangi L., Koos S. and Hadziabdic S. (2017) In unions we trust! Analysing confidence in
unions across Europe, British Journal of Industrial Relations, 55 (4), 831–858.
Freeman R. and Medoff J. (1984) What do unions do?, New York, Basic Books.
Goods V., Veen A. and Barratt T. (2019) “Is your gig any good?” Analysing job quality in
the Australian platform-based food-delivery sector, Journal of Industrial Relations.
Gorodzeisky A. and Richards A. (2019) Do immigrants trust trade unions? A study of 18
European countries, British Journal of Industrial Relations.
Hartley J.F. (1992) Joining a trade union, in Hartley J.F. and Stephenson G.M. (eds.)
Employment relations: the psychology of influence and control over work, Oxford,
Blackwell, 163-183.
Heckscher C. and McCarthy J. (2014) Transient solidarities: commitment and collective
action in post industrial societies, British Journal of Industrial Relations, 52 (4),
Hobsbawm E. (1952) The machine breakers, Past & Present, (1), 57–70.
Huiskamp R. and Smulders P. (2010) Vakbondslidmaatschap in Nederland: nooit serieus
over nagedacht!, Tijdschrift voor Arbeidsvraagstukken, 26 (2), 197–210.
Ivanova M., Bronowicka J., Kocher E. and Degner A. (2018) The app as a boss? Control
and autonomy in application-based management, Arbeit | Grenze | Fluss – Work in
Progress interdisziplinärer Arbeitsforschung Nr. 2, Frankfurt (Oder), Viadrina.
Jan A. (2018) Livrer à vélo… en attendant mieux, La nouvelle Revue du Travail [Online],
Jansen G. (2017) Solo self-employment and membership of interest organizations in the
Netherlands: economic, social, and political determinants, Economic and Industrial
Jehin A. (2018) Coursiers à vélo et Deliveroo : les enseignements d’un combat social,
Smart, 6 July 2018.
Jesnes K., Ilsøe A. and Hotvedt M.J. (2019) Collective agreements for platform workers?
Examples from the Nordic countries, Nordic future of work Brief 3, Oslo, Fafo.
Kalleberg A.L. and Dunn M. (2016) Good jobs, bad jobs in the gig economy, Perspectives
on Work, 20, 10–14.
Lenaerts K. (2018) Industrial relations and social dialogue in the age of collaborative
economy (IRSDACE), National report Belgium, Brussels, Centre for European Policy
Lenaerts K., Beblavy M. and Kilhoffer Z. (2017) Government responses to the platform
economy: where do we stand?, CEPS Policy Insights 2017-30, Brussels, Centre for
European Policy Studies.
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
34 WP 2019.06
Milkman R. (2013) Back to the future? US labour in the new gilded age, British Journal
of Industrial Relations, 51 (4), 645–665.
Newlands G., Lutz C. and Fieseler C. (2018) Collective action and provider classification
in the sharing economy, New Technology, Work and Employment, 33 (3), 250–267.
Oswalt M.M. (2016) Improvisational unionism, California Law Review, 104 (3), 597-670.
Pasquier V. and Wood A.J. (2018) The power of social media as a labour campaigning
tool: lessons from OUR Walmart and the Fight for 15, Policy Brief 10/2018, Brussels,
Rubery J. and Grimshaw D. (2001) ICTs and employment: the problem of job quality,
International Labour Review, 140 (2), 165-192.
Strikwerda C. (1997) A house divided. Catholics, socialists, and Flemish nationalists in
nineteenth-century Belgium, Lanham, Rowman and Littlefield.
Swyngedouw M., Abts K. and Meuleman B. (2016) Syndicats et syndicalisme :
perceptions et opinions, Courrier hebdomadaire No. 2298, Bruxelles, Centre de
recherche et d’information socio-politiques.
Tailby S. and Pollert A. (2011) Non-unionized young workers and organizing the
unorganized, Economic and Industrial Democracy, 32 (3), 499-522.
Tapia M. and Turner L. (2018) Renewed activism for the labor movement: the urgency of
young worker engagement, Work and Occupations, 45 (4), 391–419.
Tassinari A. and Maccarrone V. (2017) The mobilisation of gig economy couriers in Italy:
some lessons for the trade union movement, Transfer, 23 (3), 353–357.
Tassinari A. and Maccarrone V. (2018) Varieties of unionism meet the platform economy:
a comparison of gig workers’ organizing practices and trade union responses in Italy
and the UK, paper presented at Industrial Relations in Europe Conference (IREC),
10-12 September 2018, Leuven, KU Leuven.
Vandaele K. (2017) Belgique. Les syndicats sur le qui-vive pour soutenir les travailleurs
des plateformes : l’exemple des livreurs de repas, Chronique internationale de l’IRES,
160, 85-100.
Vandaele K. (2018a) Will trade unions survive in the platform economy? Emerging
patterns of platform workers’ collective voice and representation in Europe, Working
Paper 2018.05, Brussels, ETUI.
Vandaele K. (2018b) How can trade unions connect with young workers?, in O’Reilly J.,
Leschke J., Ortlieb R., Seeleib-Kaiser M. and Villa P. (eds.) Youth labor in transition:
inequalities, mobility, policies in Europe, New York, Oxford University Press, 660-688.
Van Rie T., Marx I. and Horemans J. (2011) Ghent revisited: unemployment insurance
and union membership in Belgium and the Nordic countries, European Journal of
Industrial Relations, 17 (2), 125–139.
Vendramin P. (2007) Les jeunes, le travail et l’emploi : enquête auprès des jeunes
salariés en Belgique francophone, Namur, Fondation Travail-Université (FTU).
Wartel L. (2018) Les enjeux politiques et juridiques du travail dans l’économie des
plateformes : état des lieux et perspectives syndicales, Working paper #3, Louvain-la-
Neuve, Chaire Travail-Université.
Wood A.J. (2015) Networks of injustice and worker mobilisation at Walmart, Industrial
Relations Journal, 46 (4), 259-274.
Wood A.J., Graham M., Lehdonvirta V. and Hjorth I. (2019) Networked but commodified:
the (dis)embeddedness of digital labour in the gig economy, Sociology,
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
Xhauflair V., Huybrechts B. and Pichault F. (2018) How can new players establish
themselves in highly institutionalized labour markets? A Belgian case study in the
area of project-based work, British Journal of Industrial Relations, 56 (2), 370–394.
Zamponi L. (2018) Bargaining with the algorithm, Jacobin, 9 June 2018.
All links were checked on 01.04.2019.
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
36 WP 2019.06
‘Algorithm breakers’ are not a different ‘species’: attitudes towards trade unions of Deliveroo riders in Belgium
WP 2019.06
Note: *The platform cooperative Coopcycle is also active in these countries.
Source: authors’ own typology, based on Degner and Kocher (2018), Dufresne (2018) and grey literature.
Table 5 A non-exhaustive overview of trade unions active in platform-based food
delivery in western Europe
Long-standing union: WhatsApp groups of Foodora riders are supported by Vida
Alt-union: Riders Collective
−— Long-standing unions: BBTK/Setca, BTB/UBT, CNE, Horval and LBC-NVK
Alt-union: Finnish Courier Collective (FCC), running the ‘justice4couriers’ campaign (Oikeutta läheteille).
The Foodora Take Responsibility campaign (Foodora Vastuuseen) was a predecessor of this campaign.
Long-standing union: Service Union United (Palvelualojen ammattiliitto, PAM)
−— Alt-unions: Bikers Nantais (from Nantes), Collectif des coursiers de Lille Métropole (CCLM, Lille), Collectif des
livreurs autonomes de Paris (CLAP) and Syndicat des coursiers à vélo de la Gironde (SCVB, Bordeaux).
Coordination d’actions vers l’autonomie des livreurs (CAVAL) is a national structure providing national
coordination for the different rider-based alt-unions in terms of mobilisation and information.
−— Grassroots union: SUD commerces et services (SUD stands for solidaires, unitaires, démocratiques)
−— Long-standing union: French Democratic Confederation of Labour (Confédération française démocratique du
travail, CFDT) and General Confederation of Labour (Confédération générale du travail, CGT)
−— Grassroots union: Deliverunion, set up by the Free Workers’ Union (Freie Arbeiterinnen und Arbeiter-Union, FAU)
−— Long-standing union: Liefern am Limit, as a project of the Food, Beverages and Catering Union (Gewerkschaft
Nahrung-Genuss-Gaststätten, NGG)
−— Alt-unions: Deliverance Project (from Turin), Deliveroo Strike Riders (from Milan), Deliverance Milano, Riders on
the Storm – Padova, Riders Union Bologna and Riders Union Roma
−— Grassroots union: (initially) SI-COBAS (in Turin)
−— Long-standing union: The Riders Union, which first started as an alt-union, joined the Federation of Dutch Trade
Unions (Federatie Nederlandse Vakbeweging, FNV).
−— Long-standing union: Oslo Transportworkers’ Union (Oslo Transportarbeiderforening), being part of the
Norwegian Transport Workers’ Union (Norsk Transportarbeiderforbund, NTF)
−— Alt-unions: Riders X Derechos
−— Grassroots union: Intersindical Alternativa de Cataluña
−— Grassroots union: Örestad LS
−— Alt-unions: Collectif des Coursiers/Livreurs de Genève and Velo-Kurierplatform Notime Zürich
−— Long-standing unions: Syndicom and UNIA, which supports the alt-unions.
United Kingdom
−— Grassroots unions: Independent Workers Union of Great Britain (IWGB) and Industrial Workers of the World
Kurt Vandaele, Agnieszka Piasna and Jan Drahokoupil
38 WP 2019.06
Table 6 Non-litigation successes of riders in platform-based food delivery in
western Europe
−— A works council was founded in Foodora in Vienna in 2017.
A works council was elected in Foodora in Cologne in 2017 and established one year later. Riders have also
elected a works council in Deliveroo in Cologne in 2018, but its establishment was obstructed by means of a
phasing-out of riders’ employment status from being based on part-time contracts to ‘self-employed contractors’.
Foodora riders also elected a works council in Hamburg in 2018. Equally, it looks like a works council is on its
way in Nuremberg as an election committee (Wahlvorstand) has recently been set-up, in March 2019.
A charter was signed between Riders Union Bologna in May 2018, the three main trade union confederations,
the centre-left city council and the local food delivery platform Sgnam e MyMenu. The charter, labelled Charter
on fundamental digital work rights in the urban context (Carta dei diritti fondamentali del lavoro digitale nel
contesto urbano), sets, on a voluntary basis, a framework of minimum standards covering remuneration, working
time and insurance cover to be respected by the signatory platforms. International platforms, such as Deliveroo,
Foodora and JustEat, have not signed the Charter.
The Norwegian Transport Workers’ Union and with Foodora Norway are currently, i.e. March 2019, negotiating a
collective agreement. The riders are accepted by Foodora Norway as employees (see Jesnes et al. 2019).
−— Syndicom signed a collective agreement with the courier employer’s association Swissmessengerlogistics (SML)
in February 2019.15 It is claimed that the agreement is the first for delivery riders in urban settings in western
Europe. The agreement sets minimum standards for approximately 600 riders. Since the agreement has not (yet)
been extended, platform-based riders are not covered by it, although the agreement aims to avoid ‘social
dumping’ by the digital labour platforms engaged in food delivery.
15. For details, see
Source: authors’ own compilation, based on grey literature.
ETUI Working Papers
A European Social Semester? The European Pillar of Social Rights in practice
Working Paper 2019.05
Björn Hacker
Phasing out coal – a just transition approach
Working Paper 2019.04
Bela Galgóczi
Work in deregulated labour markets: a research agenda for precariousness
Working Paper 2019.03
Valeria Pulignano
How to redesign the fiscal regime of the Eurozone: an alternative take on lessons
from US and Eurozone experiences
Working Paper 2019.02
Jörg Bibow
Work in the platform economy: Deliveroo riders in Belgium and the SMart
Working Paper 2019.01
Jan Drahokoupil and Agnieszka Piasna
Corporate income tax in the EU, the Common Consolidated Corporate Tax Base
(CCCTB) and beyond: is it the right way to go?
Working Paper 2018.06
Christian Valenduc
Will trade unions survive in the platform economy?
Emerging patterns of platform workers’ collective voice and representation in
Working Paper 2018.05
Kurt Vandaele
Sectoral social dialogue in professional football: social partners, outcomes and
problems of implementation
Working Paper 2018.04
Berndt Keller
Drifting into labour market insecurity? Labour market reforms in Europe after 2010
Working Paper 2018.03
Sotiria Theodoropoulou
These publications can be downloaded free of charge from our website.
Please visit:
... The job quality of the food delivery platform industry, which includes salary amount, enjoyment, autonomy, or other diverse factors, is not so good (Goods, Veen, and Barratt 2019), (Vandaele, Piasna, and Drahokoupil 2019). Although deliverer is an emerging occupation resulting from the booming of online commerce, social controversies, such as the high rate of road accidents in China due to the prevalence of unsafe riding and poor working conditions of delivery riders in Korea, occur (Pan, Wu, and Olson 2017). ...
ABSTRACT This study aims to answer the following research question: What is the difference in the open innovation and multi-homing of the smart delivery industry among Cardiff, Daegu, and Nanjing according to the maturity of the restaurant industry of the capitalist economies they belong to? By comparing open innovation and multi-homing of delivery platforms of the three cities representing different alongside the maturity of the restaurant industry in capitalist economies, the evaluation dynamics and focal points of the delivery platform industry, this study is based on an interview method in combination with participatory observation of deliverers, customers and restaurants of the three cities: (1) Cardiff with matured restaurant industry; (2) Daegu with an unmatured restaurant industry; (3) Nanjing with the growth of the restaurant industry. The findings of this research are as follows: (1) Existing industries can disturb the growth of the delivery platform industry; (2) Multi-homing motivates a high labour state of deliverers, the acceptance of restaurants by customers and customer surplus; (3) Motivating open innovation in delivery platforms can maintain a high level after maturity stage. The study concludes that the balance between open innovation and the multi-homing of a three-sided delivery platform is the way to sustainable development to conquer the effects of the gig economy.
... The new forms of work also make collective action difficult, weakening the effectiveness of traditional ways of worker organising (cf. Vandaele et al. 2019). At the same time, internet and platform work can benefit workers by lowering barriers to employment, enabling wider labour market participation and potentially providing a stepping stone to the labour market (Mandl 2019). ...
... The new forms of work also make collective action difficult, weakening the effectiveness of traditional ways of worker organising (cf. Vandaele et al. 2019). At the same time, internet and platform work can benefit workers by lowering barriers to employment, enabling wider labour market participation and potentially providing a stepping stone to the labour market (Mandl 2019). ...
Full-text available
This working paper presents the results of the ETUI Internet and Platform Work Survey conducted in Bulgaria, Hungary, Latvia, Poland and Slovakia in 2018-2019. The objective is to map the extent of digital labour in central and eastern Europe (CEE). We analyse two types of online sources for generating income: internet work; and its subset, platform work. We find that past experiences with generating income on the internet are relatively common among working age adults. However, the prevalence of regular internet and platform work remains very low in all five CEE countries; indeed, lower according to our estimates than in other comparative surveys. We attribute the differences to the inconsistent quality of non-representative samples of internet users that were deployed in other studies and, in particular, the use of paid, opt-in online surveys which themselves are examples of online gig work. We do not find evidence that internet and platform work is creating a qualitatively new labour market that encroaches on traditional age and gender segmentation. Neither is it a market of ‘student jobs’. Moreover, the labour market situation of internet and platform workers was somewhat more precarious than that for employed people generally, with a higher incidence of non-standard and fragmented employment. Finally, services requiring higher skills and creativity were among the least prevalent forms of internet work, suggesting little overlap with the knowledge-based economy.
... Comme le reconnaît l'OCDE(OECD 2019 : 199), les « données ne confirment pas les assertions relatives au faible intérêt des jeunes travailleurs pour l'action collective, qui expliquerait les asymétries dans la pyramide des âges des membres ». Les travailleurs, jeunes pour la plupart, du secteur de la livraison de plats préparés ne se différencient pas de manière significative de leurs homologues travaillant en dehors de l'économie de plate-forme et n'adoptent généralement pas une attitude négative vis-à-vis des syndicats (en Belgique en tout cas)(Vandaele et al. 2019). Plutôt qu'une absence de convictions et de valeurs collectives, il existe d'autres raisons plus importantes pour expliquer les difficultés des syndicats à recruter et à mobiliser les jeunes travailleurs : ces jeunes sont pour la plupart employés dans des lieux de travail, des emplois et des secteurs où la norme sociale d'une affiliation syndicale est pratiquement absente. ...
Full-text available
Ce rapport présente des données descriptives concernant les évolutions du nombre de membres et le taux de syndicalisation au niveau agrégé, dans 32 pays européens depuis 2000, en se focalisant tout particulièrement sur la pyramide actuelle des âges au sein des syndicats. Les variations entre pays en termes de taux de syndicalisation demeurent élevées, et les pays les moins bien classés des années 2000 sont très généralement restés dans le bas du classement du palmarès du taux de syndicalisation dans les années 2010 ; les pays présentant un taux moyen ou élevé de syndicalisation ont conservé leur position au milieu ou en tête de ce classement, quels que soient les développements de la syndicalisation. Les pays présentant un taux de syndicalisation moyen ou élevé ont généralement été en mesure de limiter l’érosion des membres, voire de les augmenter dans certains cas. Toutefois, de manière générale, le nombre des membres et le taux de syndicalisation s’inscrivent à la baisse dans la plupart des pays, tout particulièrement en Europe centrale et orientale. Non seulement les syndicats doivent lutter pour que le taux de syndicalisation suive le rythme d’un taux d’emploi à la hausse, mais ils doivent également faire face à un phénomène de vieillissement menaçant leur existence. Si l’écart en termes de syndicalisation entre travailleurs plus jeunes et plus âgés n’est pas nouveau, le nombre de jeunes qui se sont affiliés à un syndicat au cours de ces dernières années s’est encore réduit dans la plupart des pays européens. L’âge moyen des membres des syndicats augmente et il est supérieur à la moyenne d’âge des travailleurs et des salariés en général. Simultanément, même si les jeunes ont généralement une opinion positive des syndicats, leurs connaissances à ce sujet sont insuffisantes. Comme la réduction des effectifs va de pair avec une baisse des revenus, il convient de voir dans quelle mesure et comment les syndicats relèveront à la fois les défis de la baisse des effectifs et ceux du changement de génération. Beaucoup dépendra de l’identité des syndicats, de la manière dont ces défis seront encadrés, de la conception que les syndicats se font de l’affiliation et de leur dépendance aux cotisations des membres pour leur pérennité en tant qu’organisation. Dans tous les cas, l’importance du défi générationnel et de celui des affiliations forcera les syndicats, à côté de leurs propres initiatives de recrutement et d’organisation, à s’adresser à d’autres acteurs pour contribuer à préserver leur existence même.
... Tal como reconoce la OCDE (OCDE 2019:202): los "datos no respaldan de manera clara que el menor interés de los trabajadores jóvenes por la acción colectiva sea el factor que impulsa la diferencia de edad en la afiliación". Asimismo, los jóvenes que predominan en el ámbito de la distribución de alimentos mediante una app no son significativamente diferentes de aquellos que trabajan fuera de la economía de plataforma, que no tienen, por lo general, una actitud negativa sobre la sindicalización (el menos en el contexto belga) (Vandaele et al. 2019). Más que una deficiencia de creencias y valores colectivistas, detrás de la dificultad que tienen los sindicatos para atraer y organizar a los trabajadores jóvenes hay otros motivos: están predominantemente empleados en lugares de trabajo, ámbitos profesionales e industrias en los que la norma social de la afiliación sindical es prácticamente inexistente. ...
Full-text available
Las diferencias en los niveles de sindicalización entre los diversos países siguen siendo sustanciales y, en general, los países menos sindicalizados en el periodo 2000-2009 han seguido a la cola en la década posterior, mientras que los países con un nivel medio o alto también han mantenido esta posición, con independencia de la forma en que ha evolucionado la sindicalización. Los países que presentan una densidad sindical moderada o alta han sido capaces, por lo general, de limitar el descenso de la afiliación o incluso de aumentarla. Sin embargo, en conjunto, las tasas de afiliación y de densidad sindical están disminuyendo en la mayoría de los países, y muy drásticamente en Europa Central y Oriental. Los sindicatos no solo están teniendo dificultades para que la afiliación evolucione en línea con las crecientes tasas de desempleo, sino que están también envejeciendo. Aunque la brecha existente entre la sindicalización de los trabajadores jóvenes y mayores no es nueva, la cifra de personas jóvenes que han accedido a un sindicato ha retrocedido en los últimos años en la mayoría de los países europeos. La edad media de los afiliados está creciendo y supera a la de los trabajadores asalariados en general. Simultáneamente, aunque los jóvenes suelen tener una visión positiva de los sindicatos, carecen de conocimientos sobre ellos. Ante esta pérdida simultánea de afiliación e ingresos, queda por conocer en qué medida serán capaces los sindicatos de abordar los retos generacionales y de afiliación, algo que dependerá en buena medida de la identidad del sindicato y de la formulación de los desafíos, la concepción de la afiliación y su recurso a las cuotas de afiliación como medio para lograr su sostenibilidad organizativa. En cualquiera de los casos, la magnitud de los desafíos generacionales y de afiliación también exige que los sindicatos tiendan la mano a otros interesados en ayudar a reforzar la seguridad sindical, además de aplicar las propias iniciativas (de captación y organización).
... As the OECD (OECD 2019: 202) acknowledges: the 'data do not support strong claims about young workers' weaker interest in collective action driving the age-related membership differential'. Also, the predominantly young workers in app-based food delivery are not significantly different from their counterparts outside the platform economy, generally not holding negative union attitudes (at least in the Belgian context) (Vandaele et al. 2019). Rather than a deficiency of collectivist beliefs and values, there are other, more significant reasons for unions' difficulties in engaging and organising young workers: they are predominantly employed in workplaces, occupations and industries in which the social norm of union membership is nearly absent. ...
Full-text available
This report provides descriptive evidence of shifts in trade union membership and density at the aggregated level in 32 European countries since 2000, with a particular focus on the current age structure within trade unions. Substantial cross-country variation in unionisation levels remains, and the least unionised countries in the 2000s have largely stayed at the bottom of the ‘unionisation league’ in the 2010s, while countries with a medium and high average level have maintained their positions in the middle or at the top, irrespective of developments in unionisation. Countries with a moderate to high union density have generally been able either to limit membership shrinkage, or to increase membership. Yet, overall, membership and density rates are heading downwards in most countries, and very drastically in Central and Eastern Europe. Trade unions are not only struggling to keep membership developments in line with growing employment rates but are also ‘greying’. While the unionisation gap between younger and older workers is not new, fewer young people have joined a union over the past ten years in most European countries. The average age of union members is increasing and is higher than the average age of wage- and salary-earners in general. Simultaneously, while young people generally hold positive views about trade unions, they lack knowledge about them. With the loss in membership going hand in hand with a decrease in revenues, it remains to be seen to what extent and how trade unions will address the membership and generational challenges. Much will depend on trade union identities and the framing of the challenges, unions’ conception of membership, and their reliance on membership dues for their organisational sustainability. Whatever the case, the magnitude of the generational and member challenges is also leading other players to help bolster union security, in combination with unions’ own (recruitment and organising) initiatives.
Full-text available
Conditions in the sharing economy are often favourably designed for consumers and platforms but entail new challenges for the labour side, such as substandard social-security and rigid forms of algorithmic management. Since comparatively little is known about how providers in the sharing economy make their voices heard collectively, we investigate their opinions and behaviours regarding collective action and perceived solidarities. Using cluster analysis on representative data from across twelve European countries, we determine five distinct types of labour-activists, ranging from those opposed to any forms of collective action to those enthusiastic to organise and correct perceived wrongs. We conclude by conjecturing that the still-ongoing influx of new providers, the difficulty of organising in purely virtual settings, combined with the narrative of voluntariness of participation and hedonic gratifications might be responsible for the inaction of large parts of the provider base in collectivist activities.
Full-text available
In the framework of the so-called “sharing economy”, the number of on-demand companies matching labour supply and demand is on the rise. These schemes may enlarge opportunities for people willing to find a job or to top up their salaries. Despite the upsides of creating new peer marketplaces, these platforms may also be used to circumvent employment regulation, by operating informally in traditionally regulated markets. Literature showed how, by 2009, over 2 million worker accounts had been generated within these frameworks. Productivity may be fostered but, at the same time, a new version of Taylorism is disseminated (i.e. the fragmentation of labour into hyper-temporary jobs – they call them microtasks – on a virtual assembly line), strengthened by globalisation and computerisation. All these intermediaries recruit freelance or casual workers (these continue to be independent contractors even though many indicators seem to reveal a disguised employment relationship). Uncertainty and insecurity are the price for extreme flexibility. A noteworthy volume of business risk is shifted to workers, and potential costs as benefits or unemployment insurance are avoided. Minimum wages are often far from being reached. This paper will present a case study analysis of several “on-demand work” platforms, starting from the Amazon Mechanical Turk, one of the first schemes founded in 2005, which is arguably “employing humans-as-a-service”. It splits a single service in several micro “Human Intelligence Tasks” (such as tagging photographs, writing short descriptions, transcribing podcasts, processing raw data); “Turkers/Providers” (workers) are selected by “Requesters” to rapidly accomplish assignments online, are then rated according to an internal system and are finally paid (also in gaming credits) only if delivery is accepted. After having signed up and worked within some platforms, I comment upon TaskRabbit (thousands people on the service who bid to do simple manual tasks), Handy and Wonolo (personal assistance at a local level), oDesk and Freelancer (online staffing), Uber and Lyft (peer-to-peer ridesharing), Airbnb (hosting service), InnoCentive (engineering solutions), Axiom (legal research or service), BitWine (consultancy). Finally I highlight downsides and upsides of work in these platforms by studying terms of service or participation agreements to which both parties have to agree. I look into several key features such as (i) means of exchange/commodities, (ii) systems of payment, (iii) demographics, (iv) legal issues concerning status and statutory protection of workers, indicators of subordination, treatment of sickness, benefits and overtime, potential dispute resolution, and deprived “moral valence of work” and I discuss potential strategies to address these issues.
Do lower membership fees encourage young people to join a trade union? Evidence from Belgium In this article we investigate young people's motives to join a trade union and the extent to which a lower membership fee can encourage membership based on a mixed methods design combining data from face-to-face expert interviews and a web survey among 218 Belgian young union members who benefit of a lower fee. The results demonstrate that young employees predominantly join the union because of both instrumental (linked to the services trade unions provide) and social motives (relating to union members in their social environment). Only a minority of members joined the union because of the lower membership fee. Our results also indicate that motives to join the union differ between various social groups. Specifically, vulnerable groups (e.g. young people on temporary contracts or with low wages) may join due to getting easier access to unemployed benefits or because they can follow training. This emphasizes the diverse needs of young employees which trade unions should take into account.
Migrants form growing proportions of national workforces in advanced capitalist societies. Yet little is known about their attitudes towards the principal agents of worker representation in their host countries, the trade unions, much less via cross‐national research. Using European Values Survey data, we redress this imbalance by examining migrants’ levels of trust in unions, compared to native‐born. We find higher levels of trust in unions by migrants (compared to native‐born) in general and especially by migrants during their first decades after arrival and whose countries of origin are characterized by poor quality institutions. These findings have significant implications for unionization strategies towards migrants, especially given received wisdom portraying migrants as indifferent or distrustful towards unions.
This article investigates the (dis)embeddedness of digital labour within the remote gig economy. We use interview and survey data to highlight how platform workers in Southeast Asia and Sub-Saharan Africa are normatively disembedded from social protections through a process of commodification. Normative disembeddedness leaves workers exposed to the vagaries of the external labour market due to an absence of labour regulations and rights. It also endangers social reproduction by limiting access to healthcare and requiring workers to engage in significant unpaid ‘work-for-labour’. However, we show that these workers are also simultaneously embedded within interpersonal networks of trust, which enable the work to be completed despite the low-trust nature of the gig economy. In bringing together the concepts of normative and network embeddedness, we reconnect the two sides of Polanyi’s thinking and demonstrate the value of an integrated understanding of Polanyi’s approach to embeddedness for understanding contemporary economic transformations.
This qualitative industry case study evaluates job quality in the Australian platform-based food-delivery sector, one part of the growing gig economy where workers, as independent contractors, engage in digitally-enabled and controlled work that is remunerated on a piece rate basis. Using a multi-dimensional framework, we draw on worker accounts of economic security, autonomy and enjoyment to assess job quality. This study posits that to achieve a more refined picture of job quality, both objective and workers' subjective understandings of work need to be understood in the context of their respective 'fit' in terms of individual circumstances, labour market alternatives and the broader socio-political context. This multi-level analysis problematises individual accounts that risk overemphasising the positive elements of platform-based work. Moreover, rather than sitting neatly in a Post-or Neo-Fordist extension of job quality, the findings reveal that the gig economy is a new juncture in capitalist production, the consequences of which need to be taken seriously by regulators, scholars, workers and other relevant stakeholders.
In this article, the authors consider the findings of a multi-year, case study-based research project on young workers and the labor movement in four countries: France, Germany, the United Kingdom, and the United States. The authors examine the conditions under which young workers actively engage in contemporary labor movements. Although the industrial relations context matters, the authors find the most persuasive explanations to be agency-based. Especially important are the relative openness and active encouragement of unions to the leadership development of young workers, and the persistence and creativity of groups of young workers in promoting their own engagement. Embodying labor’s potential for movement building and resistance to authoritarianism and right-wing populism, young workers offer hope for the future if unions can bring them aboard.