Content uploaded by Martin Krzywdzinski
Author content
All content in this area was uploaded by Martin Krzywdzinski on Jun 18, 2020
Content may be subject to copyright.
Varieties of platform work.
Platforms and social inequality in
Germany and the United States
Martin Krzywdzinski & Christine Gerber
May 2020
Weizenbaum Series #7
Working Paper
Weizenbaum Series
Edited by
Weizenbaum Institute for the Networked Society –
The German Internet Institute
Project Coordination:
Wissenschaftszentrum Berlin für Sozialforschung
Reichpietschufer 50
10785 Berlin
Visiting Address:
Hardenbergstraße 32
10623 Berlin
Email: info@weizenbaum-institut.de
Web: www.weizenbaum-institut.de
Persistent long-term archiving of this series is ensured by the So-
cial Science Open Access Repository and the DOI registration ser-
vice in Germany for social science and economic data da|ra.
DOI 10.34669/wi.ws/7
This series is available open access and is licensed under Creative
Commons Attribution 4.0 (CC-BY 4.0):
http://creativecommons.org/licenses/by/4.0/
Varieties of platform work.
Platforms and social inequality in
Germany and the United States1
Martin Krzywdzinski2 & Christine Gerber3
May 2020
1 This is the authors‘ version of an article submitted in German to Nina Baur & Jürgen Beyer
(2020), Wirtschaft und soziale Ungleichheit, Berlin: Springer.
2 Martin Krzywdzinski is director at the Weizenbaum Institute for the Networked Society, head of
the research group „Globalization, Work and Production“ at the WZB Berlin Social Science Center
and professor of international labor relations at the Helmut Schmidt University Hamburg.
3 Christine Gerber ist research fellow at the WZB Berlin Social Science Center.
Weizenbaum Series #7
Working Paper
4
Abstract
The platform economy has been criticized for exacerbating social inequalities in va-
rious ways. This study draws on these discussions and examines the extent to which
social inequalities are being reproduced, reduced, or even increased within platform
work. The rst central question is that of the precariousness of this form of work and
the vulnerability of the platform workers as a group. This is followed by a second
question about the role of classical dimensions of inequality of education and gender
within the group of platform workers. The study focuses on inequalities related to
income, workload, and the subjective perception of platform work. It follows a com-
parative approach, building on institutionalist analyses developed in labor market and
inequality research. The empirical analysis is based on case studies of 15 crowdwork
platforms in the United States and Germany and on an online survey of crowdworkers
in both countries. While platforms represent a global organizational model, they are
embedded in different models of capitalism. The study shows that existing labor mar-
ket segmentation and social welfare systems determine who works on platforms and
to what extent. The weaker the social safety net, the more likely platform work is to
be both a curse and a blessing: It offers a much needed and exible source of income,
albeit under extremely precarious conditions. The stronger the social safety net, on the
other hand, the greater the market power of workers vis-à-vis the platforms.
Varieties of platform work 5
Introduction
The development of the platform economy has engendered both high expectations
and erce criticism. The great expectations came, for the most part, in the early ye-
ars, when this economic form was still typically called the “sharing economy,” a title
that emphasized how a peer-to-peer economy could generate income and change the
world of work (cf. Vallas/Schor 2020). As the platform economy has come of age,
however, criticism has grown. A number of studies have accused it of exacerbating
social inequalities in various ways. For example, some have criticized it for creating
a precarious employment sector characterized by low wages and a lack of social pro-
tection (MacDonald/Giazitzoglu 2019; De Stefano 2016) and for lacking mechanisms
to represent workers’ interests (Hoose et al. 2019). Scholars have also criticized the
frequent discrimination based on gender, age, or ethnicity (Cherry 2019a; Van Doorn
2017; Piasna/Drahokoupil 2017). The platforms’ far-reaching control mechanisms pla-
ce their workers, who are supposed, at least theoretically, to operate independently of
the company, in a position of high dependence (Cutolo/Kenney 2019; Gerber/Krzyw-
dzinski 2019; Wood et al. 2018).
This study draws on these discussions and examines the extent to which social inequa-
lities are being reproduced, reduced, or even increased within platform work. The rst
central question is that of the precariousness of this form of work and the vulnerability
of the platform workers as a group. This is followed by a second question about the
role of classical dimensions of inequality of education and gender within the group of
platform workers. We are interested in inequalities related to income, workload, and
the subjective perception of platform work.
The major contribution of the present study lies in its comparative approach, building
on institutionalist analyses developed in labor market and inequality research (Benassi
et al. 2016; Emmeneger et al. 2012). The empirical analysis is based on case studies of
15 crowdwork platforms in the United States and Germany and on an online survey of
crowdworkers in both countries. Our starting point is the expectation that country-spe-
cic institutional arrangements (especially with regard to labor market regulation and
social security) shape inequality structures within platform work. This perspective
has hardly been pursued in the eld of platform work to date, not least because the
platform economy explicitly presents itself as a global model that is not limited by
institutional constraints. Our analysis picks up on and continues initial work on the
institutional conditions under which the platform economy emerged (Rahman/Thelen
2019; Friedman 2014).
6
Platform work is a broad term that has not been used entirely consistently in the lite-
rature. Hence, before continuing, we will give a brief overview of the concept. What
all forms of platform work share is their use of an “on-demand” market of formally in-
dependent workers organized via internet-based platforms. There are, however, major
differences between the platforms regarding the importance of space. Platforms such
as Uber, Deliveroo, MyHammer, or Helpling act as intermediaries for service activi-
ties that are delivered locally and are therefore location-dependent. Platforms such as
Amazon Mechanical Turk (AMT) or Upwork, on the other hand, focus on location-in-
dependent jobs that are done online and by a largely anonymous mass of people, the
so-called crowd. This latter model is referred to as crowdwork. There are certain diffe-
rences between location-dependent and location-independent forms of platform work.
For example, collective worker organization is much more possible in location-depen-
dent forms of platform work than in crowdwork. At the same time, however, there are
also great similarities between the different forms of platform work. Thus, although
our own empirical analysis focuses on crowdwork, we discuss the state of the research
and our conclusions in a broader context and use the more general term platform work.
State of the research
Platform work and social inequality
From a formal point of view, platform work is a special kind of digitally organized solo
self-employment.4 What unites the various models of platform work is that the wor-
kers are registered as self-employed, they usually receive a piece wage, and they can
be called up “on-demand.” Crowdworking has further special features due to its so-
cio-technical conguration; unlike Uber drivers, for example, crowdworkers can work
at any time and from any place. This limits their opportunities to talk to one another,
collectively organize, and engage in interest representation (Gerber 2020a; Howcroft/
Bergvall-Kåreborn 2018; Lehdonvirta 2016).
With regard to the connection between platform work and social inequality, three
aspects can be distinguished. The rst question is which social groups take up plat-
form work and how they differ from other groups. Second, it is important to consider
the conditions of platform work: Is it a particularly precarious form of employment or
does it also offer opportunities – in other words, is platform work an “elevator to the
top,” a guarantee of social status, or an “elevator to the bottom”? Third, it is possible
4 Jonas Ferdinand (WZB) supported us in the writing of this section.
Varieties of platform work 7
to investigate differences between different groups of platform workers in terms of
income, working hours, and other variables.
Regarding the rst question, namely of the socio-demographic characteristics of plat-
form workers, the statistics collected to date are very imprecise. Depending on the met-
hods and denitions used, the estimated size of this employment segment in Germany
and the United States varies between less than 1% and about 4% of the workforce
(Bonin/Zinne 2017; Sering 2018; Jackson et al. 2017). Few systematic country-com-
parative studies on the socio-demographic composition of platform workers have been
conducted to date. With regard to crowdwork, the available literature has focused on
individual countries or platforms; for example, it has looked at crowdworkers on AMT
in the United States and India (Diffalah et al. 2018; Ipeirotis 2010) or professional free-
lancers on Upwork (Popiel 2017; Gandini 2016). In addition, there are some studies
of crowdwork in Germany (Schneider-Dörr 2019; Sering 2018; Bonin et al. 2017;
Pongratz/Bormann 2017), while the US studies mostly consider platform work in a
broader sense (Bureau of Labor Statistics 2018; Smith 2016; Farrell/Greig 2016). For
Germany, studies indicate that most crowdworkers tend to be male and have higher
educational attainment than the population average. Compared to other solo self-em-
ployed people, crowdworkers are also signicantly younger. Initial estimates of the
global crowdwork workforce and its composition are available in Berg et al (2018).
For the second question concerning working conditions on the platforms in compari-
son with other forms of employment, existing studies have reached different conclu-
sions. Some studies point to the opportunities for weaker groups in the labor market,
e.g., ex-convicts, physically or mentally disabled people, or even single parents (Car-
chio 2019; Zyskowski et al. 2015). Crowdwork also offers the opportunity to reduce
regional and even global employment and income inequalities by enabling people to
take advantage of opportunities in labor markets beyond their region (Berg et al. 2018;
Beerepoot/Lambregts 2015). In this way, people in structurally weak areas or in the
Global South can benet from a demand for labor generated in metropolitan areas or
in the Global North (Graham et al. 2017; Thies et al 2011). Yet, this literature on the
benets of platform work is contrasted by a rapidly growing body of research that
addresses its negative effects, particularly on social inequality. Scholars have argued
that platform work is creating a particularly precarious form of solo self-employment
(De Stefano 2016), i.e., that a new large category of marginalized work is being es-
tablished, thus increasing social inequality. The mechanisms of this precarization are
exibilization and disempowerment.
8
The exibilization argument refers to the form of remuneration on platforms and the
lack of social welfare benets (Berg et al. 2018). Platform work is usually remune-
rated per task performed, which can amount to a few cents for tagging a picture or a
few euros for a short product description; on so-called competition platforms, such as
99designs, platform workers take part in competitions and go away empty-handed if
their work is not among the selected winners (De Stefano 2016). Since platform wor-
kers are formally registered as self-employed, they are also not entitled to social wel-
fare benets, health insurance, paid holidays, or protection against dismissal (Huws et
al. 2016). For high-wage countries, this means that part of the labor market is being
disembedded and de facto integrated into a global low-wage labor market: “the sun
never sets on Amazon’s technology platform” (Irani 2015: 726).
Flexibility is also promoted by ranking systems, which are a key means of labor regu-
lation on platforms (Kellog et al. 2020; Gandini 2016; Lee et al., 2015). A core com-
ponent is the constant assessment of work quality, either by the customer or (in online
work) by other crowdworkers. Depending on the platform and type of work, further
criteria may be added, and these criteria may or may not be transparent. On many plat-
forms, the ranking systems take into account factors such as activity on the platform,
speed of answers to customer queries, or engagement in community discussions (Ger-
ber/Krzywdzinski 2019). This leads to crowdworkers being forced to be active on the
platforms on weekends or in the evening (Berg et al. 2018; Wood et al. 2018).
The disempowerment argument points to the special forms of control on platforms;
these differ from both regular employment relationships and classic self-employment
and contribute to the vulnerability of platform workers in comparison to other social
groups. Many studies on both location-dependent and location-independent platform
work emphasize the information and power asymmetries between platforms, clients
and workers (Gerber 2020b; Ivanova et al. 2018). Cutolo and Kenney (2019: 13–18)
describe the mechanisms that cause the disempowerment of platform workers as fol-
lows:
1) Platforms control the terms of trade and can change them unilaterally.
2) Platforms control workers’ access to information on available jobs, prices etc.,
while they themselves have complete information on the entire market.
3) Platforms control the interface between customers and workers, i.e., they can (de-
pending on the platform and type of work) withhold all information about the
customers from the workers.
4) Platforms can bar workers from accessing the platform, and thus customers, at
will.
Varieties of platform work 9
5) Platforms dene the ranking and reputation systems, which have an enormous
inuence on platform workers’ access to jobs. The criteria of the ranking and re-
putation systems are often kept secret to prevent opportunistic behavior and ma-
nipulation.
With regard to the third question, namely that of inequalities between different groups
of platform workers, the research argues that platforms reproduce and sometimes in-
tensify existing discrimination and thus social inequality on the labor market. Many of
the markets in which platforms act as intermediaries have historically been characteri-
zed by gender inequalities (Van Doorn 2017). Platforms that allow crowdworkers to be
selected based on their proles may reproduce stereotype-based selection mechanisms
(Cherry 2019a; Foong et al. 2019). Adams and Berg (2017) compared the wages of
crowdworkers in the United States by gender and showed that women earn about $1
less per hour than men. Foong et al (2018) showed that such wage differences also
exist for crowdworkers on Upwork. Leung and Koppman (2015) showed that women
have a considerable disadvantage when working on programming platforms due to
gender stereotypes among the clients. Huws et al. (2016), however, argued in their
comparative study of Germany, Sweden, Austria, and Great Britain that platform work
sometimes enables women to take on typically male-connoted work – so platforms
could also help reduce gender differences in the labor market.
Platform work and country-specic institutional
systems
In the research, the image that pervades is that of platform work as a global model that
dees national regulation and is thus completely unembedded (Cherry 2019b; Berg
et al. 2018). This means that the research on platform work contrasts with existing
research on social inequality, which emphasizes the role of institutional arrangements
(Häusermann et al. 2019; Benassi et al. 2016).
In particular, research on the dualization of labor markets and societies has emphasi-
zed the role of country-specic institutional arrangements (Schwander 2019; Emme-
negger et al. 2012). The key argument of this research is that, in contemporary socie-
ties, we are observing a tendency to segregate regular and secure employment from
a growing margin of precarious employment in response to globalization processes
and competition for location. However, this tendency varies from country to country
and depends on the institutional arrangement of labor market regulation and welfare
state protection. Labor market regulation determines how far dualization processes can
go through rules for the use of temporary work, xed-term contracts, and protection
10
against dismissal. The welfare state in turn inuences the vulnerability of precarious
groups and thus their position on the labor market (Gerstung 2019).
The institutional differences in labor market regulation and welfare state security in
Germany and the United States – which are the countries this study focuses on – have
been well researched (Kaufmann 2003). There has certainly been a rapprochement
between the two welfare state regimes (Seeleib-Kaiser 2014). In both countries, claims
under the pension and unemployment insurance systems have declined in recent de-
cades. At the same time, precarious forms of employment have increased in Germany
due to the deregulation of the labor market (e.g., with regard to the possibilities of
using temporary work), the decline in collective bargaining coverage, and the out-
sourcing of workers by companies. As a result, the low-wage sector has expanded and
the number of self-employed workers has also increased (Grabka/Schröder 2019; ILO
2017). Nevertheless, there are a number of regime differences between Germany and
the US that are relevant to our analysis:
• Protection in the event of unemployment (especially for the long-term unemploy-
ed) and illness – i.e., the decommodication of labor – is still signicantly higher
in Germany than in the United States, where there is no universal minimum welfa-
re benet system and a large proportion of employees do not have health insurance
at all (Seeleib-Kaiser 2014). The German system offers parental leave benets that
do not exist in the United States. Moreover, the costs of childcare are signicantly
lower in Germany (Kröger 2011; Henderson/White 2004). We therefore expect
platform work to be an important source of income for the groups of precarious
solo self-employed persons in the United States as well as for parents and single
parents, whereas in Germany, these groups can rely on social welfare benets and
may not need to do platform work. The large gaps in welfare-benet coverage of
the population in the United States will also likely lead to greater geographical in-
equalities. Especially in poorer rural regions, platform work could be an important
source of income.
Varieties of platform work 11
• The institutional features described here have shaped a culture in the United States
that emphasizes entrepreneurship and rejects collective responsibility for citizens’
social wellbeing (Richter et al 2018; Seeleib-Kaiser 2014). In addition, the distinc-
tion between dependent employment and solo self-employment is less clear-cut in
the United States, as the legal protection against dismissal is very weak and most
dependent employees are not entitled to benets like for instance the continued
payment of wages during sickness leave. On this basis, we can expect platform
work to be more accepted in the United States than in Germany. On the one hand,
such precarious independence is considered “normal” due to the lack of alternati-
ves. On the other hand, it is consistent with a model of individual assumption of
responsibility.
• The United States has been characterized by a particularly pronounced long-term
trend of weakening trade unions (encouraged by state policy) and by rather an-
tagonistic relations between companies and trade unions (Cutcher-Gershenfeld/
Kochan 2004). By contrast, cooperation between trade unions, companies, and the
state is much more pronounced in Germany, despite all existing conicts. We can
expect these differences to be reected in the nature of the debate on the regulation
of platform work. While we could expect more open conicts in the United States,
we would expect attempts at a negotiated approach between the state, companies,
and trade unions in Germany.
Comparative analyses of the regulation of the platform economy and the role of in-
stitutional frameworks are still a relatively new phenomenon. Recently, some initial
analyses of how institutional factors inuence platform work have been presented.
Some authors have pointed out that platform work has emerged from a specically
US regulatory context (Van Doorn 2017). Drawing on Friedman (2014), we can also
understand platform work as a digital intensication of the atypical work that has been
spreading since the 1970s. Rahman and Thelen (2019) have also described platforms
as a further development of the outsourcing strategies that sought to enhance share-
holder value. Using the example of Uber, Thelen (2018) analyzed the special role US
institutions and regulation played in the development of the platform. Although Uber
met with resistance from taxi companies in the United States, it was able to win the
conict by successfully mobilizing users (cf. Thelen 2018: 945). However, this led to
regulatory responses at the municipal or even state level, for example, in New York
or California (De Stefano 2016; Isaac et al. 2015). In the case of Germany, the few
currently existing studies on the subject have discussed the role of trade unions such as
ver.di and IG Metall in regulating platform work (Haipeter/Hoose 2019). These unions
12
have started organizing activities for platform workers, initiated dialogue processes
with German platforms, and brought the issue to the political stage.
The present study builds on existing analyses but undertakes an explicit comparative
analysis of crowdwork and its regulation in the United States and Germany.
Data
In the analysis, we used data from the project “Between digital bohemia and precarity.
Work and performance in the crowd” (2016–2019), which was funded by the Thyssen
Foundation. The analysis focused on crowdwork as a specic form of platform work.
The project combined qualitative and quantitative methods. In the qualitative part of the
data collection and analysis, 15 case studies of crowdwork platforms in Germany and the
United States were conducted. The central question concerned the extent to which the
different institutional regimes in Germany and the United States inuenced the work and
performance regulation on the platforms. In particular, the study examined approaches
to the recruitment of crowdworkers, the control and regulation of performance (incen-
tive and remuneration systems), and communication with crowdworkers. The case stu-
dies were based on an analysis of platform websites and ofcial documents, their work
interfaces for crowdworkers, and their communication forums. The study also involved
conducting 32 one- to two-hour semi-structured interviews with representatives of the
platforms (Gerber/Krzywdzinski 2019; Gerber 2020a).
The platform-selection process took place in several steps. Our central aim was to inclu-
de platforms that focused on different types of work (in terms of task content and skill
requirements). We distinguished between two task types: micro- and macrotasks. Micro-
tasks can be dened as routine tasks (e.g., image categorization, lead data verication,
short audio transcriptions) or as tasks that do not require specic knowledge (e.g., short
product descriptions, testing of apps). The nature of these tasks allows them to be broken
down into short and standardized components that can be completed within seconds
or minutes. Several crowdworkers can work simultaneously on the same task without
having to interact. Macrotasks, by contrast, are complex and require a high degree of
creativity and specic, sometimes professional knowledge (e.g., design activities, soft-
ware programming, development of product concepts, scientic problem solving). The-
se tasks cannot be broken down and are therefore organized as projects lasting several
days or weeks. The focus is on quality. Platforms often organize competitions to generate
proposals, from which the client, a jury, or the crowd community can select the winners.
To select the platforms, we compiled a list of all platforms in Germany and selected
metropolitan regions in the United States (Bay Area, Boston, Chicago, New York) on
Varieties of platform work 13
the basis of internet research. We identied 60 platforms, which we contacted in several
rounds by phone and email. We continued to contact them until we reached a sample of
15 case studies that equally represented micro- and macroplatforms.
The qualitative analysis showed that the central differences in the design of work and
performance regulation related to differences in task types between platforms (micro-
versus macrotasks); there were hardly any differences between German and US plat-
forms. Apparently, crowdwork is indeed a global model, in that the platforms apply si-
milar organizational principles (Gerber/Krzywdzinski 2019).
In the following, we will analyze the quantitative survey of crowdworkers. The survey
was sent to crowdworkers who registered as German or US residents on the platforms.
The total sample consists of 1,131 participants; however, those participants who stated
in the survey that their main residence was not in Germany or the United States were
excluded from the analysis; this may be the case if the crowdworkers work as “global
nomads.” Accordingly, the sample used here comprises 1,088 persons.
On four platforms, the survey was placed on the platform as a paid task. On the other
platforms, the survey invitation was sent to the crowdworkers via the platform itself or
(if the platforms refused to participate in the survey) via social media (Facebook, Linke-
dIn, Reddit, Xing) and the platforms’ communication forums.
In our case studies, we received information from the platforms about the composition
of the crowdworkers by gender and age. We used this information to calculate plat-
form-specic weighting variables. In terms of gender, our sample composition deviated
only slightly from the actual composition of the crowdworkers on the platforms survey-
ed, but crowdworkers under 35 years of age were noticeably underrepresented. We used
a weighting variable to correct for this.
Our data therefore represent a “convenience sample,” which means we cannot general-
ize beyond the platforms we surveyed. However, the platforms we surveyed represent
a total of 14 million registered crowdworkers, which is a considerable proportion of
the total number of workers in this sector. Our sample is also broadly similar in com-
position to those of the available studies on Germany (Bonin/Zinne 2017; Leimeister
et al. 2016) as well as on the United States (Difallah et al. 2018; Popiel 2017).
An advantage of our sample is that we covered platforms with different types of tasks
in one survey, thus avoiding a focus on individual platforms, which is the dominant
approach in the current research. Second, we surveyed German and US crowdworkers
on the same platforms, which is an important basis for this comparison. The following
table, Table 1, provides information on our sample.
14
Table 1. Socio-demographic composition of the sample
Germany United States
Gender
Women 44.8% 53.8%
Men 55.0% 45.2%
Other/third gender 0.2% 1.0%
Age
Up to 21 years 10.1% 8.5%
22–29 37.0% 31.4%
30–39 30.0% 30.2%
40–49 12.6% 16.1%
50–64 10.3% 11.5%
Above 65 0.9% 2.4%
Education High-school diploma
and below
44.4% 31.1%
Tertiary degree 55.6% 68.9%
Platform type Microtasks 88.0% 62.2%
Macrotasks 12.0% 37.8%
N 576 512
Source: Krzywdzinski/Gerber 2019. Weighted data.
In the US subsample, the majority of workers were women; in the German one, the
majority were men. This corresponds with the ndings of existing studies. There is
thus a certain imbalance in our sample, which results from the fact that macrotask
platforms in Germany had lower response rates than those in the United. This also
means that workers with a tertiary education are more strongly represented in the US
subsample than in the German one. This makes it necessary to consider the platform
type (micro/macro) when doing country comparisons.
The questionnaire used in the survey consists of 29 questions and covered the follo-
wing topics: (a) basic information about the work situation (working hours, type of
tasks etc.); (b) the form of performance monitoring on the platform; (c) reputation and
ranking systems; (d) interaction with other crowdworkers; (e) perception of working
conditions and stress; and (f) socio-demographic information (age, gender etc.).
Varieties of platform work 15
The working conditions and income situations of the
crowdworkers
With regard to the time spent by crowdworkers on platform work, there were clear
differences between Germany and the United States (Table 2). In Germany, crowdwor-
kers spent considerably less time on platforms: two thirds of the respondents worked
on their platforms for a maximum of 10 hours per week. This suggests that crowdwork
was a secondary activity for this group. In the United States, on the other hand, 35%
of those surveyed worked more than 20 hours a week on the platforms – here, it can be
assumed that platform work was the main source of income. Only 40% of respondents
performed platform work as a small part-time job of 10 hours or less per week.
Table 2. Typical weekly working time on platforms of crowdworkers in Germany
and the United States by gender and educational attainment
1–10
hours
11–20
hours
21–30
hours
31–40
hours
More than
40 hours
Germany
Men 66.1% 20.8% 10.1% 3.0% 0%
Women 63.2% 20.6% 10.8% 3.8% 1.7%
High-school or less 63.5% 23.4% 8.4% 3.3% 1.4%
Tertiary education 65.7% 18.0% 11.7% 3.4% 1.2%
United States
Men 40.2% 24.2% 19.0% 8.9% 7.6%
Women 40.8% 23.5% 18.8% 11.5% 5.4%
High-school or less 38.5% 18.6% 22.4% 12.6% 7.9%
Tertiary education 41.5% 26.0% 17.5% 9.1% 5.8%
Source: Krzywdzinski/Gerber 2019. Weighted data.
These differences may have been related to the different compositions of the German
and US subsamples with regard to platform type: Microtask platform workers are cle-
arly more represented in the German than in the US subsample. We thus took the plat-
form type into account. However, even when we did so, the picture hardly changed.
If only microtask platforms were considered, the share of German crowdworkers who
worked a maximum of 10 hours per week was 64.1%; in the US case, it was 45.6%. In
Germany, 13.5% of crowdworkers on microtask platforms worked 20 hours or more
compared to 37.0% in the United States.
16
With regard to working time, there were no signicant gender differences on the plat-
forms in either country. In Germany, there were also no statistically signicant differen-
ces between different education levels. In the United States, on the other hand, people
without tertiary education had higher weekly working hours on platforms; this gure
was statistically signicant (p<0.05). This suggests that crowdwork is more often a main
source of income for the non-tertiary-educated group in the United States.
Table 3. Primary employment status of crowdworkers in Germany
Men Women High-school
education
Tertiary
education
Microtask platforms
Employed (Full time) 39.3% 21.7% 30.8% 32.4%
Employed (Part time) 12.1% 11.6% 8.7% 14.6%
Self-employed 17.9% 24.3% 17.7% 24.0%
University/Education 22.2% 20.1% 29.9% 13.4%
Other (retired, registered
unemployed)
8.5% 22.3% 12.9% 15.6%
Macrotask platforms
Employed (Full time) 23.1% 6.0% 8.3% 16.7%
Employed (Part time) 10.3% 33.3% 8.3% 23.3%
Self-employed 59.0% 42.4% 58.3% 50.0%
University/Education 2.6% 3.0% 8.3% 1.7%
Other (retired, registered
unemployed)
5.1% 15.1% 16.7% 8.3%
Source: Krzywdzinski/Gerber 2019. Weighted data.
The similarities and differences between Germany and the United States are evident
when we look at employment status (Tables 3 and 4). Here, we separately considered
crowdworkers on micro- and macrotask platforms. The majority of crowdworkers on
microtasks platforms in both countries also had a regular job or attended university; in
both countries, the majority of crowdworkers on macrotask platforms were self-employ-
ed. In both Germany and the United States, the employment status also reected the clas-
sic gender inequalities on the labor market. Men who did crowdwork were much more
likely than women to have a regular full-time job, while women working on platforms
were much more likely than men to have a regular part-time job or no other job at all.
Varieties of platform work 17
Table 4. Primary employment status of crowdworkers in the United States
Men Women High-school
education
Tertiary
education
Microtask platforms
Employed (Full time) 44.9% 27.3% 18.7% 43.4%
Employed (Part time) 16.1% 21.5% 16.3% 21.4%
Self-employed 21.4% 23.2% 28.8% 18.7%
University/Education 7.7% 5.9% 9.5% 5.4%
Other (retired, regis-
tered unemployed)
9.8% 22.1% 26.8% 11.1%
Macrotask platforms
Employed (Full time) 29.3% 19.2% 15.4% 26.8%
Employed (Part time) 13.0% 8.1% 7.7% 11.1%
Self-employed 43.5% 49.5% 48.7% 45.7%
University/Education 2.2% 5.0% 7.7% 2.6%
Other (retired, regis-
tered unemployed)
12.0% 18.2% 20.5% 13.7%
Source: Krzywdzinski/Gerber 2019. Weighted data.
There were differences between Germany and the United States, especially with regard to
the role of education. First, a much larger proportion of crowdworkers in Germany than in
the United States used this activity as a secondary source of income during their university
education. Second, there were differences between education levels. In Germany, the dif-
ferences in employment status between crowdworkers with and without tertiary education
appear relatively small – at least if we focus on crowdworkers on microtask platforms and
ignore crowdworkers on macrotask platforms because of the small number of cases. In the
United States, on the other hand, there were large differences between people with diffe-
rent education levels. University-educated crowdworkers were much more likely to have
a regular full-time or part-time job while those without a university education were much
more likely to be self-employed or without any other employment.
Table 5 conrms the ndings presented so far. In Germany, only 16% of the surveyed
crowdworkers reported that their platform income accounted for more than 50% of
their total income. In the United States, the gure was just under 33%. Here, too, this
difference between Germany and the United States applied to crowdworkers on both
microtask and macrotask platforms.
18
Table 5. Share of platform income in total income of crowdworkers in Germany and
the United States
Less than 25% 25-50% 51-75% More than 75%
Germany 71.4% 12.5% 5.0% 11.1%
United States 47.2% 20.2% 9.0% 23.6%
Source: Krzywdzinski/Gerber 2019. Weighted data.
How can we interpret these ndings? They conrm our initial expectation that the
different levels of decommodication of labor in Germany and the United States are
related to the role of crowdwork as a source of income. In the United States, which has
low social welfare benets, crowdwork is an important opportunity to generate income
if no other options are available on the local labor market. Despite its precariousness,
crowdwork is often the primary source of income in the United States because no alter-
natives are available. In Germany, on the other hand, crowdwork is much more often a
secondary activity. This could be explained by the fact that there are fewer incentives
to use crowdwork as a primary source of income in the German context, where indivi-
duals enjoy relatively high social welfare benet levels. Another explanation could be
the strong growth of part-time employment and of so-called “mini jobs” in Germany,
which, given their low wages, create the need for additional income.
In both countries, women and crowdworkers without tertiary education were more de-
pendent on income from platform work than men and college graduates. On the one
hand, this reproduces the classical inequalities. On the other hand, crowdwork can also
be interpreted as an opportunity: It represents a new kind of access to income for these
groups, which makes them a little more independent of the local labor market.
Subjective perceptions of platform work
To what extent are the differences in crowdworkers’ objective work situations also
reected in their subjective perceptions of platform work? An interesting nding in Ta-
bles 6 and 7 is that crowdworkers in the United States are signicantly more satised
with their platform work than those in Germany and they also more frequently see this
type of work as a long-term prospect. This difference applies to both crowdworkers on
micro- and macrotask platforms, i.e., regardless of the type of task.
Varieties of platform work 19
Table 6. “Overall I am satised with working on the platform.”
Strongly
agree
Agree Partly Disagree Strongly
disagree
Germany 19.7% 46.3% 24.7% 8.5% 0.7%
United States 31.0% 46.9% 17.5% 3.2% 1.3%
Source: Krzywdzinski/Gerber 2019. Weighted data.
Table 7. “I see my personal future in platform work.”
Strongly
agree
Agree Partly Disagree Strongly
disagree
Germany 6.5% 17.1% 20.9% 23.6% 31.8%
United States 15.7% 26.4% 28.3% 18.9% 10.7%
Source: Krzywdzinski/Gerber 2019. Weighted data.
Research on job satisfaction has shown that measurements of subjective satisfaction
do not provide information on objective working conditions (cf. Brown et al. 2012) –
people are often satised with forms of work and employment that are poor and stress-
ful according to objective criteria. It is therefore important to reect on the reasons for
satisfaction. There are two possible explanations for the ndings presented here. On
the one hand, the differences in satisfaction between German and US crowdworkers
may represent the commodication of labor. In the US system, which provides limited
social welfare benets, platform work reduces dependence on the local labor market;
thus despite all the burdens, it can be perceived as a gain in autonomy and can lead to
higher satisfaction. On the other hand, our ndings may also reect general cultural
differences between Germany and the United States. In international surveys, Ame-
ricans tend to be slightly more satised with their work than Germans (Souza-Posa/
Souza-Posa 2000). In addition, self-employment enjoys greater recognition in the Uni-
ted States than in Germany and the uncertainties of self-employment are perceived as
more socially normal than in Germany.
Interestingly, however, the following two tables, Table 8 and 9, show that US workers’
greater satisfaction with crowdwork is accompanied by a higher level of perceived inse-
curity and a higher degree of work stress than is evident in the German subsample. Given
the greater importance of platform incomes for US crowdworkers, this is not surprising.
20
Table 8. “The uncertainty of whether I will get enough jobs in a month and earn
enough money worries me a lot.”
Strongly
agree
Agree Partly Disagree Strongly
disagree
Germany 7.7% 10.8% 24,5% 23.6% 33.2%
United States 19.0% 19.0% 24.0% 23.4% 14.6%
Source: Krzywdzinski/Gerber 2019. Weighted data.
Table 9. “I organize my working time according to the work that is available – even
if this means working late in the evening, at night, or on weekends.”
Strongly
agree
Agree Partly Disagree Strongly
disagree
Germany 20.7% 31.5% 23.9% 11.7% 12.2%
United States 39.3% 33.4% 18.0% 5.9% 3.3%
Source: Krzywdzinski/Gerber 2019. Weighted data.
To what extent do the factors that inuence the perception of crowdwork differ bet-
ween Germany and the United States (Tables 10 and 11)? Overall, very similar pat-
terns can be observed within the two countries studied. With regard to the question
of whether crowdwork represents a long-term model for the respondents, the current
level of platform income proved particularly relevant. People who earned relatively
high incomes on the platforms planned to continue working there in the future. In both
countries, people without a tertiary education were also much more likely to see their
personal future in crowdwork. In terms of employment status, self-employed people,
unemployed people, and retired persons in both countries were more likely to see their
future in crowdwork than those with regular employment.
Varieties of platform work 21
Table 10. Ordinal logistic regression, German subsample (odds ratios in brackets)
“I see my per-
sonal future in
platform work.”
“The uncertain-
ty of whether I
will get enough
jobs in a month
and earn enough
money worries
me a lot.”
“I organize my
working time
according to the
work that is avai-
lable – even if
this means wor-
king late in the
evening, at night,
or on weekends.”
Gender (women compa-
red to men)
.262 (1.300) .053 (1.019) .340 (1.405)
Tertiary education (com-
pared to persons without
tertiary education)
-.606 (.545)** .116 (1.123) .044 (1.045)
Platform income .417 (1.518)* -.151 (.860) -.052 (.949)
Number of inhabitants of
the place of residence
-.043 (.958) .111 (1.118) .062 (1.064)
Employed part time
(compared to full-time
employment)
.435 (1.545) .667 (1.949)* -.061 (.941)
Self-employed (compared
to full-time employment)
.638 (1.893)* 1.177 (3.245)** .551 (1.736)*
In education (compared
to full-time employment)
-.134 (.875) .266 (1.305) -.383 (.682)
Unemployed/retired/other
(compared to full-time
employment)
.934 (2.545)** 1.297 (3.658)** .783 (2.188)*
Age -.011 (.989) -.015 (.985) -.020 (.980)*
Number of children .086 (1.090) -.024 (.976) .078 (1.081)
N 534 547 554
Pseudo R² 0.047 0.046 .034
Signicance level * <0.05, **<0.01. Controls: platform, duration of work experience
on platforms, number of platforms on which the person is registered. Robust standard
errors. Weighted data.
22
Table 11. Ordinal logistic regression, US subsample (odds ratios in brackets)
“I see my per-
sonal future in
platform work.”
“The uncertain-
ty of whether I
will get enough
jobs in a month
and earn enough
money worries
me a lot.”
“I organize my
working time
according to the
work that is avai-
lable – even if
this means wor-
king late in the
evening, at night,
or on weekends.”
Gender (women compared
to men)
-.017 (.983) .263 (1.301) .282 (1.325)
Tertiary education (com-
pared to persons without
tertiary education)
-.411 (.663) .210 (1.233) -.598 (.550)*
Platform income .324 (1.383)** -.139 (.870) -.002 (.998)
Number of inhabitants of
the place of residence
-.289 (.749)** -.008 (.992) .042 (1.042)
Employed part time
(compared to full-time
employment)
-.271 (.763) .556 (1.744) .061 (1.063)
Self-employed (compared
to full-time employment)
.603 (1.827)* .704 (2.022)* .394 (1.483)
In education (compared to
full-time employment)
-.356 (.700) .523 (1.687) -.035 (.965)
Unemployed/retired/other
(compared to full-time
employment)
.844 (2.326)* .319 (1.376) -.009 (.991)
Age .011 (1.011) -.007 (.992) .004 (1.004)
Number of children .224 (1.251)** .021 (1.022) .082 (1.085)
N 431 445 446
Pseudo R² 0.062 0.016 .046
Signicance level * <0.05, **<0.01. Controls: platform, duration of work experience
on platforms, number of platforms on which the person is registered. Robust standard
errors. Weighted data.
Varieties of platform work 23
There were certain differences between Germany and the United States regarding the
long-term importance of crowdwork. In the United States people living in smaller
cities were more likely to accept this form of work than people living in metropolitan
areas – this is plausible, because crowdwork offers particular opportunities in rural
labor markets. Interestingly, however, this variable was not important in Germany. In
the United States, the number of children a person had also clearly inuenced their
perception of crowdwork: The more children they had, the more likely they were to see
crowdwork as their future form of work. This, in turn, can be explained by the lack of
social welfare benets, which puts particular pressure on large families, as well as the
high cost of childcare in the United States.
When it comes to the question of uncertainty, there were again extensive similarities
between Germany and the United States. In both countries, having a high income from
platform work counteracted the burden of uncertainty. In both countries, self-employ-
ed people were much more likely to feel insecure than platform workers in regular
full-time employment; to a lesser extent, this also applied to part-time employees, the
unemployed, and pensioners. Most other variables had little or no impact.
Likewise, with regard to the allocation of working time, the countries were largely
similar. In both Germany and the United States, women were slightly more likely than
men to strongly gear their crowdwork working hours according to demand. People
in regular full-time employment and students tended to have lower probabilities of
working on-demand, while the opposite was true for the self-employed. The level of
platform income had no inuence on this. However, a clear difference can be seen with
regard to education levels. In Germany, the level of education had no inuence on how
exible and on-demand working hours were. In the United States, by contrast, people
without a college degree were about twice as likely to report having on-demand wor-
king hours than people with a college degree.
Regulation of platform work in Germany and the
United States
Particularly in the debate on dualization, scholars have pointed out that labor market
inequalities and segmentation have an inuence on policy formulation (Häusermann/
Schwander 2012). This argument – as we will illustrate below – also applies to the
platform economy. There have been clear differences in policy responses between
Germany and the United States. To explain these differences, it was necessary to take
into account the legal framework for collective action by platform workers. Although
24
our empirical analysis focused on crowdwork, in the following we summarize the re-
gulatory responses to platform work in a broader sense.
The rst clear difference in terms of policy responses to the development of platform
work in Germany and the United States concerns litigation through the courts. In Ger-
many, there has only been one publicly known lawsuit to date: A crowdworker sued
the Roamler platform (see Zeit Online 2019). He had been working on the platform
for two years and earned his income from this platform alone, however, the platform
ultimately terminated his access. In the lawsuit, the crowdworker argued that he was
de facto employed by the platform and therefore a termination of his access without
notice was unlawful. The State Labor Court in Munich ruled that there were no charac-
teristics of dependent employment because the crowdworker was not directly bound
by instructions of the platform and could have refused to accept the offered jobs.
In the United States – and especially in California – there have been a large number of
class action lawsuits led by platform workers, almost all of which revolve around the
question of whether platform workers are in fact dependent employees (Seiner 2017).
However, the most publicly discussed lawsuits against Uber, Lyft, and Crowdower
ended with a settlement in which the platforms paid millions of dollars to the platform
workers, thus avoiding a court decision on the question of dependent employment. A
number of other claims have also been dismissed by the courts using similar arguments
as in Germany (ibid.).
In both countries, the legal disputes have thus revolved around the question of plat-
form workers’ dependence on the platforms. The legal demarcations between depen-
dent employment and solo self-employment differ only slightly between Germany and
the United States. The central criterion for dependent employment in Germany is the
direct obligation to follow instructions (Weisungsgebundenheit), which enables the
employer to determine working hours and work content (cf. Kocher 2012; Müller-Glö-
ge 2020). This is interpreted relatively strictly. In the United States, the primary criteri-
on is the degree of “control” of the employee by the employer – the greater the control,
the more the relationship between two parties is to be interpreted as an employment
relationship (Seiner 2017).
There are several explanations for the different numbers of legal disputes in Germany
and the United States. Differences in the legal system certainly play an important role.
US law permits class actions under certain conditions, and California has particular-
ly favorable conditions for class actions in the eld of labor law (ibid.). Under these
conditions, and due to the long-standing weakening of trade unions, a trend of moving
labor disputes from the arena of industrial relations to the courts can be observed in the
Varieties of platform work 25
United States (Colvin 2012). In Germany, there is no such possibility of class actions,
which makes the conditions for a court-based collective dispute much more difcult. A
second explanation, however, pertains to the differences in the importance of platform
work for the respective labor markets. In the United States, platform work is much
more widespread, and it often represents the main source of income for people. This is
a much stronger motivation and basis for collective disputes, even if they are carried
out in court.
Differences between Germany and the United States in terms of political responses
to platform work have not only been evident in legal disputes, but also in state po-
licies. In the United States, there has been a conict between the Republican states
and federal government – and the National Labor Relations Board (NLRB) control-
led by it – on the one hand and progressive, Democratic states such as California
on the other. The NLRB, which oversees compliance with the rules for unionization
and recognition, considers platform workers to be freelancers and denies them the
right to unionize (Griswold 2019). A number of Republican states have passed legis-
lation dening platform workers as independent contractors. The conict is currently
escalating in California (but also in Massachusetts, New Jersey, and Connecticut). In
2019, California Assembly Bill 5 laid down stricter rules for recognition as indepen-
dent contractors (Smith 2019).5 This immediately led to new class actions by platform
workers (Scheiber 2019) – but also a petition by the three major platforms, Uber, Lyft,
and DoorDash. In this petition, they formulated an alternative that would exempt the
platforms from the law and at the same time create additional protections for self-em-
ployed platform workers. To date, they have already collected 1 million signatures and
can therefore present their alternative to California’s voters in November 2020 (Kerr
2020).
In Germany, the disputes have been less erce. The German government has commis-
sioned a number of expert reports and studies on the topic of platform work (Leist et
al. 2017; Greef/Schröder 2017; Bonin/Rinne 2017) but has not yet decided on specic
regulatory steps. The “Green Book Working 4.0” (Grünbuch Arbeit 4.0) published by
the Federal Ministry of Labour and Social Affairs (BMAS) formulated two central
positions. First, the ministry emphasized that the growth of platform work in Germany
has been limited so far and that there has been no sign of precarization through the
platforms (BMAS 2015: 175). Second, BMAS has pointed out that the existing ru-
les for determining bogus self-employment have so far been sufcient (BMAS 2015:
5 Only platform workers who really work without control, who are not active in the same industry as
the platforms, and who offer their services to several clients are to be classied as self-employed.
26
172). At the same time, however, BMAS has stressed that the government must conti-
nue to review the need to protect platform workers (BMAS 2015: 176).
Finally, third – in addition to legal disputes and state action – different approaches
have also been evident among the trade unions. Both in the United States and in Ger-
many, trade unions have tried to support the organization of platform workers and thus
to limit and push back against the development of an employment sector that is unre-
gulated by collective agreements. In the United States, unions are mobilizing to tighten
the regulations on bogus self-employment and support the strikes and class actions of
platform workers against the platforms (Ghaffary/Campbell 2019). In Germany, too,
the trade unions are demanding a tightening of the law (Haipeter/Hoose 2019) and
have supported the above-mentioned lawsuit. However, in Germany, the only example
of a collective labor dispute organized by platform workers so far involved riders for
food delivery platforms Deliveroo, Foodora and Lieferando (Degner/Kocher 2018;
Schreyer/Schrape 2018). Due to the weak mobilization power of the platform workers
and unions’ lack of experience with platform work, the two strongest trade unions –
ver.di and IG Metall – have been cautious in their approaches so far and are counting
on dialogue with the platforms.
IG Metall launched the “Fair Crowd Work” (FCW) portal in 2015. The portal allows
crowdworkers to nd advice, evaluate platforms, and exchange information. In addi-
tion, IG Metall, in cooperation with eight major German platform companies and the
German Crowdsourcing Association, adopted a code of conduct in 2017 that lays down
rules for fair platform work. This includes a voluntary commitment to fair payment
modes and remuneration levels, to appropriate task and time planning, and to feedback
and transparency. An ombudsman’s ofce was set up to ensure implementation and to
settle disputes. To date, this ofce has dealt with around 30 cases. The services trade
union ver.di has also set up a consulting center for crowdworkers, which is part of the
union’s broader advisory services for the self-employed. Like IG Metall, ver.di is also
in contact with individual platforms. As the conclusion of the code of conduct makes
clear, platforms in Germany are willing to talk to trade unions. This attempt to regulate
platform work between companies and trade unions is certainly typical of the traditi-
ons of German industrial relations but may also be related to the fact that platforms in
the German labor market have to make much greater efforts to ensure the legitimacy
of platform work than is the case in the United States.
Varieties of platform work 27
Conclusions
Does platform work represent a global model that makes the role of national and in-
stitutional differences disappear? Certainly, there are some arguments in favor of this
strong globalization thesis. Most platforms operate across national borders. Moreover,
there are hardly any differences between the approaches to labor and performance re-
gulation on the various platforms in different countries (Gerber/Krzywdzinski 2019).
The platform as an organizational form thus has its origins in the United States (Rah-
man/Thelen 2019) but has become a global phenomenon. However, the platforms are
embedded in different models of capitalism. This comparison between Germany and
the United States has shown that institutional factors play a role in two ways.
First, it is apparent that existing labor market segmentation and social welfare systems
determine who works on platforms and to what extent: In the United States, platform
work is more often a person’s main source of income; in Germany it is more likely to
be a secondary source of income, especially during college and university education.
Accordingly, crowdworkers in the United States perceive the insecurity and burdens
of platform work much more than their German counterparts. There are also differen-
ces between Germany and the United States with regard to education: In the United
States, the situation of crowdworkers without tertiary education is much more preca-
rious compared to crowdworkers with tertiary education, while these differences are
smaller in Germany. In contrast, with regard to gender inequality, there are similarities
between Germany and the United States. Women working on crowdwork platforms
are often in a particularly precarious position: They are more likely to be in part-time
employment or otherwise unemployed, while men more often engage in crowdwork
alongside full-time employment.
Against this background, it may be surprising to hear that crowdworkers’ satisfaction
with their work is higher in the United States than in Germany. However, existing re-
search shows that job satisfaction is not an indicator of objective working conditions.
In the case of crowdwork, it can be assumed that US crowdworkers’ higher satisfac-
tion is due to the fact that they have few alternatives to platform work and therefore
perceive the exibility of platform income as a gain in autonomy. In addition, there is
a greater cultural acceptance of self-employment and insecurity in the United States.
Second, institutional differences can help explain the policy responses. The spread of
platform work in the United States and its importance as a main source of income has
led to a large wave of lawsuits in the courts – this is typical for conict resolution in
the United States (Colvin 2012). The importance of platform work as a main source of
income for many platform workers has led individual states such as California to push
28
ahead with regulatory projects. Despite intensive discussions, by contrast, the German
government has not yet decided to regulate platform work. One reason for this is that
platform work in Germany is still often an additional income source earned alongside
regular employment. At the same time, the German government considers the existing
social welfare system and labor market regulations as sufcient.
Our analysis thus shows that the platform economy can play a very different role in
different national labor markets. The weaker the social safety net, the more likely
platform work is to be both a curse and a blessing: It offers a much needed and exib-
le source of income, albeit under extremely precarious conditions. The stronger the
social support, on the other hand, the greater the market power of workers vis-à-vis
the platforms. At the same time, however, the possibilities for individual exit limit the
willingness of platform workers to organize themselves. It also reduces the pressure
for regulatory responses.
Despite these differences between Germany and the United States, we are convinced
that the regulation of platforms should be an issue on the political agenda in both
countries. First, it is necessary to consider introducing social welfare benets for plat-
form work, even if this often represents “only” a secondary income source. Second,
there is a need for regulation with regard to power and information asymmetries in the
algorithmic management of platforms. This includes securing rights for platform wor-
kers to have a say in determining rules on the platforms (e.g., with regard to the terms
of trade, performance control, or the design of ranking systems). We need to develop
ideas for how this participation of the platform workers could be organized.
A limitation of our analysis is our empirical focus on crowdwork, an area in which
jobs are completed online and thus independent of location. Although we expect that
our results will essentially also apply to platforms that organize more location-bound
activities, this requires further research.
References
Adams, Abi & Berg, Janine (2017). When home affects pay: An analysis of the
gender pay gap among crowdworkers. Available at: https://ssrn.com/abs-
tract=3048711 (29.4.2020).
Beerepoot, Niels & Lambregts, Bart (2015). Competition in online job marketplaces:
towards a global labour market for outsourcing services?. Global Networks, 15(2),
236–255.
Varieties of platform work 29
Benassi, Chiara; Doellgast, Virginia & Sarmiento-Mirwaldt, Katja (2016). Institu-
tions and inequality in liberalizing markets: Explaining different trajectories of
institutional change in social Europe. Politics & Society, 44(1), 117–142.
Berg, Janine; Furrer, Marianne; Harmon, Ellie; Rani, Uma & Silberman, M. Six
(2018). Digital Labour Platforms and the Future of Work: Towards Decent Work
in the Online World. Geneva: International Labour Ofce.
Bonin, Holger & Rinne, Ulf (2017). Omnibusbefragung zur Verbesserung der Daten-
lage neuer Beschäftigungsformen. IZA Research Report No. 80, Bonn: IZA.
BMAS (2015). Grünbuch Arbeiten 4.0, Berlin: BMAS.
Brown, Andrew; Charlwood, Andy & Spencer, David (2012). Not all that it might
seem: why job satisfaction is worth studying despite it being a poor summary
measure of job quality. Work, Employment & Society, 26(6), 1007–1018.
Carchio, Claudia (2019). Platform work as a chance for a more inclusive labour mar-
ket. In Valeria Filì & Federico Costantini (eds.), Legal Issues in the Digital Eco-
nomy. The Impact of Disruptive Technologies in the Labour Market (pp. 73–104).
Newcastle: Cambridge Scholars Publishing.
Cherry, Miriam (2019a). Age discrimination in the on-demand economy and Crow-
dwork. Berkeley Journal of Employment & Labor Law, 40(1), 1–29.
Cherry, Miriam (2019b). A Global System of Work, a Global System of Regulation?:
Crowdwork and Conicts of Law. Tulane Law Review, 94; Saint Louis University
School of Law Legal Studies Research Paper No. 2019-11. Available at: https://
ssrn.com/abstract=3523303 (29.4.2020).
Colvin, Alexander (2012). American Workplace Dispute Resolution in the Indivi-
dual Rights Era. International Journal of Human Resource Management, 23(3),
459–475.
Cutcher-Gershenfeld, Joel & Kochan, Thomas (2004). Taking Stock: Collective Bar-
gaining at the Turn of the Century. Industrial and Labor Relations Review, 58(1),
3–26.
Cutolo, Donato & Kenney, Martin (2019). Playing in God’s Garden: Dependent
Entrepreneurs in a Platform Economy, BRIE Working Paper 2019-3, Berkeley:
University of California in Berkeley.
Degner, Anne & Kocher, Eva (2018). Arbeitskämpfe in der „Gig-Economy“? Die
Protestbewegungen der Foodora- und Deliveroo-„Riders“ und Rechtsfragen ihrer
kollektiven Selbstorganisation. Kritische Justiz, 51(3), 247–265.
30
De Stefano, Valerio (2016). The rise of the „just-in-time workforce”: On-demand
work, crowdwork, and labor protection in the „gig-economy”. Comparative Labor
Law and Policy Journal, 37(3), 461–471.
Difallah, Djellel; Filatova, Elena & Ipeirotis, Panos (2018). Demographics and Dy-
namics of Mechanical Turk Workers. Proceedings of the Eleventh ACM Interna-
tional Conference on Web Search and Data Mining, Marina Del Rey (CA), USA,
February 05–09, 2018, 135–143.
Emmenegger, Patrick; Häusermann, Silja; Palier Bruno & Seeleib-Kaiser, Martin
(eds.) (2012). The Age of Dualization: The Changing Face of Inequality in Dein-
dustrializing Societies. Oxford: Oxford University Press.
Farrell, Diana & Greig, Fiona (2016). Paychecks, paydays, and the online platform
economy. Proceedings. Annual Conference on Taxation and Minutes of the Annu-
al Meeting of the National Tax Association, 109, 1–40.
Friedman, Gerald (2014). Workers without employers: shadow corporations and the
rise of the gig economy. Review of Keynesian Economics, 2(2), 171–188.
Foong, Eureka; Vincent, Nicholas; Hecht, Brent & Gerber, Elizabeth M. (2018). Wo-
men (still) ask for less: Gender differences in hourly rate in an online labor mar-
ketplace. Proceedings of the ACM on Human-Computer Interaction, 2(CSCW),
1–21.
Gandini, Alessandro (2016), The Reputation Economy: Understanding Knowledge
Work in Digital Society. London: Palgrave Macmillan.
Gerber, Christine & Krzywdzinski, Martin (2019). Brave New Digital Work? New
Forms of Performance Control in Crowdwork. In Steven Vallas & Anne Kova-
lainen (eds.), Work and Labor in the Digital Age (Research in the Sociology of
Work, Vol. 33) (pp. 121-143). Bingley: Emerald.
Gerber, Christine (2020a). Community building on crowdwork platforms: Autonomy
and control of online workers?. Competition & Change, advanced access, 1–21.
Available at: doi.org/10.1177/1024529420914472.
Gerber, Christine (2020b). Crowdworker*innen zwischen Autonomie und Kontrolle.
Die Stabilisierung von Arbeitsteilung durch algorithmisches Management. WSI
Mitteilungen, forthcoming June 2020.
Gerstung, Valeska (2019). Niedriglohnbeschäftigung im Wohlfahrtsstaat: Der Ein-
uss von Ideen und Institutionen auf den Niedriglohnsektor. Wiesbaden: Springer
VS.
Ghaffary, Shirin & Campbell, Alexia (2019). A landmark law disrupted the gig eco-
nomy in California. But what comes next for Uber drivers? Available at: https://
Varieties of platform work 31
www.vox.com/recode/2019/10/4/20898940/uber-lyft-drivers-ab5-law-califor-
nia-minimum-wage-benets-gig-economy-disrupted (29.4.2020).
Grabka, Markus & Schröder, Carsten (2019). Der Niedriglohnsektor in Deutschland
ist größer als bislang angenommen. DIW-Wochenbericht, 86(14), 249–257.
Graham, Mark; Hjorth, Isis & Lehdonvirta, Vili (2017). Digital labour and develop-
ment: impacts of global digital labour platforms and the gig economy on worker
livelihoods. Transfer, 23(2), 135–162.
Greef, Samuel & Schröder, Wolfgang (2017). Plattformökonomie und Crowdwor-
king: Eine Analyse der Strategien und Positionen zentraler Akteure, Forschungs-
bericht 500, Berlin: BMAS.
Griswold, Alison (2019). The future for Uber drivers won’t be decided by the US
government. Available at: https://qz.com/1621141/trumps-nlrb-says-uber-drivers-
are-independent-contractors/ (29.4.2020).
Häusermann, Silja; Enggist, Matthias & Pinggera, Michael (2019). Sozialpolitik in
Hard Times, in: Herbert Obinger & Manfred G. Schmidt (eds.), Handbuch Sozial-
politik (pp. 33–54). Wiesbaden: Springer VS.
Häusermann, Silja & Schwander, Hanna (2012). Varieties of dualization? Labor mar-
ket segmentation and insider-outsider divides across regimes. In Patrick Emme-
negger, Silja Häusermann, Bruno Palier & Martin Seeleib-Kaiser (eds.), The Age
of Dualization: The Changing Face of Inequality in Deindustrializing Societies
(pp. 27–51). Oxford: Oxford University Press.
Henderson, Alisa & White, Linda (2004). Shrinking welfare states? Comparing
maternity leave benets and child care programs in European Union and North
American welfare states, 1985–2000, Journal of European Public Policy, 11(3),
497–519.
Hoose, Fabian; Haipeter, Thomas & Ittermann, Peter (2019). Digitalisierung der
Arbeit und Interessenvertretungen. Arbeit, 28(4), 423–444.
Howcroft, Debra & Bergvall-Kåreborn, Birgita (2019). A typology of crowdwork
platforms. Work, Employment & Society, 33(1), 21–38.
Huws, Ursula; Spencer, Neil &Joyce, Simone (2016). Crowd work in Europe:
Preliminary results from a survey in the UK, Sweden, Germany, Austria and
the Netherlands. Available at: http://www.uni-europa.org/wp-content/uplo-
ads/2016/12/2016-12-Crowd-work-in-Europe.pdf (29.4.2020).
ILO (2017). Non-standard employment around the world. Geneva: International
Labour Ofce.
32
Irani, Lilly (2015). The cultural work of microwork. New Media & Society, 17(5),
720–739.
Ipeirotis, Panagiotis (2010). Demographics of mechanical turk, NYU Working Paper,
CEDER-10-01.
Isaac, Mike; Wingeld, Nick & Scheiber, Noam (2015). Seattle Considers Measure
to Let Uber and Lyft Drivers Unionize. New York Times, Dec. 13, 2015. Available
at https://www.nytimes.com/2015/12/14/technology/seattle-considers-measu-
re-to-let-uber-and-lyft-drivers-unionize.html?partner=IFTTT (29.4.2020).
Ivanova, Mirela; Bronowicka, Joanna; Kocher, Eva & Degner, Anne (2018). The App
as a Boss? Control and Autonomy in Application-Based Management. Available
at: https://opus4.kobv.de/opus4-euv/frontdoor/deliver/index/docId/388/le/Arbeit-
GrenzeFlussVol02.pdf (29.4.2020).
Jackson, Emilie; Looney, Adam & Ramnath, Shanthi (2017). The Rise of Alternative
Work Arrangements: Evidence and Implications for Tax Filing and Benet Cover-
age. Ofce of Tax Analysis Working Paper 114, Washington: The Department of
the Treasury.
Kaufmann, Franz-Xaver (2003). Varianten des Wohlfahrtsstaats. Der deutsche So-
zialstaat im internationalen Vergleich. Frankfurt am Main: Campus.
Kellogg, Katherine C.; Valentine, Melissa A. & Christin, Angele (2020). Algorithms
at work: The new contested terrain of control. Academy of Management Annals,
14(1), 366–410.
Kerr, Dara (2020). Uber, Lyft, DoorDash’s gig worker ballot initiative heads to voters
in November. CNET. Available at: https://www.cnet.com/news/uber-lyft-door-
dashs-gig-worker-ballot-initiative-heads-to-voters-in-november/ (29.4.2020).
Kocher, Eva (2012). Hausarbeit als Erwerbsarbeit: Der Rechtsrahmen in Deutsch-
land: Voraussetzungen einer Ratikation der ILO-Domestic Workers Convention
durch die Bundesrepublik Deutschland. Düsseldorf: Hans-Böckler-Stiftung.
Kröger, Teppo (2011). Defamilisation, dedomestication and care policy, International
Journal of Sociology and Social Policy, 31(7/8), 424–440.
Krzywdzinski, Martin & Gerber, Christine (2019). Working conditions of crowdwor-
kers. Survey of crowdworkers in Germany and the United States, Version 1.00,
WZB Berlin Social Science Center, unpublished data set.
Lee, Min Kyung; Kusbit, Daniel; Metsky, Evan & Dabbish, Laura (2015). Wor-
king with machines: The impact of algorithmic and data-driven management
on human workers. Proceedings of the 33rd Annual ACM conference on hu-
Varieties of platform work 33
man factors in computing systems, New York, NY: ACM, 1603–1612. doi:
10.1145/2702123.2702548.
Lehdonvirta, Vili (2016). Algorithms that divide and unite: delocalisation, identity
and collective action in ‘Microwork’. In Jörg Flecker (Hrsg.), Space, Place and
Global Digital Work. Dynamics of Virtual Work (pp. 53–80). London: Palgrave
Macmillan.
Leimeister, Jan Marco; Durward, David & Zogaj, Shkodran (2016). Crowdworker in
Deutschland. HBS Study Nr. 323. Düsseldorf: Hans-Böckler-Stiftung.
Leist, Dominik; Hießl, Christina & Schlachter, Monika (2017). Plattformökonomie –
Eine Literaturauswertung. Forschungsbericht 499, Berlin: BMAS.
Leung, Ming D. & Koppman, Sharon (2018). Taking a pass: How proportional preju-
dice and decisions not to hire reproduce gender segregation. American Journal of
Sociology, 124(3), 762–813.
MacDonald, Robert & Giazitzoglu, Andreas (2019). Youth, enterprise and precarity:
or, what is, and what is wrong with, the ,gig economyʻ?, Journal of Sociology,
55(4), 724–740.
Müller-Glöge, Rudi; Preis, Ulrich & Schmidt, Ingrid (eds.) (2020), Erfurter Kom-
mentar zum Arbeitsrecht, 20th edition, § 611 Rn. 50-58, Munich: C.H. Beck.
Piasna, Agnieszka, & Drahokoupil, Jan (2017). Gender inequalities in the new world
of work. Transfer: European Review of Labour and Research, 23(3), 313–332.
Pongratz, Hans (2018). Of crowds and talents: discursive constructions of global on-
line labour. New Technology, Work and Employment, 33(1), 58–73.
Pongratz, Hans & Bormann, Sarah (2017). Online-Arbeit auf Internet-Plattformen:
Empirische Befunde zum „Crowdworking“ in Deutschland. AIS-Studien, 10(2),
158–181.
Popiel, Pawel (2017). „Boundaryless“ in the creative economy: assessing freelancing
on Upwork. Critical Studies in Media Communication, 34(3), 220–233.
Rahman, K. Sabeel & Thelen, Kathleen (2019). The rise of the platform business mo-
del and the transformation of twenty-rst-century capitalism. Politics & Society,
47(2), 177–204.
Richter, Nancy; Jackson, Paul & Schildhauer, Thomas (2018). Entrepreneurial be-
haviour and startups: The case of Germany and the USA. In: Nancy Richter, Paul
Jackson & Thomas Schildhauer (eds.), Entrepreneurial Innovation and Leader-
ship. Preparing for a Digital Future (pp. 1-14). Cham: Palgrave Pivot.
34
Scheiber, Noam (2019). New Lawsuit Against Uber Is Set to Test Its Classication
of Workers. New York Times (Sept. 12, 2019). Available at: https://www.nytimes.
com/2019/09/12/technology/uber-drivers-california.html (29.4.2020).
Schneider-Dörr, Andreja (2019). Erwerbsarbeit in der Plattformökonomie. Eine kriti-
sche Einordnung von Umfang, Schutzbedürftigkeit und arbeitsrechtlichen Heraus-
forderungen. Working Paper 116, Düsseldorf: Hans-Böckler Stiftung.
Schreyer, Jasmin & Schrape, Jan-Felix (2018). Algorithmische Arbeitskoordination
in der plattformbasierten Gig Economy: Das Beispiel Foodora. AIS Studien,
11(2), 262–278.
Schwander, Hanna (2019). Labor Market Dualization and Insider–Outsider Divides:
Why This New Conict Matters. Political Studies Review, 17(1), 14–29.
Seeleib-Kaiser, Martin (2014). Wohlfahrtssysteme in Europa und den USA: Annähe-
rung des konservativen deutschen Modells an das amerikanische? WSI-Mitteilun-
gen, 4/2014, 267–276.
Seiner, Joseph (2017). Tailoring Class Actions to the On-Demand Economy. Ohio
State Law Journal, 78(1), 21–71.
Sering, Oliver (2018). Crowdworking Monitor Nr. 1, Kleve: Hochschule Rhein-
Waal.
Smith, Aaron (2016). Shared, collaborative and on demand: The new digital econo-
my. Washington D.C.: Pew Research Center.
Smith, Rebecca (2019). Washington State Considers ABC Test for Employee Status.
National Employment Law Project (Apr 1, 2020). Available at: https://www.nelp.
org/blog/washington-state-considers-abc-test-employee-status/ (29.4.2020).
Sousa-Posa, Alfonso & Sousa-Posa, Andres (2000). Well-being at work: a cross-na-
tional analysis of the levels and determinants of job satisfaction. Journal of So-
cio-Economics, 29(6), 517–538.
Thelen, Kathleen (2018). Regulating Uber: the politics of the platform economy in
Europe and the United States. Perspectives on Politics, 16(4), 938–953.
Thies, William; Ratan, Aishwarya & Davis, James (2011). Paid crowdsourcing as a
vehicle for global development, Workshop paper for the CHI 2011 Workshop on
Crowdsourcing and Human Computation, (8 May, 2011), Vancouver, BC, Canada.
U. S. Bureau of Labor Statistics (2018). Contingent and alternative employment
arrangements. News Release. Available at: https://www.bls.gov/news.release/co-
nemp.nr0.htm (29.4.2020).
Varieties of platform work 35
Vallas, Steven & Schor, Juliet (2020). What Do Platforms Do? Understanding
the Gig Economy. Annual Review of Sociology, 46. doi.org/10.1146/annu-
rev-soc-121919-054857.
Van Doorn, Niels (2017). Platform labor: on the gendered and racialized exploitation
of low-income service work in the ‘on-demand’ economy. Information, Communi-
cation & Society, 20(6), 898–914.
Wood, Alex; Graham, Mark; Lehdonvirta, Vili & Hjorth, Isis (2018). Good gig, bad
gig: autonomy and algorithmic control in the global gig economy. Work Employ-
ment and Society, 33(1), 56–75.
Zeit Online (2019). Crowdworker sind Selbstständige, Available at: https://www.zeit.
de/wirtschaft/2019-12/landesgericht-muenchen-crowdworker-selbststaendige-ur-
teil (29.4.2020).
Zyskowski, Kathryn; Ringel, Meredith; Bigham, Jeffrey; Gray, Mary & Kane, Shaun
(2015). Accessible crowdwork? Understanding the value in and challenge of mic-
rotask employment for people with disabilities. Proceedings of the 18th ACM con-
ference on computer supported cooperative work and social computing, (14–18
March, pp. 1682–1693). Vancouver, BC, Canada. doi: 10.1145/2675133.2675158.
