Skills for the Planetary Labour Market: Indian Workers in the Platform Economy - Fairwork
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
1 Skills for the Planetary Labour Market: Indian Workers in the Platform Economy
2
Table of Contents
1. List of Abbreviations 3
2. List of Boxes, Figures and Tables 3
3. Introduction and Summary 4
3.1. Key Findings 4
3.2. Policy Recommendations for Development Cooperation 7
4. Research Context 8
4.1. Cloudwork and the Planetary Labour Market 8
4.2. Skills for Cloudwork 9
4.3. Platform Workforces in India 11
4.4. Skills Training Systems in India 12
5. Methods 16
5.1. Research Questions 16
5.2. Research Design 15
5.3. Sample 18
6. Impacts of the COVID-19 Pandemic 19
6.1. The COVID-19 Pandemic’s Impacts on Cloudworkers and Cloudwork 19
6.2. Interpreting Research Findings 22
7. Developing Skills for Cloudwork 23
7.1. Useful Skills for Cloudwork 24
7.2. Skills Developed on-the-job 24
7.3. Skills Developed during Leisure Time 25
7.4. Learning Activities 29
8. Strategies and Obstacles for Skills Development 32
8.1. Self-directed vs Collaborative Learning Activities 32
8.2. Obstacles to Undertaking Skills Development 34
8.3. Strategies to Overcome Obstacles 35
9. The Platform Economy and Skills Training 37
9.1. Cloudwork and the Indian Skills Ecosystem 37
9.2. Continuities and Discontinuities 42
10. Conclusion: Policy Recommendations for Development Cooperation 49
11. Appendix A: Methodology Note 54
12. Credits and Funding 62
13. References 63
3
1. List of Abbreviations 2. List of Boxes, Figures and Tables
BMZ Federal Ministry of Economic Cooperation and
Development (Germany) Boxes
CEO Chief Executive Officer
In Focus: Platform Profile Ratings
DGT Directorate General of Training In Focus: Working while Having Caring Responsibilities
FAQ Frequently Asked Questions
GDP Gross Domestic Product Figures
GIZ Deutsche Gesellschaft für Internationale
Zusammenarbeit GmbH Figure 1. The Distribution of the Indian Internet Users
by Gender and Age in 2019
HIT Human Intelligence Task (a task on Amazon
Figure 2. Yearly Central Government Budget Allocation
Mechanical Turk)
for Skills Development
HR Human Relations
Figure 3. Geographic Distribution of Survey Respondents
ILO International Labour Organisation Figure 4. Sign-ups on Cloudwork Platforms during the
IT Information Technologies COVID-19 Pandemic
Figure 5. Time Spent on Cloudwork Platforms during the
ITeS Information Technology Enabled Services
COVID-19 Pandemic
ITI Industrial Training Institute
Figure 6. Skill Categories Perceived Useful for Work on
MSDE Ministry of Skill Development and the Platform
Entrepreneurship Figure 7. Skill Categories Developed while Working on the
MTurk Amazon Mechanical Turk Platform
Figure 8. Skill Categories Developed during Leisure Time
NASSCOM National Association of Software and Service
Companies Figure 9. Learning Activities Undertaken by Workers
Figure 10. Learning Activities that Workers were Unable to
NCVET National Council of Vocational Education and
Undertake
Training
Figure 11. Screenshot from the eSkill India Portal Website
NGO Non-Governmental Organisation
Figure 12. Attitudes about the Indian Government
NSDC National Skill Development Corporation
Figure 13. Screenshot from Upwork Blog
NSQF National Skills Qualification Framework Figure 14. Screenshot from Freelancer Website
PLU Platform-led Upskilling Figure 15. Screenshot from the GigIndia Website
PM Prime Minister Figure 16. Mixed-Methods Research Design
Figure 17. Sample Characteristics
PMKVY Pradhan Mantri Kaushal Vikas Yojana
RASCI Retailers Association’s Skill Council of India
SSC Sector Skill Council
Tables
TVET Technical and Vocational Education and Training Table 1. Summary of Included Platforms
UNESCO United Nations Educational, Scientific and Table 2. Platform and Gender Distribution of Survey
Cultural Organization Respondents
UNEVOC UNESCO International Centre for Technical and Table 3. Typology of Skills
Vocational Education and Training Table 4. Typology of Learning Activities
UNFPA United Nations Population Fund Table 5. Platforms, Informal Work and Vocational
Training: Implications for TVET
WEF World Economic Forum
Table 6. Overview of Stakeholder Interview Participants
4
3. Introduction
Methodologically, we structure our
empirical inquiry of the five research
questions around a mixed-methods
and Summary research approach, combining quantitative
and qualitative methods to collect and
analyze our data. The qualitative phase
included semi-structured interviews
With 560 million internet users, India is the second
with 12 leading TVET stakeholders,
largest digital market in the world. People between including experts from India’s National
20 and 39 years of age represent just over half of the Skill Development Cooperation and
country’s internet users. This means that workers sector skill councils, as well as platform
located in India may overcome some of the constraints managers, among others. We additionally
carried out 31 in-depth interviews with
of their local labour markets by using digital labour
Indian platform workers. The quantitative
platforms to take part in a planetary labour market. So, phase involved the development of a
for many of these young internet users, digital labour novel survey instrument, which was
platforms provide an essential income and much- administered to over 400 Indian platform
needed opportunity to improve their livelihoods. Given workers across four platforms, and
representing the two different types of
the potentially important job opportunities offered
cloudwork: microwork and freelancing. We
by this planetary labour market, there is a need to included one microwork platform (Amazon
understand how Indian workers view and expand their Mechanical Turk) and three freelancing
skills portfolios in relation to their platform work, and platforms—which included two global
platforms (Upwork and Freelancer) and
unveil the points of leverage where existing training
one Indian platform (Truelancer).
initiatives by governmental, private and third-sector
actors may need to be adapted to meet their needs. Before presenting our empirical
evidence, we first set the stage for our
This report focuses on five key research study by situating it vis-a-vis relevant
literature and insights on the relationship
questions: between cloudwork platforms and skills
development in India, in section 4. We
1. What are the impacts of the COVID-19 pandemic on Indian pay particular attention to defining
cloudworkers and the cloudwork landscape? cloudwork and skills for cloudwork,
discussing the demographics of India’s
2. What skills are particularly relevant for cloudworkers in India and platform workforce, and introducing the
how are they acquired? country’s extant skills training systems.
We distinguish between government-led
3. What factors support or limit cloudworkers in acquiring the skills skills training, platform-led skills training,
they need for their work, and what strategies do they use to private skills training, and skills training
overcome obstacles? beyond organisations.
4. What connections and interactions exist between the platform On that basis, we describe the impacts
economy, informal work and skills training? that the COVID-19 pandemic has had
on Indian cloudworkers (RQ1) in section
5. What conclusions can be drawn for future skills training 6. In section 7, we elucidate which skills
approaches, in particular for development cooperation? are particularly relevant for cloudwork in
India while investigating the differences
across the four platforms and between
women and men (RQ2). Section 8 closely
5
examines what factors support or will grow in importance compared to Upwork focusing on learning skills, and
constrain cloudworkers in acquiring the traditional employment arrangements MTurk workers spending their leisure
skills they have identified as important to as a result of the pandemic. This time developing computer literacy and
develop for their work and what strategies corresponds with the finding that analytical skills.
they employ in order to overcome more companies are beginning to
obstacles (RQ3). In section 9, we outline outsource parts of their business • Similarly, differences in skill
the connections between the platform operations to temporarily engaged development between genders were
economy, informal work, and skills training cloudworkers. also somewhat more pronounced
(RQ4). Section 10 draws conclusions for skills developed during leisure
from the findings and offers practical time than during cloudwork. Women
recommendations for development 3.1.2. Skills for Cloudwork were slightly more likely to develop
cooperation, as well as to inform the both hard and soft skills on each
emerging policy landscape that governs • Indian cloudworkers find a variety platform while working, with the gap
cloudwork in India (RQ5). of skills to be useful for their somewhat wider when considering
work on online labour platforms. communication and organizational
3.1. Key Findings Communication skills, technical skills, skills. During leisure time female survey
platform-related skills, and learning respondents focused on language skills,
skills are particularly valued. computer literacy, and organizational
3.1.1. Impacts of COVID-19 on skills, while the men tended to focus
Cloudwork
• Indian cloudworkers tend to develop on improving communication, analytical,
different kinds of skills during their and technical skills.
work on platforms and during leisure
• Since the pandemic started, there
has been a marked surge in new
time. Respondents indicated a greater • Respondents reported that their
overall focus on soft skills developed rationales for on- and off-platform
sign-ups on cloudwork platforms.
during cloudwork with communication skills development varied, as
Across platforms, 18% of the survey
skills, personal dispositions, and skills developed while working are
respondents reported that they only
organizational skills being developed either needed to complete the job
started working on the platform
at least weekly by over 80% of the successfully or otherwise directly
where they were surveyed after
respondents. During leisure time benefit platform work, but developing
the pandemic started, often due to
respondents instead focused on skills during leisure time entails a
difficult circumstances that arose as a
developing technical skills and learning further investment of resources into
result of the pandemic.
skills, as well as platform-related skills. training skills that are more challenging
• 62% of veteran cloudworkers reported to develop during platform work,
that they are spending more time
• While the board patterns of skill but which have expected utility for
development are related across the platform work.
working or looking for work since the
four platforms, we saw more variation
pandemic started—though workers’
ability to dedicate more time to
between platforms in terms of skills • Across platforms, workers had similar
developed during leisure time than motivations for developing skills:
platform work may be constrained
with skills developed on-the-job. for instance whereas technical skills
by their working environment, caring
During cloudwork, the workers on were developed with an expectation
responsibilities, etc.
MTurk tend to differ in relation to the of leading to more clients and higher
workers on the other three platforms, paying jobs, languages were developed
• A recurring experience that survey
developing language skills and with the hope of facilitating finding
respondents and interviewees from
computer literacy relatively more often work and building a better rapport
across all platforms reported was that
and technical skills and most soft skills with clients.
they were facing substantially more
less often than the workers on other
competition, and were able to secure
platforms. During leisure time the
fewer jobs, during this past year.
focus of skill development varies more
with workers on Truelancer tending
• 79% of all survey respondents agreed
to focus on soft skills, workers on
or strongly agreed that cloudwork
6
3.1.3. Strategies and enable them to locate and access others. Platforms do often
Obstacles for Skills low-cost learning activities that have mechanisms for workers
Development align with their goals. to showcase their skills on their
profiles, but portability of platform-
• Workers face significant challenges • Survey respondents were also specific skills credentials (as well as
accessing and undertaking learning asked to reflect on the reasons work history and ratings) is limited,
activities, particularly in the absence why they were unable to undertake and leads to ‘locking in’ workers onto
of support from platforms. Since some of the learning activities they particular platforms.
cloudwork can be highly independent, indicated they wanted to take. Time
the exchange of tacit knowledge, like emerged as the central constraint • Whether workers can make use of
strategies for managing cloudwork to undertaking learning activities their “off-platform” experiences
through on-platform interaction across all platforms. Time in these and their educational background
can be limited. Platforms tend to be instances referred to several issues to increase their success “on-
designed to support individualised, including a lack of time due to caring platform” depends on a variety of
not collaborative work. Furthermore, responsibilities and the need to aspects: the platform interface and
learning outside of platform work make an income. This was followed its work allocation method; the type
hours can come at a high cost to closely by financial constraints to of task or gig; the expectations of
workers who are aware of the undertaking paid activities or missing clients; and the form of education
trade-off between investment work opportunities. Another key or experience. There is no simple
and income, one that does not issue was a lack of information and answer to the question of how
always even out. The willingness to awareness of potential places to education and experience shape,
invest in developing skills in leisure undertake learning activities, which and are shaped by, the interactions
time can be tied to the limited was also linked to a lack of time to between platforms, informal work
development opportunities offered look for these resources as well as and skills training, requiring tailored
by platforms, and is indeed one area a lack of guidance and mentorship. evidence-based solutions for
where workers expressed significant Together, these three reasons were building one’s profile when joining
interest in having the third sector given, in combination or alone, by and fostering upskilling on the basis
provide support. the majority survey respondents of different platform types.
when asked why they were unable
• Respondents reported generally to undertake the learning activities • Irrespective of workers’ education and
favouring self-directed learning despite expressing an interest in background and significant differences
activities over collaborative learning doing so. between platforms when it comes to
activities. Among the groups of reputation systems a common theme
learning activities, workers engaged to emerge from these interviews is
3.1.4. The Platform Economy therefore the need for a standard,
most often in individual self-
and Skills Training preferably platform-specific, skills
directed learning activities and in
platform-specific learning activities. training course that provides basic
• While skills development is a key information about profile building.
Taking paid online courses, webinars
priority for the Indian government, Against the backdrop of workers’
or tutorials was a specific challenge,
cloudworkers have been a blind spot lived experiences, it’s vital to improve
and one that was also particularly
in government-led TVET approaches. the portability of their profiles and
felt by women survey respondents.
There are numerous initiatives reputation or create external portals
Even when respondents indicated
spread across the Indian skills in which workers can show off their
the desire to undertake learning
ecosystem aiming to impart digital qualifications and experiences, without
activities, they faced challenges
skills, but none of them specifically being restricted to one particular digital
finding the financial and logistical
target cloudworkers, nor impart labour platform.
support to develop skills, an
platform-specific skills.
investment that workers did not
always feel able to make. This may
indicate that workers require more
• Platforms engage in very limited
direct skills training, though
structural forms of support to
some platforms do more than
7 3.2. Policy Recommendations for Development Cooperation 1. Engage with, and encourage the growth of, cloudworker networks and communities. 2. Convince platforms to support and invest in cloudworkers’ skills development. 3. Support the adjustment of existing institutional mechanisms to better cater for cloudworkers. 4. Support the creation of platform-specific training courses. 5. Advocate for a jurisdiction-spanning policy environment that would improve cloudworkers’ access to skills development. 6. Include access to skills development for cloudworkers in legislative action to foster due diligence in global supply chains. 7. Support further research and develop pedagogical tools to share knowledge about cloudwork. An Upwork worker at her desk
8
4. Research Context
4.1.1. Selective Formalisation
Cloudwork must be situated in an
empirical context in which huge numbers
4.1. Cloudwork and
When referring to platforms, it is vital of workers around the world are engaged
to be clear about definitions. Different
the Planetary Labour ideas of what platforms are and what
in non-standard and vulnerable forms
of employment characterized by low pay
Market they do entail various assumptions and no social protection—accounting
about what development actors should for two billion people, or almost
There are now over 50 million platform do about them. In other words, how we two thirds of the global workforce
workers who live all over the world, define platforms doesn’t just shape (ILO, 2021). Standard employment is
doing work that is outsourced via digital our analytical perspectives but also our traditionally defined as work that is
labour platforms or apps. It has been policy repertoire. In this report, we focus full-time, indefinite, and that includes
predicted that by 2025, one-third of all on the second type of platform work, statutory benefits and entitlements. In
labour transactions will be mediated cloudwork, and therefore use the term many places, including India, cloudwork
by digital labour platforms (Standing, ‘cloudworkers’ to refer to people who find is typically regarded as less secure
2016). A ‘digital labour platform’ is a work remotely through labour-broking than standard employment, owing to
company that uses digital resources cloudwork platforms, regardless of their the overwhelming proportion of self-
to mediate value-creating labour employment status (e.g. employees employed workers on these platforms.
interactions between consumers and or independent contractors).1 Digital This status quo frequently leaves
individual service-providing workers. platforms like Airbnb or eBay—where workers poorly integrated into social
goods are exchanged—or social media protection schemes, uncovered by key
platforms like Instagram are explicitly labour law legislation governing minimum
Digital labour platforms include two
not included within this definition. wages, unable to collectively bargain,
broad types (Woodcock & Graham,
2019). In the first, geographically- and with limited opportunities for skills
Cloudwork has also been called development—all things that may be
tethered work, platforms require a
‘crowdwork’, ‘online gig work’, and ‘online acutely dangerous for the economically
job to be done in a particular place
freelancing’ (Schwellnus et al., 2019). A precarious (Heeks et al., 2020).
(e.g. delivering food to an apartment
common denominator of all these labels
or driving a person from one part of
is that digital labour platforms are seen Research shows that cloudwork
town to another). As Shyamal Majumdar,
to be functioning as cross-jurisdictional platforms exacerbate particular
Former Head of the UNESCO-UNEVOC
infrastructures that mediate labour features of informal work by carefully
International Centre for Technical and
transactions, bringing together workers deploying narratives of flexibility,
Vocational Education and Training, puts
in one country with clients anywhere on freedom, and entrepreneurship while
it, these location-based types of work
the planet. Some commentators have systematically devolving social and
‘are executed in real world and locally but
promoted platform-mediated forms of economic risks onto workers, and
channeled through online independent
labour transactions and value creation positioning themselves as neutral
contracts.’ Cloudwork, in contrast, is
as a solution to reducing structural and intermediaries (Donner et al., 2019;
work that can, in theory, be performed
geographic inequalities by fostering Anwar & Graham, 2020; ILO, 2021).
from anywhere via the internet. This
inclusive development around the world That is, platforms formalise certain
means that cloudworkers located in
(Schwab, 2017). However, acting as a aspects of informal work by centralising
India can use digital labour platforms
counterweight to this optimism, a rich exchange between workers and
to take part in a ‘planetary labour
body of qualitative and quantitative clients while perpetuating the
market’ (Graham & Anwar, 2019) and
research has pointed to troubling precarity associated with informal
may overcome some of the constraints
features of cloudwork, including: labour markets—a phenomenon
of their local labour markets. Cloudwork
selective formalisation, exploitative that van Doorn (2020) has called
tends to be less well-known compared
working conditions, and oversupply of ‘selective formalisation.’ At the same
to more visible and urban forms of
labour (Graham et al., 2017; Anwar & time, cloudwork deviates from other
platform work.
Graham, 2020). types of work in that it is difficult to
1
We will use the terms cloudworker, platform worker, and worker interchangeably throughout the document.
9
contain within clearly defined, and of working conditions on cloudwork same freelancer platforms are careful
conventionally articulated, occupational platforms along five dimensions of to frame payment for platform
standards and job roles. work quality—pay, conditions, contracts, workers as optional (Graham & Anwar,
management and representation 2019). Guru.com prominently notes on
(Fairwork, n.d.). An upcoming report their website: ‘Pay only for a job well
4.1.2. Exploitative Working (Fairwork, 2021) that surveys working done.’ Upwork.com similarly states:
Conditions conditions on major cloudwork ‘Upwork…helps ensure that an hour
platforms across the world, echoes the paid is an hour worked’. Platforms put
While cloudwork platforms are findings from previous research that effort into highlighting the ephemeral
frequently celebrated for their the cloudwork model can frequently temporalities of the labour relationship
ability to offer workers access to a result in poor working conditions. to clients, and workers are often seen
global clientele and expanded work However, it also notes importantly that as an anonymous, replaceable group
opportunities, this can give rise to there exists a diversity of platform that can be brought together with their
working conditions that are frequently models, with some cloudwork platforms clients rather seamlessly.
characterized by long and irregular implementing policies to ensure fair
working hours, high levels of stress, and working conditions, and others falling In this study, we focus on a particular
other detrimental psychosocial risks short. dimension of cloudwork: the
and hazards (Wood et al., 2018; Berg, relationship between skills development
2016; Berg et al., 2018). Due to their and digital labour platforms. This
access to a huge number of spatially 4.1.3. Oversupply of Labour intersection is highly relevant because
dispersed workers, cloudwork platforms it relates to all three features
are sensitive to the fact that their With a planetary labour market linking mentioned above. It is hard now to
clients need some mechanism to be able together billions of people, most of imagine a world without some form
to trust that workers—who are often whom live in low- and middle-income of digital intermediaries that bring
on the other side of the planet—are countries, we have a system of truly together workers and clients. For this
indeed doing the work that they are planetary competition (Graham reason, the task of understanding
paying for. Platforms’ solution to this and Anwar 2019). In an already cloudwork is pivotal for anticipating
trust issue is sometimes an automated largely unregulated labour landscape, future challenges in the world of
surveillance system, for example cloudwork may further erode workers’ work—particularly skills development
which captures screenshots from structural power and fuel intense in an informal, and perhaps precarious,
workers’ computers at random intervals, competition between workers, in a context. Before presenting the results
contributing to stressful working digitally mediated marketplace that in sections 6-9, we will situate and
conditions. Most platforms additionally makes workers more easily replaceable contextualise our research questions
employ ratings-based performance (Huws, 2014). For almost all types of by discussing skills for cloudwork, the
management systems whereby clients digital work, there are fewer jobs than demographics of platform workforces
can ‘rate’ workers. Such ratings there are workers able and willing to do in India, and skills training systems.
systems can be opaque and arbitrary, them, creating an oversupply of labor
especially when they are based on that drives a global race to the bottom
4.2. Skills for
Cloudwork
algorithmic decisions, thus contributing (Graham & Anwar, 2019). This means
to the psychosocial risks and hazards of that when competing for jobs and tasks,
platform work. workers are well aware that others may
underbid them. The skills and abilities required for
There are several ways to conceptualise cloudwork are impacted by larger
the fairness of working conditions Cloudwork platforms are also careful transformations that have taken place
on cloudwork platforms (Harmon & to frame what does and does not over recent decades. The changing
Silberman, 2019; Whiting et al., 2019; count as working time. Indeed, one nature and organization of work
Graham et al., 2020). The Oxford study found that online freelancers combined with the accelerating speed
Internet Institute’s Fairwork Project spend an average of 16 hours a week of technological change have had
(of which the authors are members) (unpaid) looking for new jobs (Wood diverse impacts on societies around
periodically assesses the fairness et al., 2019). At the same time, those the world. As jobs have shifted from
manufacturing and agriculture to the
10
service, communication, transportation, facilities for importing professional or basis. Immigrants and women reported
and creative sectors, and alternative educational histories from outside, or developing their skills while doing
work arrangements have become more to export profile information from the cloudwork more often than men.
common, labour markets have come to platform, securing job opportunities
demand and reward different skill sets largely operates through algorithms Due to the scarcity of research in this
than they did in the past (Ojanperä matching workers to available jobs, area, little is known about the skill
et al., 2018). Many jobs call for human relying on platform-generated metrics acquisition of Indian cloudworkers.
capital and involve cognitive and socio- to rank search results and sort Based on interviews carried out with
emotional skills as well as adaptability submitted bids and applications. In 32 MTurk workers in 2013 in India,
and resilience. The introduction of new order to succeed on these platforms— Gupta (2017) identifies particular
technologies has permeated both i.e. to win bidding contests for projects skills developed by these cloudworkers,
innovative sectors that are driven by and acquire favourable reviews from including communication skills,
technological development as well as employer—workers must familiarize language skills, and technical skills.
more traditional sectors where new themselves with and learn to utilize Investigating various platform-led
technologies have been integrated into a wide array of practical skills and upskilling initiatives across 13 cloud-
existing practices, thus demanding strategies ranging from communication based and geographically tethered
the ability to navigate software and etiquette with often foreign clients platforms in Africa and Asia, Donner
hardware (Sudakov et al., 2016). to financial considerations about et al., (2019) find training content
budgeting and bidding for jobs with types to generally include technical
The emergence of the platform remuneration levels that are neither skills, general digital literacy, financial
economy offers unique opportunities too low to be suspicious nor too high literacy, platform proficiency, and
and challenges in terms of professional to be discouraging, to navigating the soft skills. Interviewing 24 Indian
skills and education. The multitude of many platform functionalities that are freelancers working on Elance-oDesk
different kinds of platforms host a subject to frequent revision as the (since renamed to Upwork), D’Cruz &
great diversity of job opportunities with platform updates its interface. Noronha (2016) respondents generally
varying levels of required skills—ranging reported being able to utilize and
from the microtasks offered by Amazon There is little research so far on the develop their skills while working on
Mechanical Turk, to the freelancing work kinds of skills that workers use or the platform. Surveying the attitudes
offered in a range of sectors by Upwork, develop for their work on the platform. of cloudworkers in India and Sri Lanka,
Freelancer, or Fiverr, to legal work Surveying European freelancers, Bandaranayake et al. (2020) noted
offered by sector-specific platforms like Cedefop (2020)—an agency supporting that women generally have positive
UpCounsel. While cloudwork platforms development of European vocational attitudes towards cloudwork, but are
differ in the kinds of work that are education and training policies— forced to take on smaller projects due
available through them, they defy the developed a typology of 10 skills that to the time constraints of their caring
mechanisms of more traditional labour cloudworkers tend to develop while duties.
markets by challenging the importance working on platforms. They note that
of formal education qualifications and successful cloudworkers are already
rather compensating for metrics of in possession of strong digital skills,
expertise and success on the platform technical skills related to the kind
in the form of a profile history of of work they do, communication and
completed jobs and favourable ratings organizational skills, and positive
(Herrmann et al., 2019). Relatedly, personal dispositions such as
formal educational qualifications are confidence and resilience, before
often designed to deliver a fixed set of they start working on a platform. The
skills and knowledge, agility for learning majority of the survey respondents
and a mentality for life-long learning reported that they did not further
while emphasizing labour-market develop their digital skills while working
related content to a lesser extent. on platforms, but two-thirds continued
Given their nature as self-contained developing their professional and
marketplaces for work with limited technical skills at least on a weekly11
4.3. Platform
Whereas agriculture formed the base by the Online Labour Observatory, the
of India’s economic activity three report notes a growth in India’s share
Workforces in India decades ago, employment has been of the labour supply on these platforms
steadily shifting towards industry from 26% in 2018 to 34% in 2020.
Given the potentially important job and services, which together account The country also provides the fifth-
opportunities offered by platforms, for over half of employment in 2016 largest share of labour demand, having
there is a need to understand how and contribute approximately 95% to increased from 5% in 2018 to 8% in
Indian platform workers view their skill gross value-added (Fairwork, 2020). 2020. The jobs posted by clients based
portfolios in relation to their platform The country has a large population of in India tend to mainly concern roles in
work, and where existing training highly educated English speakers, and software development and technology,
initiatives may need to be amended with earning a livelihood from cloud-based creative and multimedia, writing and
novel resources. The emergence of the gigs is an opportunity that appeals translation, and sales and marketing
platform economy and an associated to many Indians. In a country with support. The jobs carried out by Indian
rise in the platform workforce has a working-age population of nearly workers also mainly include tasks in
certainly been met with both policy and 900 million (Office of the Registrar software development and technology,
research interest in India. Estimates by General and Census Commissioner, creative and multimedia, and sales and
the Indian Readership Survey (MRUC, 2018) job creation is a key issue, in marketing support, with the proportion
2020) suggest that with 560 million particular for the younger cohorts of jobs carried out within the two first
internet users, India is the second joining the workforce, which grows by categories increasing over the two-year
largest digital market in the world. With approximately one million new workers period.
people between 20 and 39 years of every month.
age representing just over half of the The ILO report also analyzed the gender
country’s internet users (Figure 1). Men While it is difficult to estimate the size of of a random sample of workers from
make up two-thirds of the country’s the Indian cloud-based workforce, the the Online Labour Observatory across
internet users, with the gender gap ILO (2021) assessed the distribution of occupational categories and found that
being more marked in rural areas (Figure the global cloudwork supply and demand Indian women accounted for about a
1). For many of these young internet on four major cloudwork platforms, fifth of the country’s labour supply
users, platforms provide an essential finding India to be the largest supplier across the four platforms, which is
income and much-needed opportunity of labour on Fiverr, Freelancer, Guru, and lower than the other countries included
to improve their livelihoods. PeoplePerHour. Using data collected in the study—Ukrainian women formed
Figure 1. The Distribution of the Indian
Internet Users by Gender and Age in
2019. Source: MRUC, 2020
The Distribution of the Indian Internet Users by Gender and Age in 2019
Share of respondents, in % of subgroup
All India Rural Urban
70%
All India
Percentage of Internet users
Percentage of Internet users
60%
Location
50%
Rural
40%
30%
20% Urban
10%
100% 0% 20% 40% 60% 80% 100%
0%
50+ Female Male Female Male Female Male 12-15 16-19 20-29 30-39 40-49 50+12
39% and women from the United provides a brief and general overview development system was judged to be
States 41% of the collected sample. of the various organisations involved in non-responsive to labour market needs,
While Ukrainian and American women imparting skills training in the country, in that there was a demand-supply
tend to dominate men in jobs related distinguishing between government- mismatch in numbers of people trained,
to writing and translation, and reach led initiatives, platform-led initiatives, quality of training and skill types; skills
parity with men in jobs related to private initiatives, and self-driven skills curricula were ‘inflexible and outmoded’,
creative and multimedia, the proportion training that happens outside these and unsuited to market needs in the
of Indian women participating in these organisational pathways. Section 9, absence of systematic industry-faculty
categories remains comparatively low drawing on our fieldwork, delves more interaction; training programmes
at 47% and 21%, respectively. deeply into recent developments and resulted in low labour market outcomes
priorities in the Indian TVET ecosystem, for graduates; and there were poor
Much of the research on the Indian particularly as pertaining to cloudwork. testing, certification and accreditation
platform economy focuses on the systems for formal training, and a
Indian workers and clients on the complete lack of certification for
largest global platforms, based in the 4.4.1. Government-led Skills informally acquired skills (which was the
United States (Upwork), Australia Training case for the majority of the workforce).
(Freelancer), the United Kingdom There has since been substantial
(PeoplePerHour), and Israel (Fiverr). Skills development has been a core government investments and
Guru.com is a leading platform partially priority for the Indian government institutional reorganisation to reform
based in India and there additionally since the mid-2000s, and the this fragmented and ineffective system.
exist smaller Indian cloudwork institutional structures servicing this
platforms, such as Truelancer, which priority have evolved substantially The skills ecosystem in India is
contract work globally. Despite their over this period. India’s Eleventh now centrally coordinated, with a
global outlook and importance for Five-Year Plan (Government of India, clearer division of responsibilities
Indian freelancers and clients, even less 2008), spanning the 2007-2011 across different bodies. 2014 saw
is known about these local platforms, period, foregrounded this priority, with the establishment of a separate
which is why we have included an entire chapter devoted to ‘Skill Ministry for Skill Development and
Truelancer in our study. Development and Training’.2 Prior to Entrepreneurship (MSDE, n.d.), tasked
2009, some skills training was imparted with coordinating all skills development
4.4. Skills Training via formal TVET channels, though efforts across the country. The MSDE
Systems in India
most was done informally; across both, coordinates and works with the
only about 10% of the Indian labour following key organisations:
force was estimated to have received
The Technical and Vocational Education vocational training (Singh, 2012). The • Directorate General of Training
and Training (TVET) sector in India is formal channels were multiple, involving (DGT): The DGT develops and
complex, and comprises a wide array of vocational education in secondary coordinates long-term vocational
programmes and organisations. In this schools, higher technical education training programmes across
project, following UNESCO-UNEVOC through professional colleges (teaching the country, through a network
(2015) definition, TVET is understood engineering, medicine, agriculture, of around 15,000 private and
as ‘comprising education, training and etc.), technical training at specialised governmental Industrial Training
skills development relating to a wide institutions, and apprenticeship Institutes (ITIs), and 33 National
range of occupational fields, production, programmes. These were managed Skills Training Institutes (NSTIs),
services and livelihoods’. As such, by numerous governmental bodies National Skills Training Institutes
TVET includes the full gamut of skills spanning central ministries and for Women (NSTI-W) and other
development activities undertaken departments, and state governments. national-level institutes (MSDE,
at secondary, post-secondary, and
n.d.). DGT also oversees the ‘Bharat
tertiary levels, and includes work-based The Eleventh Five-Year Plan deemed Skills’ portal, an online self-learning
learning, and continuous training and this system not fit-for-purpose in platform for ITI students and
professional development. This section many respects (pp. 87-89): the skills teachers.
2
Between 1947 and 2017, the Indian Economy was organised through a series of Five-Year plans that outlaid the government’s priorities and the allocation of state
resources for the respective five-year period13
• National Skill Development and among vocational education, 2015, and the Pradhan Mantri Kaushal
Corporation (NSDC): The NSDC vocational training, general education Vikas Yojana (PMKVY) scheme (Press
is a public-private partnership and technical education, thus linking Trust of India, 2015). As recently as
that seeks to bridge industry and one level of learning to another in July 2020, PM Modi underscored
government in the skills ecosystem. higher level” (NSDA, n.d.). the skills development priority for his
A key role of the NSDC is to administration, announcing in an address,
coordinate and liaise with the 38 • Other government agencies
Sector Skill Councils (SSCs) that it and third sector organisations: “People ask me that in these
established in 2015. SSCs are non- The MSDE also collaborates times when businesses and
governmental trade associations in with other central ministries markets are changing so fast
different economic sectors, ranging and state governments, private how to stay relevant. This
question is even more important
from construction and agriculture to sector companies, international
during this COVID-19 pandemic.
Information Technology (IT) and IT organisations and NGOs on skills
The mantra to be relevant is to
enabled Services (ITeS). The NSDC development initiatives.
skill, reskill and upskill” (ANI,
works with SSCs to link the skills
2020).”
ecosystem to industry needs, for In accordance with the high priority
example through the development that the central government places on
of industry-specific occupational skills development, union budgets have
4.4.2. Platform-led Skills
standards, skills gap analyses, and also seen increasing yearly allocations
Training
training curricula. Particularly for jobs, skill development, and
relevant for this report is the IT- livelihoods schemes (see Figure 2).
Previous research on skills development
ITeS SSC NASSCOM, which is the
in the gig economy indicates that
national standard-setting body for The government, under Prime Minister
digital platforms may provide certain
IT skills across the country. The Narendra Modi’s stewardship, has in
minimal forms of direct skills training
NSDC additionally aims to play the recent years also launched numerous
to workers or other users. This has
role of a ‘market-maker’ by providing skills initiatives, including the ‘Skill India’
been referred to as ‘platform-led
subsidised financing for private mission, the National Policy for Skill
upskilling’ (PLU) (Donner et al., 2019)
training providers, and ensuring Development and Entrepreneurship
and ‘last-mile training’ (Craig, 2019).
that they make use of industry-
developed standards, curricula
and resources (The Pathways for Yearly Central Government Budget Allocation
Prosperity Commission, 2019). for Skills Development
• National Council for Vocational
Education and Training (NCVET):
Actual expenditure in Crores of Rupees
7,000 6,608
NCVET acts as an overarching
6,042
regulator to provide quality 6,000
assurance across the skills
5,000
ecosystem, by recognising and
regulating assessment agencies 4,000
and awarding bodies, and overseeing 2,723
3,000
and updating the National Skill
1,817
Qualification Framework (NSQF). 2,000
1,177
The NSQF oversees a “nationally 1,000
integrated education and
-
competency based skill framework 2015-16 2106-17 2017-18 2018-19 2019-20
that will provide for multiple
pathways, horizontal as well as Figure 2. Yearly Central Government Budget Allocation for Skills Development
vertical, both within vocational Source: Compiled from Union Budgets (The Government of India, 2017; 2018; 2019;
education and vocational training 2020; 2021).14
Such an approach to skilling workers benefits, and other specific aspects maintain an arms-length relationship
finds historical parallels in employer- of the organisation and operations with workers, so as not to become
led upskilling where companies train (Felstead & Jewson, 2014). Beyond this embroiled in questions of worker
their employees in occupational and minimum level, it has been argued that misclassification. One way that this
workplace skills. Access to in-work ultimately, “employees’ access to and arms-length relationship manifests is in
training varies vastly across workplaces, involvement in meaningful training and platforms’ limited involvement in direct
based on factors like the nature of the skill formation will be contingent on the skills training. The recent Cedefop study
employment contract (where longer- power relations which characterise the on digital work and learning found that
term ‘employees’ doing higher-skill work employment relationship and the quality online platforms usually skill workers only
receive more training), the flexibility of the job” (Unwin, 2017: 222). in indirect ways, such as by publishing
workers have over their work processes, information on which skills are in demand
and the degree to which workers are As explained in Section 4.1, (so workers can better develop their
involved in decision making (Unwin, cloudworkers—who are typically profiles), getting clients to provide
2017). Most employers, at a minimum, classified as self-employed workers or feedback on submitted work, referring
meet a ‘training floor’ (if only to comply independent contractors—often face workers to off-platform training
with regulatory requirements) where poor job quality, and rarely have a say resources, and facilitating worker
they may impart information to workers in how work is organised and managed communities where peer learning may
on orientation, safety, employee on platforms. Platforms are careful to take place (Cedefop, 2020). Concurrently,
An Upwork worker at his
desk15
however, platforms limit engagement in from platform proficiencies, financial private companies like Microsoft, IBM,
direct skills training activities: and digital literacies, technical skills, SimpliLearn, and others.
and soft skills, and examples of such
“Online labour platforms generally do PLU included webinars, Facebook live
not see a business case for more sessions, video tutorials, and Frequently 4.4.4. Skills Training beyond
direct involvement in training their Asked Questions (FAQ) webpages. Organisations
workers [...] training such workers is However, such training appears to
expensive and risky, because they may be less frequent in the case of the While not typically thought of as
take their skills elsewhere. Rather cloudwork platforms that are the focus forming a part of the TVET ecosystem,
than investing in directly supporting of this project and the Cedefop study previous research indicates that there
freelancers’ skill development, discussed above. are myriad informal ways that platform
platforms therefore invest in their workers acquire the skills, knowledge,
crowdworkers’ satisfaction, community and resources they need for their work,
promotion and marketing initiatives to 4.4.3. Private Skills Training without reliance on government, private
attract and retain skilled workers from training providers, or platforms. They
outside. Platform companies are also Alongside government-led efforts, can learn the skills they need for their
concerned that too much involvement numerous private education technology work from free resources like YouTube
in skill development and training could companies have emerged in the past tutorials, or by relying on advice from
risk them being potentially reclassified decade, offering Indians an array of friends, family, and mentors (Partnership
as employers, which they wish to avoid. ‘e-learning’ opportunities. Examples for Finance in a Digital Africa, 2019).
This is because, in many jurisdictions, include companies like UpGrad, Coursera, Workers also rely on communities of
the provision of training is considered Udemy, Skillshare and SimpliLearn, other workers found on messaging
one of the hallmarks of an employment where learners can pay for certified apps like Whatsapp, and social media
relationship; it could potentially be used courses on a variety of subjects. websites like Reddit or Facebook; for
to argue that platform workers should There are also an increasing number example, the ‘Turker Nation’ subreddit
be classified as employees in a lawsuit of global technology companies that caters specifically to MTurk workers, as
challenging their employment status. have developed platforms to impart does the ‘Turkers 101’ Facebook group
Overall, the platforms see themselves technical skills to learners, such as (Sherry, 2020). These avenues of self-
as having only a limited and indirect role Cisco’s ‘Networking Academy’ which organised, self-driven learning can be
to play in supporting crowdworkers’ skill was developed as a corporate social crucial for workers who lack access to
development” (Cedefop, 2020: 36-37). responsibility undertaking, or Google’s formal and institutionalised channels for
Developer Students Club. There also skills development.
Despite there being little incentive exist collaborative ventures between
for platforms to engage in extensive private training companies and the
direct skills training, other researchers Indian government; for example, the
have found some examples of such Director General of Training (DGT)
training. Donner et al., (2019) studied and MSDE signed an agreement with
a range of geographically tethered and Cisco and Accenture in late-2019
e-commerce platforms in Sub-Saharan to make available a digital skills
Africa and Asia, and found instances module to ITI students through the
of ‘platform-led upskilling’ (PLU) that Bharat Skills portal. Another notable
happened face-to-face, through online collaborative venture is the ‘e-Skill
training, and through in-workflow India Portal’, a multilingual e-learning
moments, that is, through ‘a subset of aggregator platform that the NSDC
nudges or other cues in which interface coordinates and maintains. The e-Skill
and experience design has been India portal is a web-based platform
deployed specifically to transfer skills that consolidates free and paid
to the user’. Training subjects ranged
3 digital learning content created by
3
It should be noted that ‘in-workflow moments’ where workers are ‘nudged’ to change their behaviours also has the potential to be manipulative and detrimental to
workers’ welfare; Noam Scheiber (2017) discusses how Uber nudges drivers to pursue work and earning targets that may be inimical to their interests.16
5. Methods 5.2. Research Design
needed to perform them? How has
the global upheaval caused by the
COVID-19 pandemic impacted these
structures? What lessons does the We structured our inquiry of the five
research questions around a mixed-
5.1. Research
intersection between TVET, digital
labour platforms, and (in)formal methods research approach, combining
Questions employment hold for development quantitative and qualitative methods
cooperation in a post-pandemic to collect and analyze our data and an
While the burgeoning platform economy context? initial theoretical phase consisting of
offers the promise of jobs for India’s desk research to ground our inquiry.
growing workforce, it also entails In order to begin addressing these We included four cloudwork platforms
sweeping challenges for how Technical broad questions and examine the in our research, representing the two
and Vocational Education and Training socio-economic factors grounding the different types of cloudwork: microwork
(TVET) programmes can be developed perceptions, choices, and actions of and freelancing. We included one
and implemented for these workers. Indian cloudworkers’ education, training, microwork platform (Amazon Mechanical
Given that cloudworkers are typically and skills development, our project Turk) and three freelancing platforms,
classified as independent contractors adopted five research questions: with two global platforms (Upwork and
rather than employees, there is a lack Freelancer) and one Indian platform
of formalised upskilling pathways RQ1: What are the impacts of (Truelancer). Table 1 describes the
officially accredited by India’s National the COVID-19 pandemic on main types of tasks carried out on the
Skills Development Cooperation (NSDC). Indian cloudworkers and included platforms.
Additionally, the diversity of cloudwork the cloudwork landscape?
is difficult to contain within clearly The quantitative phase involved
defined occupational standards and job RQ2: What skills are particularly development of a novel survey
roles, which tend to underpin many TVET relevant for cloudworkers instrument, which was administered
initiatives and projects in the country. in India and how are they to over 400 Indian platform workers
acquired? across the four platforms. We required
The increase in job opportunities the survey respondents to have
on cloudwork platforms, and the RQ3: What factors support worked for at least one month on one
multiplicity of platforms through which or limit cloudworkers of these platforms, be based in India,
freelancers can find opportunities, in acquiring the skills and be at least 18 years of age. Beyond
pose important questions for the they need for their work, comparing the populations working on
actors involved in the TVET landscape. and what strategies do the four included platforms, we were
For instance, how can skills training they use to overcome interested in interrogating whether
programmes be reformed and expanded obstacles? inequalities arise along other metrics
in ways that account for this shift that are relevant to the Indian context
online when it comes to requesting, RQ4: What connections and and may explain why some gig workers
accessing, and performing jobs and interactions exist between are able to acquire skills and progress
tasks? Which parties should be involved the platform economy, along cloudwork career pathways
in shaping policies, projects, and informal work and skills while others are not. Given the gender
initiatives? Is a new social contract training? disparities among Indian cloudworkers
between labour, business, and society that previous research has highlighted,
required, as global institutions like the RQ5: What conclusions can be we recruited roughly equal numbers of
International Labour Organisation drawn for future skills respondents who identified as ‘male’
and the World Economic Forum have training approaches, in and as ‘not male’ (including female, non-
implied? How does the overall structure particular for development binary, and other genders) in order to be
of the Indian cloud-based workforce cooperation? able to compare these groups as well.
look, including the types of jobs that
cloudworkers perform, and the skills17
The qualitative phase included 10 of their working environments (home the quantitative data. We additionally
semi-structured interviews with leading office, desk space, etc.), for use in this address our research questions with
TVET stakeholders, including experts report. None of the survey or interview more subjective reflections arising from
from India’s National Skill Development participants were among these ten the interviews with workers from the
Corporation (NSDC) and Sector Skill workers, in order to protect the former four platforms, which allow us to further
Councils, and platform managers, group’s confidentiality. explore the central themes that emerge
among others. We additionally carried from the survey data. Finally, we spread
out 31 in-depth interviews with Indian Throughout the report we draw our the stakeholder interviews through
cloudworkers from across the four findings from these varied levels of data the research project and used them to
platforms that we researched. The and we discuss insights from the survey gain insight into the central themes,
interviewees were drawn from the data in terms of the entire sample, and processes, and stakeholders related to
sample of survey respondents. A more comment on averages across platforms skills development of Indian platform
detailed description of the interview and or between men and women where workers.
survey methodologies, and the list of particular patterns emerge along those
stakeholder interviewees is presented in divisions. Where our survey questions
Appendix A, while Table 2 describes our have recorded open-ended responses, we
sample. In addition to the surveys and have identified common themes among
interviews, we also commissioned ten the respondents, which we employ to
Indian cloudworkers to take photographs add insight into the patterns we see in
Table 1. Summary of Included Platforms.
Source: Project data set
Platform Headquarters Platform Type Main Task Types
Amazon Seattle, Washington, USA Microwork Surveys, image identification, transcription
Mechanical Turk and annotation; content moderation; data
collection and processing; audio and video
transcription and translation, etc.
Freelancer Sydney, Australia Freelancing Web and software development; web and
graphic design; content and research writing;
Truelancer New Delhi, India Freelancing business services (e.g. accounting, HR, legal,
etc.); sales and marketing, etc.
Upwork Santa Clara, California, USA Freelancing
Table 2. Platform and Gender Distribution of
Survey Respondents. Source: Project data set
MTurk Freelancer Truelancer Upwork Total
Female 53 (51.9%) 47 (46.5%) 50 (49.5%) 50 (49.5%) 200 (49.4%)
Male 49 (48.1%) 54 (53.5%) 51 (51.5%) 51 (51.5%) 205 (50.6%)
Total 102 (25.2%) 101 (24.93%) 101 (24.93%) 101 (24.93%) 405 (100%)You can also read
