Skip to main content

Online Labour: The Construction of Labour Markets, Institutions and Movements on the Internet

Periodic Reporting for period 3 - iLABOUR (Online Labour: The Construction of Labour Markets, Institutions and Movements on the Internet)

Reporting period: 2018-09-01 to 2020-02-29

"The 'gig economy' or 'platform economy' has emerged as a key theme in debates around digitalisation and the future of work. It can be defined as ""people using apps and websites, commonly known as platforms, to sell their labour"". A recent UK government survey suggests that 4.4% of adults have worked in the gig economy in the last year and suggest that 2.4% of adults do so at least monthly; surveys from other European countries have produced similar figures.

The gig economy consists both of work that is transacted via platforms but delivered locally and thus requires the worker to be physically present, and work that is transacted and delivered remotely via platforms. Local gig work includes food delivery, couriering, transport and manual labour. Remote o r online gig work by contrast consists of the remote provision of a wide variety of digital services, ranging from data entry to programming and graphic design. Though locally delivered gig work has received much attention, surveys suggest that the remote type makes up a similar proportion of UK gig economy. Globally, 70 million workers are estimated to have registered with online labour platforms that facilitate remote forms of gig work.

The iLabour project examines how this platformization of labour markets is changing the rules or ""institutions"" that govern how labour markets work. To some extent platforms extricate workers and employers from national institutional frameworks, such as employment law and collective bargaining. But they are not free ""laissez-faire"" markets: they impose their own, often programmatically enforced rules or institutions. For example, a leading marketplace instated a global minimum wage of 4.00 USD/h. With hundreds of thousands of employers and millions of registered workers in 180 countries, this company is making important labour policy decisions that influence businesses and individuals from London to Manila.

The objective of the iLabour project is to lay bare the politics and institutions of these next-generation labour markets, focusing especially on the online gig economy. What are the most influential rules or institutions in the online gig economy? Whose interests do they reflect, and through what rule-creating processes are they made? How do workers express collective voice in this new market? And finally, two what extent are these relations still reducible to struggles between capital and labour, rather than more complex networked models of production?

We will tackle these questions through a combination of conventional social research methods and innovative digital research methods, on both virtual research sites (online labour markets and workers’ online forums) and physical research sites (platform companies' premises and worker gatherings). We survey, interview, and observe designers and workers to reconstruct processes through which online markets, institutions, and movements are shaped, and ""scrape"" online data to quantify their influence. The results will open up entirely new vistas in labour market research and policy debate."
"Work package 1: Overview and quantification of online labour markets

To obtain a quantiative overview of the online gig economy, we have built the ""Online Labour Index"", an automated data collector software system that collects data from major online platforms in real time. The system has been running at the Oxford Internet Institute and collecting data for over two years now.

Among other things, the data reveals that the utilization of online labour has grown approximately 30 percent globally over the past two years. The results are published as an automatically updated interactive visualisation and open data set at available at

Our article describing the ""Online Labour Index"" was published in the journal Techological Forecasting and Social Change. The data set has also been used in various other publications, including an OECD policy paper, a masters' thesis, and a doctoral thesis, shedding light on quantitative aspects of the online gig economy.

Work package 2: Online labour market institutions

To identify influential rules and policies (ie. institutions) in the online gig economy, and how those rules are created, we reviewed existing literature on online labor markets and the gig economy, studied the user interfaces of major online labor platforms, and negotiated access for qualitative/ethnographic fieldwork at a major online labor platform company. This allowed us to study closely how the platform company creates the rules of the platform and manages the platform.

A team member conducted a 1-year ethnographic study, combining traditional ethnographic methods of participant observation, in-depth interviews, and document analysis, with a digital ethnography of the platform’s online interactional spaces. This resulted in an extensive and unique dataset, unparallelled in its coverage - including data from a full year of etnographic fieldwork inside the platform, of the platform’s leadership team, of core strategic projects, and of the platform’s users. Analysis of this data is ongoing.

In connection with the ethnographic work inside the platform, we also collected and analyzed data on adoption of online labor markets by large enterprises. Early results from this work are published in a report titled ""Platform sourcing: How Fortune 500 firms are adopting online freelancing platforms"" (

In parallel with the ethnographic work, we have performed quantitative analyses of key online labour market institutions such as the reputation system and computer-administered skill testing system. Findings from this work will soon be published as an OECD working paper, titled 'Does Skill Testing Help New Workers: Evidence from Online Labour Markets'.

Work package 3: Online labour movements

To study how platform workers organise and express collective voice in the online gig economy, we have both analysed existing data sets as well as collected new data.

We have performed secondary analysis of existing data sets on platform worker organisation (semi-structured interviews (N=107) and survey data (N=658)), and these results have already been published in the journal New Technology, Work and Employment as “Workers of the Internet unite? Online freelancer organisation among remote gig economy workers in six Asian and African countries.”

As for new data, we started with scraping public online forum posts posted by workers, which we identified as constituting an example of platform worker voice (N=1100). This data was analysed geographically in order to map where in the world online platform worker contestation happens. This map of platform worker contestation was then used inform the choice of ethnographic fieldwork sites: San Francisco Bay, Los Angeles and New York City, London and Manila. Ethnographic fieldwork conducted in the aforementioned places in 2017 yielded 71 semi-structured interviews (20 US workers; 8 UK workers; 35 workers in the Philippines; 8 freelancer advocates) and participant observation of a dozen freelancer community events. Analysis of this data is ongoing.

Work package 4: Shifting boundaries and networks

To understand how platform work changes the nature of work, and in particular to examine whether it represents a shift towards a ""network economy"" or similar concepts put forward in earlier literature, we have so far analysed existing data sets from online freelancer interviews (N=107) and survey data (N=658). The results so far suggest that pre-existing local face-to-face networks are much more important even for platform-based workers than are virtual networks between geographically distant workers. We are currently preparing an article based on these findings."
"In Work Package 1, we have advanced the state of the art by creating the ""Online Labour Index"", which measures the online gig economy in real time, by directly tracking projects posted on online labour platforms. It provides insights that complement and go beyond what was possible with established approaches to measuring digital labour, especially online panel surveys and analysing national statistics. Among other things, the data set reveals that Amazon Mechanical Turk -style microwork, which previous research has largely focused on, is only a small subset of the total market for platform-based remote work. Before the end of the project, we further expect to link this dataset with other datasets, to yield insights on such questions as the value of flexible contracts in the gig economy, and the local economic impacts of online work.

In Work Package 2, we are expecting to significantly advance the state of the art on online labour market institutions before the end of the project, by publishing analyses from our unique ethnographic data set on how a major platform company makes platform rules and policies. In particular, we are expecting to produce articles on 1) how platforms are making important decisions on how to design and operate their marketplace, 2) how these comprise new institutions of work and labor markets, and 3) how organizations adopt and use online marketplaces to source work from freelancers.

In Work Package 3, our analysis of existing data, published in the journal New Technology, Work & Employment, already goes beyond the state of the art by presenting the first empirical findings on organising and collective voice among skilled online freelance workers. Previous research has focused almost exclusively on collective voice among microworkers. Going forward, the work package should yield the first ever detailed account and conceptualization of labour relations among online platform workers. It is expected that the analysis of our original data (combined with some further data gathering in the UK) will lead to an original understanding of how online labour platforms transform the nature of labour relations. Our initial analysis highlights that online labour platforms reduce the dependence of self-employed workers on their clients by reducing the costs of matching with new clients but in doing so these platforms also create new forms of platform dependency, reputational precarity and algorithmic insecurity. These developments seemingly reduce worker desire for traditional collective voice channels with regards to clients, such as collective bargaining, but simultaneously create desires for new forms of collective organisation which can provide platform workers with representation to platforms and governments. These findings will be of great value to policy makers, social partners, platforms and workers themselves.

In Work Package 4, our findings on collaborative networks between online workers represent an original contribution advancing the state of the art. Before the end of the project, we expect to produce additional secondary analyses of data that similarly examine and question the nature of platform-based work and the conceptualizations put forward in earlier literature, to arrive at an understanding of how and to what extent is digitalization transforming the nature of work."