Minimum wages on online labour platforms

15 June 2017 0

A response to the ETUI and IG Metall’s request for comment. Download the PDF at:

Wood, A., Graham, M., Anwar, M. A., Ramizo, G. 2017. Minimum wages on online labour platforms. Oxford Internet Institute.


Some members of our research group have recently completed a three-year study of workers on online platforms who live in Africa and Asia (140 interviews; > 600 survey responses; and six months of transaction data from one of the world’s largest platforms)1. Our research into this topic continues in two ongoing five-year projects2. The platforms that we looked at were global in nature, and in this response we focus on ‘non-geographically sticky work’ (i.e. ‘crowdwork’ or work that can, in theory, be done from anywhere).


Based on our research, we believe that any discussion of platform minimum wages is worth foregrounding with a few key points. First, for our Southeast Asian and Sub-Saharan African respondents, it was clear that pay rates were not the most important issue relating to the quality of their jobs. In fact, the pay rates were often significantly higher than what was available locally and were often considered to constitute decent pay. More important issues to emerge from our qualitative interviews and supported by our survey research were the limited social contact which workers experienced, that they often worked long or irregular unsocial hours at intense speeds, that many felt they had at little security and some had low incomes. Nevertheless, the downward pressure on pay rates created by the individualised and competitive design of online labour platforms contributed to these outcomes. However, they were also due to an oversupply of workers relative to clients meaning that there were inadequate earning opportunities to meet the needs of all workers and this in turn generally weakened the bargaining position of workers. Therefore, while implementing minimum wages on online labour platforms might alleviate some of these problems by increasing pay rates at the bottom, doing so might also exacerbate these problems by reducing the supply of clients (by making the platforms less attractive) while increasing the supply of workers (by making the work more attractive). Thus any intervention to increase a platform’s pay rates would require increases in the quality of the services provided in order not to reduce demand and exacerbate the weak position of labour.

Second, our empirical research highlights how the competition on many online labour platforms is international. What is more, we find that many workers perceive themselves as threatened with replacement by workers in other countries who are able to work for less due to the lower cost of living in that country. This international aspect is a key consideration in thinking about minimum wages, as any intervention is likely to unevenly affect workers living in diverse contexts. For example, a minimum wage set at North American or Western European levels would erode the comparative advantage of workers in lower income countries. This is not to suggest a race to the bottom in wages, but rather a need to make sure that minimum wages do not become an overly protectionist measure at the cost of workers in the Global South.

Third, our research has detailed that some platforms have implemented global minimum wages – mainly as an attempt to ensure quality by pricing out low-quality workers. However, a major issue with these minimum wages is that they relate only to hourly paid work when much of the work is paid on the basis of a fixed price per project. This means that the effective wage can be below the minimum hourly rate.

Labour Market Principles for Online Labour Platforms

There is currently insufficient empirical data to fully evaluate the likely labour market consequences of online labour platform minimum wages. Instead we suggest some general labour market principles which we believe should be applied to online labour platforms

First, all work that is done happens somewhere. Therefore, paid work undertaken through online labour platforms should fall under at least one set of national jurisdictions. There are few countries on the planet that do not have some form of regulated labour standards and minimum wage regulations. Therefore, online labour platforms must not exist as mechanisms for the avoidance of labour regulations. Just because a digital platform is used to connect a client with a worker, does not mean that the underlying economic and regulatory geography of that work should be ignored.

We should, as a starting point, adopt the principle that we do not need to reinvent the wheel. Online labour platforms should ensure that the relevant labour laws – including the classification of workers – are being followed. This is not an unusual expectation and it is widely accepted that conventional labour market intermediaries, such as employment agencies and labour brokers, have this responsibility.

When considering this issue it is useful to draw upon the discourse surrounding what is known as ‘tax dodging.’ Both tax evasion and tax avoidance are forms of tax dodging. While only tax evasion is illegal, as only these activities break the letter of the law, both evasion and avoidance are generally seen as harmful and immoral. We argue, therefore, that what matters, when thinking about labour regulation avoidance is the spirit of the law, not the letter of law.

Online labour platforms not only have a responsibility to ensure that the letter of the law is being followed but also the spirit of those laws. This is especially important regarding employment classification as minimum wages often only apply to those classified as “workers” or “employees”. In the spirit of the law, “self-employed contractors” are widely understood as being equal parties to those with whom they are entering into contracts with and thus do not require minimum wages. Conversely, “employees” are regarded as being the more vulnerable party in the relationship and in need of special protections such as minimum wages. However, in the contemporary labour market, many independent contractors are best understood as “self-employed workers” as they are in a vulnerable position due to dependence on a small number of clients and therefore in need of protections. Therefore, the spirit of these laws dictates that self-employed workers i.e. the vulnerable self-employed should be entitled to minimum wages as well as other protections outlined in relevant labour laws. However, much online gig work seems to be oblivious to, or ignore, those regulations. An employer based in Germany who sources work from a worker based in Kenya (via a platform based in the US) rarely has any knowledge of Kenyan labour law. It is also important to note that many countries’ minimum wage regulations include piece work. Under these laws employers are usually required to calculate a minimum piece rate which is not less than the hourly minimum if undertaken by a worker who is somewhat less skilled or more fatigued than the average worker.

Second, (and perhaps somewhat paradoxically), platforms should get rid of their global minimum wages. Global minimums send a message to clients that if they pay above the minimum then they are in compliance with relevant local regulations. However, it is entirely possible for workers to earn above platform minimum wages, but below their client’s national/local minimum wages

Third, we acknowledge that there might be claims that any attempts to enforce minimum wages could be unenforceable given the global and dispersed geographies and networks of online work. However, our research shows that the vast majority of demand for digital work comes from just five countries. Furthermore, a small handful of platforms mediate the vast majority of that work. These two facts demonstrate that initial barriers to regulation are not due to a dispersed geography or dispersed network of work. These topological and geographical bottlenecks in the global trade of digital work offer potential sites in which regulation can be enforced (we realise that many of the other submission to this call deal with some of the specifics of ‘how to do regulation’ and we therefore leave the details of that discussion to others).

We hope that some of these suggestions can help to bring about a fairer set of relationships between the employing class, the governing class, and the working class. Online gig work has brought income and jobs to many, but that does not mean that we should expect it to function as an unregulated labour market.

1 For some preliminary results, see the following publications:

Graham, M., Hjorth, I., Lehdonvirta, V. 2017. Digital labour and development: impacts of global digital labour platforms and the gig economy on worker livelihoods. Transfer: European Review of Labour and Research. 23(2) 135-162.

Graham, M., Lehdonvirta, V., Wood, A., Barnard, H., Hjorth, I., and Simon, D. P. 2017. The Risks and Rewards of Online Gig Work At the Global Margins. Oxford: Oxford Internet Institute.

2 See and

About Mark Graham

Mark Graham is the Professor of Internet Geography at the OII, a Faculty Fellow at the Alan Turing Institute, a Research Fellow at Green Templeton College, and an Associate in the University of Oxford’s School of Geography and the Environment. He leads a range of research projects spanning topics between digital labour, the gig economy, internet geographies, and ICTs and development.