Rollee makes it easier for platform workers to access financing. Until now, it has been difficult to obtain a loan without a steady monthly income. Lenders want a clear picture of income in order to calculate a credit score. Rollee unlocks data from platforms, banks, tax portals and other relevant data sources for lenders. This should lead to fair access to financial services for all workers, says founder Ali Hamriti in this episode of The Gig Work Podcast by the WageIndicator Foundation.

Outdated system

The idea came to Hamriti, who is Moroccan-French, when he was working at a fintech company in Paris that provides loans to small businesses and the self-employed. ‘I noticed how difficult it was for freelancers to provide financial institutions with enough certainty about their income to get a loan,’ he says. “The problem is that self-employed workers and platform workers are not employees and therefore do not receive monthly payslips. Their income varies from month to month. This makes it difficult for lenders such as banks to assess their creditworthiness, and they are often denied loans.‘

He knows that not having a payslip does not mean that they are not creditworthy. ’On the contrary: in Europe, you sometimes see that people can earn more as freelancers than in salaried employment, with more freedom,” says Hamriti. ‘Yet they are often excluded from financial services because the current systems are not organised for this. Bank statements show payments and income, but banks simply have too little context for that income. This inspired me to build a new system.’

Retrieving and sharing data

The basis of Rollee is a so-called API (Application Programming Interface): software that enables two applications to communicate with each other. ‘We link to alternative data sources about work, income and taxes,’ says Hamriti. ‘I discovered that you can get more contextual data via freelance platforms and tax portals, for example.’

The result is an open platform that allows workers to collect their platform data, income data, tax data, payslips and invoices, for example. Lenders can request and analyse all this alternative data on income and work via the API platform.

Insight into platform work

In the beginning, he focused mainly on platform workers. ‘They have more access to work via digital media, but not yet to financial services,’ he explains. ‘In emerging markets such as Africa and India, current platform workers were previously invisible to lenders because they worked informally. They found their customers through word of mouth, and payment was in cash.’

But now they are increasingly working via digital platforms such as Uber and Upwork. ‘That suddenly makes their work and income transparent,’ he says. ‘If they can share that overview with lenders, it opens up opportunities for financial services such as credit.’

100 integrations

‘Our goal is to give as many people as possible access to the financial system,’ says de Hamriti. “We started by focusing on platform workers, but it is also suitable for other workers with variable incomes. With Rollee, banks can make better-informed and fairer decisions about the creditworthiness of all types of workers. It is a solution for all kinds of financial service providers: from leasing companies and banks to insurers.”

Rollee works with all kinds of companies and agencies, such as the Tax Administration and banks. They now have more than 100 integrations with freelance platforms, tax portals, payment systems and digital wallets. No direct cooperation with platforms is required, see box.

Privacy and statistics

Rollee does not store any personal data, he explains. ‘We only facilitate the transfer of data between workers and lenders. When a worker links their bank account or platform data, the data is sent directly to the selected third party. After the transfer, we only store statistical data, such as average incomes per country.’

Rollee’s mission is to make the financial system fairer. That is why they do share statistics with financial service providers. ‘This enables us to help lenders better tailor their acceptance criteria to modern workers,’ says Hamriti. ‘For example, they can adjust their criteria if it turns out that freelancers with a slightly lower income are still financially stable.’

Hamriti is also looking to collaborate with modern lenders. ‘In the long term, we want to help workers find the best interest rates and financial terms via our fintech partners,’ he says. ‘Companies such as Revolut and Monzo can make competitive offers via our platform. This allows workers to get a fairer, better deal based on their actual income rather than general credit rules.’

International cooperation

Hamriti is currently focusing on Europe and the United States. Emerging markets such as Africa, Asia and Latin America are also very interesting for Rollee. ‘We have already helped companies in Nigeria and Kenya, where digital payments are often better developed than in Europe,’ he says. “We work with banks and lenders that operate in several countries. That makes sense, as freelancers in the platform economy often work across borders.‘

Rollee is also in dialogue with governments about regulations surrounding data sharing. Hamriti sees that France and the Netherlands, for example, are rapidly improving their digital services. ‘And there is a major European project to develop a digital identity card,’ he says. ‘This will make data exchange increasingly easier.

This offers opportunities for growth for our company. After all, financial institutions need a uniform, secure API to access different data sources. We can help with that.”

Rollee versus GigCV

What makes this conversation so interesting to me is that Rollee has both similarities and differences with my own initiative: GigCV. This is an API that allows platform workers from affiliated platforms to download an overview of their work experience in the gig economy. Via an open standard, they can easily obtain an overview of their reputation and transaction data on platforms.

The more platforms share their data, the more valuable such a CV becomes. That is why I am seeking cooperation with platform entrepreneurs. In practice, it remains difficult to convince enough parties of the strategic advantage of sharing data. Moreover, they have to integrate the GigCV API. Rollee takes a different approach. This initiative does not depend on the cooperation of the platforms, because they log in on behalf of the workers. This allows them to scale up much faster. In addition, the data is immediately usable: banks already use this information to calculate risks, but the problem was that they simply did not have access to it.

I predict that platform data and data portability will become increasingly important. Regulations such as the Digital Markets Act (DMA) and the European Platform Work Directive mean that data will no longer be shared only on request, but will be available immediately and in real time. Large platforms must give their users access to data and offer free tools for data exchange, such as APIs. This gives users more control and stimulates innovation. Article 9, paragraph 6 of the Platform Work Directive confirms this right within digital labour platforms. This is good news for the future of initiatives such as GigCV and Rollee, and therefore for platform workers.

Want to know more? Listen to the full podcast with Ali Hamriti here

Artificial intelligence (AI) is and will remain dependent on human labour. The people who do the work behind AI systems are often invisible. This carries risks of poor working conditions, low wages and inadequate protection for workers. How does this situation arise, and how can we ensure that the many invisible data workers also benefit from technological developments? For the WageIndicator Foundation‘s Gig Work Podcast, I spoke with Professor Antonio Casilli (Institut Polytechnique de Paris), author of the book Waiting for Robots, the Hired Hands of Automation.

Scooby-Doo in the world of platform work

‘Me and my team are like Scooby-Doo: we travel all over the world investigating mysteries,’ says Casilli. ‘We conduct empirical research into artificial intelligence and how it is produced. Our focus is not on the new possibilities of AI, but on the development process: who is working behind the scenes to make AI possible?

His research team is called Diplab, which stands for Digital Platform Labor. They have developed a very broad view of automation.

The myth of automation

The dream of automating work is not new: Thomas Mortimer, among others, wrote in 1801 about a machine that would be capable of making human labour ‘almost completely superfluous’.

‘Technologists and economists have been looking for ways to make labour more efficient for centuries,’ says Casilli. ‘The industrial revolution saw the emergence of the first machines, such as the steam engine and the Spinning Jenny. Every innovation came with great promises. They would save us many hours of work. But nothing could be further from the truth.’

Many predictions about automation were overstated. Studies between 2013 and 2024 claimed that robots would replace 46-47% of all jobs. Casilli: ‘Organisations such as the OECD and ILO have shown that this is not true. Even with additional crises such as climate change, geopolitical tensions and a pandemic, global unemployment has not risen. In fact, in 2025, people will be working more than ever.’

The problem lies in the methodology used by these researchers, explains the professor. ‘They take a profession and break it down into tasks. If they expect AI to replace 60% of the tasks, they conclude that the job will disappear. But that’s not how it works in practice. Often, employees simply get new tasks.’ 

See also his research Waiting for robots: the ever-elusive myth of automation and the global exploitation of digital labour.

Influence of platformisation

According to Casilli, the biggest change in recent years is not automation, but platformisation. Companies such as Uber, Amazon and Meta use huge amounts of data to connect supply and demand and organise work. They also use all this data to train AI systems. For example, they build software such as ChatGPT (the P stands for ‘Pretrained’) and the technology behind self-driving cars.

‘What is often forgotten or ignored is how many people are involved in this,‘ says the researcher. “The promise of AI is that systems can take over human cognitive tasks. But in reality, many so-called ”automatic’ processes depend on human labour. The people who do this work are often invisible and poorly paid.’ This is not a recent phenomenon: Google, for example, has had its own platform, Raterhub, since 2007, where data workers verify search results and thus improve the search engine’s algorithms. Amazon Mechanical Turk, the platform used by Amazon and also available to external customers, makes a clear reference to the myth surrounding AI and its dependence on human labour. The Mechanical Turk after which the platform is named is the ‘chess robot’ invented in 1770, which travelled the world for 84 years as an example of automation. Later, it turned out that there was a person (often described as disabled or underage, in any case not a chess master) inside the machine and there was little automation involved.

Automation does not lead to less work, but to different, degraded form of work. ‘Big tech companies prefer not to talk about that. It undermines the narrative that AI is truly intelligent. In reality, people are working more than ever, but sometimes under worse conditions than before.

Check here the full interview with Antonio Casilli on YouTube

Who are these data workers?

Data workers collect, organise and improve data. Without them, AI would not work. Take image recognition, for example: AI learns what a cat is by analysing millions of images of cats. ‘People have to label those images first. It seems like simple work, but it’s a skill in itself. Yet these data workers often receive remuneration that is not commensurate with their efforts,’ says Casilli. ‘In countries such as Kenya, the monthly wage for these data workers is around $400. That’s not enough to make ends meet.’

Casilli emphasises that this is not a temporary phase. ‘Data work will remain necessary as long as we continue to develop AI,’ he says. ‘We have to constantly train the systems, adapt them to new customer requirements and check them for errors. World Bank or Oxford estimates point towards a ballpark figure of 150 million such workers worldwide, and that number is only growing. That’s another reason why it’s important to take a critical look at their working conditions.’

You are a data worker too

In his book Waiting for Robots, Antonio Casilli mentions a group of digital workers who are often overlooked: social network labourers. This basically includes everyone with a smartphone. Through our daily online activities, we train the AI of large tech companies. We teach AI what a traffic light is by filling in ReCaptchas. When we like social media posts, we teach systems which images are attractive. So we provide value to AI systems, but we are usually not paid for it. We are both users and producers of data. This raises an interesting question: is this work or not?

Casilli ziet dat deze vorm van arbeid bestaande machtsstructuren en scheve arbeidsverhoudingen versterkt. Hij en zijn team hebben samengewerkt met beleidsmakers en vakbonden om dit aan het licht te brengen. “Tech-ingenieurs bij bedrijven zoals Google verdienen hoge salarissen, terwijl datawerkers in India, Venezuela en Madagascar onderbetaald worden. Dit volgt koloniale patronen. India voert datawerk uit voor Engelstalige landen, terwijl Franse bedrijven werk uitbesteden aan Franstalige landen in Afrika.”

What can we do?

What can we do about this? He describes this in the last chapter of his book ‘What is to be done?’, a tongue-in-cheek quote from Vladimir Lenin. According to Casilli, a systemic approach is needed to improve the conditions of all data workers worldwide. ‘A solution for a specific group will not work in the end. We need to look for a universal strategy.’

He distinguishes between three types of solutions: regulation, collective platform initiatives and a global redistribution system:

1. Regulation: Spain, for example, has introduced the Riders’ Law and the European Union is working on guidelines for platform workers. “These are steps in the right direction, but this type of regulation needs to be applied more broadly. After all, tech companies operate globally.”
2. Platform cooperatives: Workers can set up their own platforms in which they have a say in wages and working conditions. ‘This is already happening on a small scale, but deserves more attention.’
3. Redistribution: Large tech companies can be taxed and the proceeds used for a universal basic income for data workers. Importantly, Casilli states that this UBI is neither connected to a “robot tax” (as he doesn’t see robots replacing workers) nor it is intended to replace welfare assistance (as it should be paid regardless of other social benefits). ‘This will ensure greater fairness.’

By combining these three strategies, the professor hopes that we can create a fairer and more sustainable system. ‘Tech companies must take responsibility for all their workers, including the invisible ones who manufacture their data,’ says Casilli. ‘I am concerned about this situation: wages are far below the minimum and even basic health and safety rules are not always observed.’

Casilli believes that organisations such as the WageIndicator Foundation and the Fairwork project are making an important contribution. ‘These organisations set standards for fair wages and working conditions, and these are desperately needed.’

Enforcement, collective action and user responsibility

After several interviews on this topic, I personally believe that, besides the solutions that Casilli provides, it is also important to enforce existing regulations. In countries where there are many underpaid data workers, there is a lack of supervision. This is partly due to strong lobbying by tech companies. That is why it is so important for workers to take collective action, for example via trade unions. These are underrepresented, although a number of interesting grassroots initiatives have emerged.

I also believe that (large) users of AI solutions must take responsibility. There are many discussions about responsible AI use. But I can no longer take a discussion about responsible AI seriously if it does not take into account the hidden workers.

Why this is important

Casilli and his team are uncovering an important mystery: AI is not a magical ‘black box’. In reality, millions of people work behind the scenes on these so-called ‘intelligent systems’. AI is presented as completely autonomous, and the extensive manual labour involved is often forgotten or ignored. This has serious consequences for the working conditions of these data workers.

If we really want to use AI responsibly, we must also consider the people behind the technology. I try to raise awareness of this issue and highlight it wherever possible. That is why I spoke earlier with Claartje ter Hoeven about Ghostwork: the invisible world of work behind AI. I will soon be speaking to the Data Labeler Association in Kenya to gain more insight into the conditions and problems faced by workers in Kenya. After all, we can only really get started with responsible AI if we understand how AI is created.

Want to know more? Listen to the full podcast with Antonio Casilli here. Check all podcast episodes via this link.

Claartje ter Hoeven (Utrecht University) and her research team reveal the hidden world of European data or ghostworkers. They are often highly educated, seeking flexibility, but their working conditions are usually poor.What drives them? What impact do they have on algorithm development, and vice versa?

They are invisible, ubiquitous and indispensable for the development of artificial intelligence (AI): ‘ghost workers’. Millions of people worldwide annotate, check and translate texts and images so that AI can understand and process the information. Who are these people and what drives them? What about their well-being? And what impact do their poor working conditions have on the development of AI? To learn more about this, I sat down with researcher Claartje ter Hoeven of Utrecht University for The Gig Work Podcast from the WageIndicator Foundation. She is conducting research into this phenomenon with a European Research Council (ERC) grant.https://open.spotify.com/embed/episode/5km4FhBvCgU2cW29EmlDpZ?si=b165a66c63e44aa9&utm_source=oembed

Annotating, checking and correcting so AI can learn

Ter Hoeven and her team are researching the working conditions and well-being of so-called ‘ghostworkers’ finding work through online platforms. They build on the work of Mary Gray and Siddharth Suri, ‘Ghost Work: How to Stop Silicon Valley from Building a New Global Underclass’. While Gray and Suri researched working conditions of ghostworkers in the United States and Asia, Ter Hoeven and her research team focus on Europe and examine the relationship between working conditions and worker well-being. It is a five-year study and the team is now about halfway through.

‘We are investigating the hidden labour behind AI,’ she says. ‘Ghostwork is a catchy term, but in science these days we prefer to call it ‘datawork’. Dataworkers make texts and pictures readable to AI in order for the machine to learn from them. For example, they indicate what a lamppost or a bicycle is, so that the algorithm of a self-driving car learns to recognise these objects. They also check and correct the output of AI models and algorithms.’

Low-paid

Data workers often do this from home through online platforms such as Amazon Mechanical Turk, Clickworker or Microworkers. They usually get paid per mini-task or ‘microtask’: an annotation, check or translation. The pay is often low and data workers have to search for the microtasks themselves on various platforms. Searching takes time, but they do not get paid for that. They often earn less than the minimum wage.

There are also companies that employ data workers, so-called Business Process Outsourcing organisations (BPOs). They work in physical office locations, are often paid by the hour and are supplied with the tasks. Although they have no unpaid search time, their pay too is often below the poverty line. 

We still know quite little about these data workers. and many big tech companies prefer not to talk about the contribution of humans in the development of AI, because it does not fit the narrative that AI is ‘self-learning’. This is not only a shame, but also detrimental to the development of responsible AI. It is therefore beneficial that Ter Hoeven and her team are researching datawork through platforms in Europe.

Highly educated with a distance to the labour market

Ter Hoeven used various research methods to discover how working conditions of dataworkers affect their well-being. It started with a short survey. This they distributed as a microtask on various platforms in Europe to get as many responses as possible. In the end, more than 5,000 people completed the survey.

‘A striking result was that many data workers are highly educated,’ says the researcher. ‘They often have a certain distance to the labour market. Think migrants or people who combine work with caring responsibilities for family members.’

Four drivers

Among data workers working via platforms, Ter Hoeven distinguishes four groups based on their motives , based on her survey of over 5,000 respondents: explorers, enthusiasts, supplementers and dependents:

Rather an algorithm than a human as boss

Ter Hoeven and her team conducted 137 face-to-face interviews with data workers from Europe. She discovered all kinds of motivations. ‘Those who work through platforms are dependent on algorithms,’ she says. ‘Algorithms determine whether you get a task and sometimes what you earn with it. There are all kinds of drawbacks to this. It is often unclear how algorithms make decisions, and platforms make it almost impossible to complain or discuss decisions. Yet many data workers told me during interviews that they would rather work for an algorithm, than a human manager.’

For example, she spoke to a migrant in Germany who had bad experiences with nasty bosses. Thanks to microwork, he could at least work from home and decide his own working hours. Another interviewee was a neuroscientist with a medical condition, which meant she needed more time to get up in the morning. She had to stop working at the university as a result. Thanks to datawork, she can still earn money. Ter Hoeven: ‘So our research not only says something about microwork, but also raises questions about the way we organise more traditional work.’

Need for colleagues and appreciation

The researchers present their findings not only on paper, but also in a documentary. They invited six European dataworkers to participate in video recordings. ‘We asked them questions and brought them together for panel discussions,’ says the researcher. ‘This cinematic research offered very interesting insights.’

Trailer of the film ‘Ghost Workers’ by Lisette Olsthoorn in collaboration with Erasmus University Rotterdam. This film was funded by the Erasmus Initiative Societal Impact of AI and by the European Research Council (ERC) as part of the European Union’s Horizon 2020 research and innovation programme (grant agreement no. 101003134).

While most dataworkers indicated during the interviews that they generally did not miss colleagues, it became clear during the recordings that they actually needed to do so. ‘Suddenly they could complain, brainstorm and share experiences with like-minded people,’ says Ter Hoeven. ‘They had simply never had a dataworker colleague before. Furthermore, I saw their self-confidence grow during the filming now that they were the ones in the spotlight. Dataworkers may need contact and appreciation more than they themselves sometimes think.’

Data quality

Datawork raises a lot of questions. These are not only about the well-being of employees, but also about data quality. AI has an increasing impact on our daily lives. Ter Hoeven: ‘An example: some dataworkers annotate medical procedures. For example, they have to indicate whether a doctor’s hand is shaking during an operation. But these people usually have no medical background. So how reliable is that data?’

What consequences does it have if AI learns from people without sufficient expert knowledge and information about the context of the data to be ‘translated’? To improve quality, it may make sense to better match workers’ skills to tasks and provide better guidance. This is only possible if you invest more in data workers.

More transparency

In her book The Tech Coup, Marietje Schaake discusses how tech companies are conducting real-time experiments with user data to optimise their platforms, often without users’ knowledge. This can have serious consequences for privacy, democracy and personal freedom. That is why she calls for stricter regulation and more transparency.

The same applies here. In my opinion, organisations should be more transparent about the contributions of data workers and their potential risks. I therefore hope that European legislation like the Corporate Sustainability Due Diligence Directive (CSDDD) will also apply to how companies develop their AI. After all, this smart technology is increasingly affecting all kinds of processes. In short, the conversation with Ter Hoeven again leads me to many new questions. I will be seeking answers to those in this podcast in the coming months.

Want to know more? Listen to the podcast episode on Ghostwork here. This blog was also published on Gigpedia.org.

The GHOSTWORK-project received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme Grant agreement No. 101003134

The growing impact of digital technology, generative AI and algorithmic management on work is an increasingly widely explored topic. Last week during the two-day conference ‘Future of Work: Reclaiming the value of work in the digital economy’, researchers from across Europe gathered to present their work and exchange views.

It was organised by both the research group of the ERC project ‘Respect Me “ and the European Trade Union Institute” (ETUI). In this blog, I look back at the discussions and my own contribution on an upcoming paper on the Living Tariff methodology.

From focus on platform to focus on impact of technology on work

The shortcoming with many discussions on the platform economy and platform work is that it is assumed to be an isolated phenomenon. A self-contained silo. I wrote earlier that this is of course big nonsense: you cannot talk about platform work while ignoring the rest of the labour market. So that has been my biggest criticism for ages of how unions relate to this development. They often ignore what the (often poor) conditions of workers are in the same market where a platform does not provide intermediation and that the alternative for the worker is not a well-paid job with a lot of security. For me, this was most visible in the case where FNV sued domestic cleaning platform Helpling. When you have a discussion about domestic cleaning platforms without acknowledging that it takes place in a sector where informal work dominates in most countries, you deliberately miss an opportunity to do something for this group of workers.

It is therefore first important to look at what is really new. I did this in 2021 together with Jeroen Meijerink in our research ‘Online labour platforms versus temp agencies: what are the differences?’.

Also at the conference in Leuven, the call to look at what is really new was highlighted by Uma Rani of the International Labour Organisation (ILO), among others. She brought an interesting overview of a historical perspective of the use of technology in the context of work and also showed that sometimes the technology itself does not change, but the way it is applied does. This was also discussed by her ILO colleague Annarosa Pesole.

The platform economy is often seen as the ‘nursery’ or ‘sandbox’ of the labour market, where these technologies like algorithmic management are developed and tested (which raises important ethical issues) on workers. To then be applied to the wider labour market. Something that, by the way, is in line with the current development of legislation like the Platformwork Directive, which everyone understands that the passages on algorithmic management should apply not only to a specific group of platform workers, but to all workers.

And so that was the focus of this conference: the impact of digitalisation on the way we work, organise, allocate, control and (to some extent) evaluate work. Where the predominance was on the worker’s perspective, which is not surprising with the ETUI as co-organiser. There was much discussion on how to secure workers’ voices in the governance of this technology, how to make processes more transparent (and verifiable) and how to increase knowledge among workers and unions.

Insufficient use of existing regulation

There is a lot of focus on new regulation in the platform economy. In addition to the previously mentioned Platformwork Directive, a platform law to be implemented nationally in all European member states over the next 18 months, we also have the Digital Service Act (DSA), Digital Markets Act (DSA), Platform to Business Directive (P2B) and others. The focus here is on creating more transparency and accountability with the aim of improving the balance of power between the different stakeholders (platform, worker, client and society).

💡

Read a report on a meeting on workable regulation I organised earlier and a review of the DAC7 platform law

With all the attention around new regulations, you would almost forget that there is already a lot of regulation that working platforms already have to comply with. During the session ‘Representing workers rights in the platform economy’, among others, this was emphasised several times. This session was mainly about the upcoming Platformwork Directive, but also explicitly about GDPR. Since many platform workers are not employees, they are also not protected under employment law, making AVG for rights relating to data suddenly very interesting. (note: I am not a lawyer) For example, María Luz Rodríguez Fernández presented the GDPoweR project, which , according to its own website, ‘explores what worker data is collected by platform companies and how this affects workers, what strategies are used by social partners to negotiate and implement collective agreements and how the implementation of such agreements can be monitored and enforced. A central method used is the recovery of worker data through GDPR requests and the joint analysis of this data by workers and researchers.’ The slides below summarise the initial findings.

And although GDPR is an individual and not a collective right, the GDPoweR project shows that there is no reason not to use it for collective activities. Something I also described earlier in the blog and podcast ‘knowledge is power, even in the platform economy’ following an interview with James Farrar, founder of the Worker Info Exchange. All the cases his modern union brought against Uber, among others, were won on the basis of existing regulations. Something policymakers and unions should consider a lot more.

And of course, new regulations also provide opportunities. Personally, as a non-lawyer, I expect little impact from the employee part of the upcoming Platformwork Directive. It would be strange if your rights as workers depend on how you get your work handed to you. I see this part more as a stepping stone to better rights for an entire sector, although with the current political climate, the words ambition and political unity are rare. I am particularly curious to see how the excerpts dealing with the impact of technology on work are fleshed out. On the one hand, because certain things are contract-neutral (this sounds boring, but is revolutionary, as all certainties and obligations around security, among other things, are linked to being an employee) and, on the other hand, because there is a mention of ‘digital community channels’ that should break or reduce the isolation of individual workers (and thus an important instrument of platform power).

Finally, Annarosa Pesole (ILO), among others, warned on the finding that more and more parties are using ‘off the shelf’ technology and thus have little knowledge, insight or influence on the technology being deployed. A danger for the worker, but also for the one deploying the technology, as from the AI Act, among others, there is also a responsibility on the user of this technology.

Caught in the worker paradox and union dilemma

Many discussions on platforms and labour have revolved around whether the worker is a freelancer or an employee. In itself a logical thought from a Global North perspective, because there, being an employee is the dominant model and many securities and obligations are linked to this contract model. In addition to social securities, I am talking about health and safety obligations, but also a role in social dialogue and representation. If you are not an employee, you are virtually outside the scope of this.

Although I can understand the focus, I do question to what extent the discussion is not caught in an employee paradox, a term that also came up during this congress. This is because in many countries, the worker model is not the dominant model and because this focus ignores a large group of workers who also need protection. Think of freelancers, but also of the informal market.

The employee paradox also lacks room for nuance. Employing means, for many platforms, switching themselves to the temp model (with limited rights) or using existing temp agencies or subcontractors. Randstad’s CEO a few years ago, for instance, named the ‘gig staffing market’ as one of the big opportunities for the staffing company. One of the conclusions from the aforementioned paper I wrote with Jeroen Meijerink is that many types of platform work can be perfectly well organised via a temping model, with the important note that the terms ‘security’ and ‘fairness’ under these models are not of the level as they are presented to us in the public debate. And that it becomes very difficult for employment agencies to comply with the legal duty of care. Something that also came up in the contribution by Silvia Borelli (Università di Ferrara).

The conference presented several examples of how ‘worker representation’ in digital technology and AI can be organised.

A subject that will only become more important in the coming years. Here, unions face a dilemma: do they stick to the existing model, or do they initiate a change to a more broad focus on ‘the workers’ in the broadest sense of the word and build up expertise to also stand up for workers in the digital domain. Visitors to the conference agreed that unions currently give far too little priority to this, something also acknowledged in the report ‘Collective bargaining practices on AI and algorithmic management in European services sectors.’

I challenge unions to then instantly look a bit wider and examine whether they may need to target not only employers, but also mediators and the creators of technology. In the digital domain, it is more necessary than ever to fight for a better balance of power, which will ultimately also lead to better products and innovations. Trade unions could and (in my opinion) should play an important role in this.

What is a decent income? Towards a more global view

A topic that doesn’t come up much at these kinds of conferences is the topic of income. Or rather, decent income. When it comes to income, the topic is usually the intransparency of payments in on-demand platforms like delivery and taxi. An important topic, as these platforms have only made the structures of how a tariff is built more complex over the years, so the worker often does not know what the earnings are before accepting a gig and the gap between what the worker receives and what the customer pays has grown. In a market where transparency was promised, information asymmetry has only grown to the disadvantage of workers.

Besides the issue of transparency (it is questionable to what extent dynamic tariffs for work are at all desirable), there is no debate on the level of compensation a worker should receive for the service provided. A topic that is ranked number one when you look at what platform workers are taking to the streets worldwide for, according to earlier research by Leeds University.

💡

Check also the blog and podcast I produced “How and why does the platform worker protest? Scientists provide overview and insights”, based on an interview with researchers Vera Trappmann and Simon Joyce.

That was therefore the topic of my contribution: ‘Facilitating workers and policy makers in the gig economy making better informed decisions: the case study of the Living Tariff’. The basis of this concept comes from the Living Wage methodology, where a minimum wage for a worker is calculated based on the costs a worker has to incur to live a decent life in country X in region X. A wage floor based on actual costs. Where the Living Wage is of value to employees, someone who is not an employee can do little with it. Whereas certain costs and risks of work in the case of an employee are borne by the employer, a non-employee (freelancers and informal market) has to bear these costs and risks themselves. These should be included in the hourly rate to arrive at a fair minimum threshold.

💡

Last year I interviewed Valeria Pulignano, who is in the lead of the Respect Me project, for The Gig Work Podcast by the WageIndicator on her research on unpaid labour. Check the podcast and blog.

With the Living Tariff methodology, these costs and risks are taken into account, allowing the freelancer to know what they need to earn per hour to eventually arrive at a Living Wage, after deducting risks and costs. This methodology fills an important gap in the calculation of minimum income (wages and tariffs).

A next step is for this calculation to be included in consultations with various stakeholders (clients/platforms, workers/trade unions and policymakers/politics) in a discussion on fair compensation for workers. If you want to know more about this, check out the Living Tariff page, the slides I used for my presentation, or the blog and podcast I produced on this topic.

To conclude

It was a delight to attend this conference, hear the insights of researchers and then have great conversations and discussions about this. What I would like to see is a sequel to a conference like this, but with a broader stakeholder perspective. At the opening, someone in the audience asked ‘are we also going to talk about opportunities?’. A fair question, but in order to arrive at exploiting opportunities (and reducing risks), an insight from different stakeholders is needed. Insights from policymakers and platforms themselves, for example. Now platforms have an image problem due to the sometimes questionable practices of some big players, but the market of platform companies is in fact mainly an SME market. According to the ‘Monitor online platforms 2023’ (CBS, 2023), 64.2 per cent of the 1,600 Dutch platform companies have 2 or fewer employees. Only 5 per cent of these companies have more than 100 employees. Those opportunities may not come directly from the big players, but there are plenty of small(er) platform companies that are a lot more approachable where positive change could come from. Whereas now there is sometimes a bit too much emphasis on struggle, which is explainable because of the union and collective action perspective, it would also not be wrong to apply experiments on a smaller scale with parties that do want to collaborate. And to take the lessons from there and scale them up. This also includes a discussion on the more fundamental questions about the value of work. The title of the conference was ‘reclaiming the value of work in the digital economy’. This is a step that, as far as I am concerned, is often skipped, because it also involves a piece of self-reflection on parts of the (vulnerable) labour market that are not included in the debate.

I am convinced that there are still plenty of opportunities, only it is up to those involved to take responsibility and take up, test, validate and scale up these opportunities. I am happy to contribute to that.

Since 2023, platforms have had to share supplier data with tax authorities. What are the implications of this legislation? How can we achieve better regulation? Researchers Ahmed Darwish and Martijn Arets share lessons.

Since 2023, platform operators have been obliged to collect and share vendor data with the tax authorities. By imposing reporting obligations, this so-called Directive on Administrative Cooperation (DAC7) aimed to ensure more transparency regarding the income earned through digital platforms and aid tax authorities in countering fraud.

Does the legislation do what the Council of the European Union intended it to do? How can it be improved? As part of the Platform Economy Research Group at The Hague University of Applied Sciences, we, Ahmed Darwish and Martijn Arets, took a closer look at the development and effects of the legislation. While, at first glance, the legislation seems logical and practical, there are a number of questions and concerns raised by entrepreneurs. What can policymakers learn from this?

Five lessons for regulation in the platform economy:

1. Provide clear guidelines
2. Offer support
3. Consult stakeholders in advance
4. Implement regulations based on company size
5. Make sure government institutions are ready for the new legislation

What is DAC7?

DAC7 is an EU directive aimed at improving tax transparency on income via digital platforms. By 1 January 2023, all European Union member states were required to incorporate this directive into their domestic legal frameworks (DAC7, Article 2.1). Since then, digital platforms have been obliged to identify, verify, and report the income of sellers that utilize their platforms.

The legislation applies specifically to platforms that bring together the supply and demand of products and services (DAC7, Article 1.8). They must share the details of suppliers on their platform with the tax authorities. These can be sellers of new goods (such as Bol, Amazon) and second-hand goods (such as eBay, Vinted). Platforms facilitating for booking hotels, holiday homes, and hospitality (such as Airbnb, Booking) are also covered by this legislation. It may also cover service providers, such as taxi drivers (such as Uber, Lyft), babysitters (such as Charly Cares) or food delivery (such as Thuisbezorgd, UberEats).

Any platform where sellers can offer certain services or goods to other users may be subject to the directive. As such, webshops that only sell their own products are not covered by the new legislation. It does not matter whether the platform is registered in the EU, as long as the sellers operate within the EU. The tax authorities of Member States also exchange the data gathered among themselves.

Digital platforms have access to information on markets that were previously highly fragmented. The EU is thus cleverly leveraging the immense power and information held by digital platforms. Indeed, before platforms brought supply and demand together digitally, people still traded products and services in ways that were less visible to tax authorities. Although they were, in certain cases, required to declare their income, many did not. In those cases, the tax authorities had no information. Platforms can, therefore, help governments obtain information about previously invisible highly fragmented markets (European Commission, Rationale).

Why this case study?

The idea of digital platforms sharing data with public authorities seems simple and logical. In practice, however, implementation proves more complicated. There are many differences between national governments and digital platforms. Although the platform economy consists mainly of small companies, legislation seems tailored at larger companies that have more resources to store and process data. Indeed, much of the platform market is an SME market. According to the ‘Monitor online platforms 2023’ (CBS, 2023), 64.2 per cent of the 1,600 Dutch platform companies have two or fewer employees. Only 5 per cent of these companies have more than 100 employees. 

We did an exploratory case study of DAC7, focusing on the Dutch context. We interviewed all sorts of stakeholders and experts, including platform operators, representatives, experts from the Tax Administration, and academics. We analyzed the directive, its implementation, and its impact on the platform economy. As a result, we produced five recommendations for future policy initiatives around the platform economy.

Benefits

The main benefits of DAC7 are:

1. Less fraud through data sharing

By sharing data, platforms and governments can work together to prevent tax fraud. DAC7 also promotes cross-border tax compliance, as some larger digital platforms operate in multiple countries. Data sharing leads to higher effectiveness and efficiency in monitoring revenues (European Commission, Rationale). By making data related to income transparent to tax authorities, they can better assess whether someone is a ‘seller’ for tax purposes or not. DAC7 applies only within the EU but promotes international cooperation with countries outside the EU, thanks to its alignment with OECD guidelines. It is important that platforms implement DAC7 properly and inform their users about the process. Since users know that their income data will be collected and shared with tax authorities, they will be more aware that they may have to pay their taxes. DAC7 thus helps not only detect fraud but also prevent it. However, success depends on the tax authorities’ ability to collect and reliably process the data.

2. Better cooperation between governments and platform operators

Technological developments are happening at a rapid pace and, as a result, policy often lags behind. Transparent cooperation between governments and the technology industry can help reduce the information gap and promote trust. In that vein, DAC7 is a step in the right direction. With their valuable data, platforms can help governments with enforcement and regulation, as long as the processes are workable and the impact on business processes and data collection proportional.

3. Improving KYC processes

DAC7 standardizes Know Your Customer (KYC) processes, since platforms must collect, verify, and report user data (such as name, address, and tax numbers) to national tax authorities in a uniform manner. This standardization reduces compliance costs, increases predictability for users, and can serve as a basis for further regulation (Vidal, 2024).

Disadvantages

The new directive has a lot of potential benefits, but it is far from perfect. Our analysis shows the main drawbacks:

1.A too broad definition, a tight timeline and a lack of guidance for platforms
The timeline of policy decisions and implementation was quite tight. For example, the Dutch Senate took a final decision on DAC7 on the 20^th of December 2022 and companies had to have already implemented it less than two weeks later (1 January 2023). Adding to the confusion, DAC7 had to be implemented at a different deadline for each member state (Trolley, 2024). Although all member states had to be ready for DAC7 at the same time, some governments took more time than others (European Commission, 2023). This was especially difficult and costly for platforms operating in multiple member states.

Many platforms found the directive’s provisions unclear, leading to uncertainty. The directive used broad and vague wording. Moreover, the definition of what a platform meant in this legislation was abstract and unclear. Some operators were forced to incur excessive compliance costs, to avoid being fined or penalized. This uncertainty and lack of clarity also affected platform suppliers, who suddenly had to share more personal data than they initially collected and did not understand why it was necessary. 

Due to the high-pressure timeframe, the Tax Administration did give platforms much time and leniency, a so-called ‘soft landing’. The tax authorities tried to keep platforms well informed and provide them with support. With the implementation phase complete, the Tax Administration is slowly moving to stricter enforcement.

2. High compliance costs and inequality
DAC7 imposes a heavy administrative burden on platform operators, a group that mainly consists of small and medium-sized enterprises (SMEs). The directive seems to be tailored for larger platforms, but the obligations also apply to smaller players. For them, the costs of data collection, storage, and reporting are high. The labor and financial costs associated with compliance hinder growth and innovation. Whereas the OECD directive included an exception for small platforms, companies with up to €1 million in turnover, the EU decided not to include this exception in DAC7 (OECD MRDP, 2020).

3. Potentially discouraging participation
Stricter reporting requirements make digital platforms less attractive to sellers. Users become reluctant, as they cannot be certain about what the Tax Administration will do with the reported data and how it will impact their tax calculation. Some users even decided to work outside the platform out of uncertainty and fear. Research shows that knowledge of DAC7 regulations lowers the willingness to work utilizing platforms, noting a chilling effect. Users get the feeling that the government is watching them closely. This leads to distrust and fear (Mol & Molho, 2024).

4. Additional collection of privacy-sensitive data
To comply with DAC7, some platforms have to request and store additional personal data from their sellers. This is most relevant for platforms that specialize in facilitating the sales of second-hand goods. They store data that is not essential to the transaction, raising several significant privacy risks.

5. Limited capacity of Tax Administrations
Finally, platform operators doubt the Tax Administration’s ability to deal with the huge amount of data. The Tax Administration indicated that the DAC7 data is a new source of information for their inspectors. The Tax Administration aims to determine how reliable the data is, before inspectors use it for enforcement. In the future, if the Tax Administration has reason to think that people are not complying with legislation, inspectors will choose other enforcement tools, such as regular inspection. In doing so, the inspector can use the DAC7 reported data to check whether the tax information reported is correct. While the Tax Administration is optimistic about their ability to analyze large datasets, platforms are more critical. Until the Tax Administration is able to deliver on the goals set out by DAC7, compliance will feel like a disproportionate burden.

Conclusion – Five lessons for better guidelines:

1. Provide clear guidelines and a realistic timeline.
The first lesson is that legislation should be accompanied by clear, consistent, and accessible guidelines. This will make it easier for small- to medium-sized entrepreneurs to comply with regulations. Many problems that platforms encountered with DAC7 stemmed from confusion and a lack of legal certainty. This confusion concerned both the reporting requirement and the tax implications. A realistic timeline for implementation also proved important.

2. Offer support
The second lesson is that start-ups and other small businesses need support to be able to comply with complex legislation. These enterprises usually do not have a legal or compliance department. As a result, they struggle to understand how legislation can affect their business and how they can comply with regulations. Targeted support can help these businesses develop strategies for compliance. The government forced platforms to invest their own time and resources to obtain legal advice, hire compliance experts, and outsource data transmission. This discouraged their investors and hindered growth. Authorities should therefore develop targeted support for SMEs, such as:

1. Simple checklists, timelines, and step-by-step instructions, such as the information page on DAC7 set up by the Dutch Tax Administration.
2. Targeted compliance support programmes and advisory services. For example, the Dutch Tax Administration held informative presentations on DAC7 via intermediary days and umbrella organisations.
3. Access to digital tools and solutions.

It is crucial to actively inform SMEs about this support. The government should also launch campaigns to inform platform users. After all, support in communication or a public information campaign can increase acceptance and awareness.

3. Consult stakeholders in advance
It is wise to engage with platform operators about new regulations as early as possible. That way, policymakers can develop legislation and guidelines that are well aligned with practice. In the case of DAC7, stakeholders did not always feel heard. The online consultation received hardly any reactions beforehand, as SME platforms were unaware of the consultation or did not have DAC7 as a priority. Large platforms felt that little was done with their feedback. An earlier consultation at the national level could add value here.

Collaboration improves the effectiveness of legislation, especially with complex business models and in unpredictable situations. It also gives policymakers insight into the processes and variables within a seemingly homogenous group of companies. Furthermore, it increases trust and transparent communication between entrepreneurs and policymakers.

4. Implement regulations based on company size
The fourth lesson is that the regulatory burden should not be disproportionate. While you can expect companies to perform KYC reviews, excessive financial and administrative burdens can discourage new investments and hinder growth. One recommendation is to apply rules proportionally based on the size of the platform. While the provisions were more suitable for larger platforms, medium and small platforms fell within the rules. 

A better approach would be a categorised structure, where reporting obligations are tailored to a company’s size or turnover. The cost and additional workload of complying with the regulations should be proportional to the size of the business. Thus, small business owners are not overburdened or unfairly disadvantaged compared to large, established competitors.

In addition, policymakers should be careful when determining the scope of legislation. For DAC7, many question whether the burden imposed outweighs the potential disadvantages.

5. Make sure government institutions are ready for the new legislation
The fifth and final lesson is that government institutions such as the Tax Administration must be ready for the new regulations in advance. The technology and team must be capable of handling and monitoring implementation. For example, platforms experienced a high barrier due to the complex reporting system. Platform companies must therefore develop a gateway in their own software that communicates directly with the Tax Administration’s systems or use intermediaries who charge between €1,300 and €1,800 a time for this (NRC, 2025). The Dutch Tax Administration now recognizes this and is working on a simpler system. Platform operators feel as though they have wasted their time and money on adapting to methods that will have to be adjusted later on. With small-scale tests being rolled out in advance, this situation could have been avoided.

Sources and thanks
We thank the experts who contributed to this research through interviews and feedback on this article. In particular, Chantal Malfeyt (Marktplaats), Joey van Angeren (Vrije Universiteit Amsterdam), Pepijn Niesten (Booka Rentals), Jasper van Schijndel (PwC), Dion Egiyan (PwC) and Juan Manuel Vázquez (UvA). We also thank Merel Hillen of The Hague University for coordination and organisation.

Sources:

1. Anu Bradford ‘The False Choice Between Digital Regulation and Innovation’ (7 March 2024) 118 Northwestern University Law Review 2.
2. Alexander Kobakhidze ‘Challenges for Digital Platform Operators as DAC7 Looms’ (Bloomberg Tax, 2022) <https://news.bloombergtax.com/daily-tax-report-international/challenges-for-digital-platform-operators-as-dac7-looms>.
3. Anthony Mark A. Gutierrez, ‘Decode: Taxing the Digital Economy’ (2021) 65 Ateneo LJ 1028.
4. Apostolos Thomadakis and Vieri Ceriani ‘EU Corporate Taxation in the Digital Era – The Road to a New International Order’ (2023) CEPS-ECMI Task Force Report, Centre for European Policy Studies.
5. Andrew Victor Kofman ‘Small Businesses, Big Burdens: Harnessing RegTech to Help SMEs Meet Compliance Requirements’ (Utica University, May 2023) <https://www.proquest.com/openview/9362ac6fced977c55c22013bbd9f28d3/1?pq-origsite=gscholar&cbl=18750&diss=y>. 
6. Bicchieri, C., Fatas, E., Aldama, A., Casas, A., Deshpande, I., Lauro, M., Parilli, C., Spohn, M., Pereira, P., & Wen, R. (2021). In science we (should) trust: Expectations and compliance across nine countries during the COVID-19 pandemic. PLoS One, 16(6), e0252892. <https://doi.org/10.1371/journal.pone.0252892>.
7. Boer, K., & Gribnau, H. (2018). Legal aspects of behaviourally informed strategies to enhance tax compliance. In tax nudges (pp. 53–64). <https://doi.org/10.2139/ssrn.3295964>.
8. Batrancea, L., Nichita, A., Olsen, J., Kogler, C., Kirchler, E., Hoelzl, E., Weiss, A., Torgler, B., Fooken, J., Fuller, J., Schafner, M., Banuri, S., Hassanein, M., Alarcón-García, G., Aldemir, C., Apostol, O., Bank Weinberg, D., Batrancea, I., Belianin, A., & Zukauskas, S. (2019). Trust and power as determinants of tax compliance across 44 nations. Journal of Economic Psychology. <https://doi.org/10. 1016/j.joep.2019.102191>.
9. Catherine Mpolokeng Sephapo, Michael Colin Cant, Johannes Arnoldus Wiid ‘Tax Compliance: An Administrative Burden for Small Business Enterprises (SME’S)’ 13 Corporate Ownership & Control 4.
10. Daan Arends, Jesse Peeters, Wouter Kolkman ‘DAC7 adds reporting obligations for EU digital platform operators’ (International Tax Review, 2022) <https://www.internationaltaxreview.com/article/2a7cstq7ub837k2cbbm68/dac7-adds-reporting-obligations-for-eu-digital-platform-operators>.
11. Daniel Bunn ‘A Summary of Criticisms of the EU Digital Tax’ (Tax Foundation – Europe, 2018) <https://taxfoundation.org/research/all/eu/eu-digital-tax-criticisms/>.
12. ‘DAC7: Rationale’ (European Commission) <https://taxation-customs.ec.europa.eu/taxation/tax-transparency-cooperation/administrative-co-operation-and-mutual-assistance/dac7_en#rationale>.
13. ‘DAC7: Lessons Learned – The Dutch Tax Authority shows leniency, but compliance is expected.’ (RSM) < https://www.rsm.global/netherlands/en/insights/dac7-lessons-learned-dutch-tax-authority-shows-leniency-compliance-expected>.
14. Daisy Ogembo and Vili Lehdonvirta ‘Taxing Earnings from the Platform Economy: An EU Digital Single Window for Income Data?’ (2020) British Tax review (1): 82-101.
15. Dejan Ravšelj, Polonca Kovač, Aleksander Aristovnik ‘Tax-Related Burden on SMEs in the European Union: The Case of Slovenia’ 10 Mediterranean Journal of Social Sciences 2.
16. ‘EU Adopts Tax Transparency Rules for Digital platformen (DAC7)’ Ernst and Young <https://go.ey.com/39ukiw8>.
17. European Commission. (2016). A European Agenda for the collaborative economy. In The European Economic and Social Committee.
18. European Commission. (2020). Tax fraud and evasion – better cooperation between national tax authorities on exchanging information
19. Gerhard Huemer ‘Productivity growth in Europe is hampered by high administrative burdens and lack of finance’ (SMEUnited, 15 May 2024) <https://www.smeunited.eu/news/productivity-growth-in-europe-is-hampered-by-high-administrative-burdens-and-lack-of-finance>.
20. Huws, U., Spencer, N. H., Syrdal, D. S., & Holts, K. (2017). Work in the European Gig Economy: Research results from the UK, Sweden, Germany, Austria, The Netherlands, Switzerland and Italy. Foundation for European Progressive Studies. <https://uhra.herts.ac.uk/bitstream/handle/ 2299/19922/Huws_U._Spencer_N.H._Syrdal_D.S._Holt_K._2017_.pdf>.
21. Ian Bradley, ‘Automatic Exchange of Information (AEOI) services: DAC7, DAC8 and CESOP’ Ernst and Young <https://www.ey.com/en_za/financial-services/aeoi>.
22. ‘Inside DAC7: The EU’s Newest Tax Directive & Its Implementation Status’ (24 October 2024, Trolley) <https://trolley.com/learning-center/dac7-implementation-progress/>.
23. Josef Bergt, ‘A New Era of Tax Transparency: An of DAC7 and its Implications for International Tax Compliance’ (Chambers and Partners, 2023) <https://chambers.com/articles/a-new-era-of-tax-transparency-an-of-dac7-and-its-implications-for-international-tax-compliance>.
24. José Van Dijck, J. (2021). Seeing the forest for the trees: Visualizing platformization and its governance. New Media & Society, 23(9), 2801– 2819. <https://doi.org/10.1177/1461444820940293>.
25. Jane Wang, ‘Taxing the Digital Economy: Theory, Policy and Practice’ (2021) 27 Auckland U L Rev 458.
26. ‘Key Administrative Burdens Faced by Revenue’s Small and Medium Sized Business Customers’ (Revenue.ie , March 2008) <https://www.revenue.ie/en/corporate/documents/research/admin-burden-report.pdf>.
27. Kogler, C., Batrancea, L., Nichita, A., Pantya, J., Belianin, A., & Kirchler, E. (2013). Trust and power as determinants of tax compliance: Testing the assumptions of the slippery slope framework in Austria, Hungary, Romania and Russia. Journal of Economic Psychology, 34, 169–180. <https:// doi.org/10.1016/j.joep.2012.09.010>.
28. Kathleen DeLaney Thomas, ‘Taxing the Gig Economy’ (2018) 166 U Pa L Rev 1415.
29. Kirchler, E., Hoelzl, E., & Wahl, I. (2008). Enforced versus voluntary tax compliance: the “slippery slope” framework. Journal of Economic Psychology, 29(2), 210–225. <https://doi.org/10.1016/j. joep.2007.05.004>.
30. Kaido Künnapas, Begoña Pérez Bernabeu, Katariina Kuum, and Karl Oskar Pungas, ‘Taxes on the Digital Economy’ (2023) Springer International Publishing 101.
31. ‘Kleine Marktplaatsen kunnen nauwelijks aan verplichte belastingrapportage voldoen’ (16 January 2024, NRC) <https://www.nrc.nl/nieuws/2025/01/16/kleine-marktplaatsen-kunnen-nauwelijks-aan-verplichte-belastingrapportage-voldoen-a4879757”

32. KPMG Malta, ‘DAC7 – New reporting obligations for digital platformen’ (KPMG, April 2021) <https://assets.kpmg.com/content/dam/kpmg/mt/pdf/2021/04/dac7-new-reporting-obligations-for-digital-platformen.pdf> accessed 12 August 2024.
33. Laura Crockett, ‘Taxing Uncertainty: Electronic Commerce’ (2013) 4 Geo Mason J Int’l Com L 383.
34. Lidia Vidal, ‘KYC: The EU Digital Service Act Adds to platformen’ DAC-7 Duties’ (19 February 2024, Pinsent Masons) <https://www.pinsentmasons.com/out-law/analysis/kyc-the-eu-digital-services-act-adds-platformen-dac7-duties>.
35. Madeleine Merkx, ‘Exchange of Information and Administrative Cooperation between Countries in a Globalished and Digital Economy’ (2022) 15 Erasmus L Rev 73.
36. Maciej Nyka & Karolina Zapolska, ‘The Impact of the DAC7 Directive on the Functioning of platformen and Platform Operators, from the Perspective of the Legal Model of Their Collaboration with Individuals’ (2024) 29 Bialostockie Studia Prawnicze 177.
37. Mol, J.M., Molho, C. ‘Information about changes in platform economy taxation diminishes optimism regarding future use’ J Econ Sci Assoc (2024). <https://doi-org.ezproxy.hhs.nl/10.1007/s40881-024-00160-y>.
38. ‘Model Rules for Reporting by Platform Operators with respect to Sellers in the Sharing and Gig Economy’ (OECD, 2020).
39. ‘Monitor online platformen 2023 – Cijfermatig inzicht in de kenmerken en ontwikkeling van online platformen in Nederland’ (Centraal Bureau voor de Statistiek, 2024) <https://www.cbs.nl/nl-nl/longread/rapportages/2024/monitor-online-platformen-2023/2-demografie>.
40. Michael Wenzel ‘Tax Compliance and the Psychology of Justice: Mapping the Field’  Taxing Democracy 2002: 41-70.
41. Nathalie Bravo ‘The Meaning of Platform Under DAC7: More Clarity Needed’ (Wolters Kluwer, 2023) < https://kluwertaxblog.com/2023/06/27/the-meaning-of-platform-under-dac7-more-clarity-needed/>.
42. Nevia Čičin-Šain and Joachim Englisch, ‘Joint audits under the new DAC 7’ (Kluwer International Tax Blog, 2021) <https://kluwertaxblog.com/2021/06/01/joint-audits-under-the-new-dac-7/> accessed 12 August 2024.
43. ‘New Survey Reveals that SMEs Struggle with The EU’s Sustainable Finance Framework’ (Eurochambers, 2023) <https://www.eurochambres.eu/publication/new-survey-reveals-that-smes-struggle-with-the-eus-sustainable-finance-framework/>.
44. ‘Non-transposition of EU legislation: Commission takes action to ensure complete and timely transposition of EU directives’ (27 January 2023, European Commission) <https://ec.europa.eu/commission/presscorner/detail/en/inf_23_262>.
45. Pieter Buyl and Annelies Roggeman ‘Do SMEs Face a Higher Tax Burden? Evidence From Belgian Tax Return Data’ (2019) 28 Prague Economic Papers 6 pp729-747.
46. Paula Rosado Pereira ‘Online platformen and Taxes in the EU: A Compatible Match?’ (Springer, 2023) Law, Governance, Technology Series 57 < https://doi.org/10.1007/978-3-031-40516-7_8>.
47. Ramona Ciobanu, ‘Administrative Cooperation in the Field of Taxation’ (2017) 7 Law Rev 62.
48. Rebecca Harding ‘Creating a Modern Digital Trade Ecosystem: The economic case to reform UK law and align to the UNCITRAL Model Law on Electronic Transferrable Records (MLETR)’ (International Chamber of Commerce United Kingdom, 2021) <https://iccwbo.uk/products/united-kingdom-creating-modern-digital-trade-ecosystem>.
49. Roczniewska, M., & Higgins, E. T. (2019). Messaging organizational change: how regulatory ft relates to openness to change through fairness perceptions. Journal of Experimental Social Psychology, 85, 103882. <https://doi.org/10.1016/j.jesp.2019.103882>.
50. Stanislav Bilek, Iveta Kmecova, and Michal Tlusty ‘The impact of taxes and administrative activities of SMEs on their performance’ (2021) SHS Web of Conferences 92, 02006.
51. Thomas Fetzer & Bianka Dinger, ‘The Digital Platform Economy and Its Challenges to Taxation’ (2019) 12 Tsinghua China L Rev 29.
52. Thomas Poell, David Nieborg, and José van Dijck ‘Platformisation’ (2019) Volume 8 Internet Polic Review Issue 4.
53. Vazquez, Juan Manuel, Tax Reporting by Digital platforms under DAC7: A Proportionality Assessment (PhD Thesis, University of Amsterdam, Forthcoming May 2025)
54. ‘What is the Collingridge dilemma and why is it important for tech policy?’ (Demos Helsinki) <https://demoshelsinki.fi/2022/02/15/what-is-the-collingridge-dilemma-tech-policy/>.
55. Wahl, I., Kastlunger, B., & Kirchler, E. (2010). Trust in authorities and power to enforce tax compliance: an empirical analysis of the “Slippery Slope Framework.” Law & Policy, 32(4), 383–406. <https:// doi.org/10.1111/j.1467-9930.2010.00327.x>. 

Albert Cañigueral Bagó is fascinated by technology, platforms and the world of work. How did the sharing and gig economy develop? And what impact does technology have on workers now and in the future? This is what Martijn Arets discusses with him in The Gig Work Podcast of the WageIndicator Foundation.

The platform economy has undergone a considerable (r)evolution in 12 years. I myself have been exploring platforms and their impact on people and society since 2011. I was not alone: during my quest, I met several other explorers. One of my travelling companions is Albert Cañigueral Bagó. In 2011, he started the first Spanish blog on the sharing economy. Not much later, he came into contact with the organisation Ouishare, an international partnership of freelancers. “For me, the sharing economy was personally interesting because I don’t value property much,” he says. “I have a rented house, very little stuff. In addition, I saw many benefits for society and the environment. At Ouishare, I found like-minded people.”

Ouishare: a mix of visions

I first encountered Cañigueral in 2013, during the famous ‘Ouishare Fest’. This was a festival at Parc de la Fayette in Paris where 1,500 entrepreneurs, professionals and activists gathered to discuss the development of the sharing economy. “The special thing was that we brought together all the different perspectives,” Cañigueral says. “We at Ouishare were activists, we loved open source. For us, the success of big companies like Uber and AirBnb was disappointing. Still, we managed to have good conversations with these tech entrepreneurs about the future of sharing.”

This was also necessary, he says. “In Spain, delivery platforms Deliveroo and Glovo had just started and that led to conflicts immediately,” he says. “Logical. After all, the inventor of the boat is also the inventor of the sunken ship. The negative effects of platforms can never be completely avoided, but if we work together we can ensure that the positive effects prevail.”

Utopias for the world of work

When it comes to platform work, the biggest problem is the skewed power relationship between the platform and the platform workers, he says. A few years ago, he wrote a book on macroeconomic trends and their impact on labour: El trabajo ya no es lo que era (‘Work is not what it used to be’). In it, he explains, among other things, how the industrial revolution is transitioning into the digital revolution. “During the previous revolution we were children, now we are parents,” he says. “Now we also have to behave like adults. That means facing the facts and taking responsibility: we cannot ignore developments, we have to respond to them in a smart way.” 

We work differently in the digital age and platform work fits in with that, he explains. “The idea that a job always has to be 40 hours a week is outdated. Part-time, flexible or freelance work sometimes fits better with all the other pursuits in modern life, for example volunteering or taking care of our family.”

We are still in the middle of the transition, he says. His book contains seven utopian situations for the future of work and technology. “We can use these visions to approach a future where the positive effects of technology outweigh the negative.”

Collective action and cooperation

Previously, unions, for instance, protected workers from employers. Such collectives made for a less skewed balance of power. In a fragmented world, collaborations are more important than ever, Cañigueral stresses. “Platforms provide structured access to fragmented and often invisible work,” he explains. “But if the conditions in doing so are not clear or if you cannot influence them, you feel like a slave to the platform.”

Therefore, he says, new collectives are needed. As an example, he mentions the Worker Info Exchange, a non-profit organisation that helps platform workers gain insight into the data collected about them during their work. Read more in this interview with founder James Farrar.

That does not necessarily mean the end of existing unions, says Cañigueral. “Some will disappear, others will reinvent themselves. Collaborating with new initiatives is incredibly valuable. If they are willing to learn from each other and set up collective actions, it ultimately delivers more for working people.”

WorkerTech: collective facilities for those not in an employment relation

Platform workers lack collective services in addition to representation. As freelancers, they have to find their own ways to cope with retirement, disability and illness. So-called ‘WorkerTech’ are solutions to this. They are collective services especially for Glovo delivery drivers, Uber drivers or Fiverr freelancers, for example. Interestingly, WorkerTech is usually not reserved for one type of platform worker. Often such initiatives are open to freelancers working through various platforms.

For example, Pulpo.Life, a Latin American startup that arranges health insurance for platform workers and freelancers. Another example is Nippy, an Argentinian startup that offers all kinds of services for freelancers: from bank accounts to insurance. Cañigueral: “These kinds of initiatives prove that a permanent contract is not the only way to provide decent working conditions and collective services to workers. 

Cañigueral does not see WorkerTech as competing with public social security, but complementing it. “The government can learn from such initiatives. For example, such startups know better how to use data to respond to individual needs. But that requires mature politics.”

Analysis

Catching up and looking back with this good friend on the development of the platform economy gave me new insights. We both started our research journey in the sharing economy and later turned the focus to the job and platform economy and the future of work. The impact of technology on people, organisation and society was the common thread.

Cañigueral explains that we are in the midst of a major change. In my opinion, platforms are a logical testing ground for technologies that will eventually impact the entire labour market. Just look at WorkerTech, which was born out of a platform economy need but also offers many benefits for other workers. Increasingly, WorkerTech is also available to all workers: self-employed, employees or them working in the informal market. The reason these kinds of startups fit the needs of modern working people so well is because they can develop and adapt faster than existing institutions.

Finally, we discussed how social security can be shaped. It will be interesting to explore how we can ‘re-bundle’ these kinds of securities so that they do not only suit permanent salaried employees, but are of use to all kinds of workers.

Indeed, in the Western world, many securities are linked to the permanent contract, but this is not the case at all in many parts of the world. Moreover, in many countries the informal labour market is large: in Latin America, some 56 per cent of people work in the informal market. At other places in the world this percentage is even (much) higher. Currently, they are invisible to all institutions and have no protection at all. Platform work, as a gathering place, can give structure to this fragmented group of informal workers, bring them together and give them access to the securities and protection of WorkTech. So, especially in the non-Western world, this can be a substantial improvement.

This blog and podcast were also published on Gigpedia.org

Delivery riders connected to platform Glovo do not all get employment contracts, but are covered by the ‘Couriers Pledge’ in 21 countries. With this, the company promises delivery workers safety, community, equality and a fair income. Why, how does it work and what does a platform worker get out of it? Martijn Arets travelled to Barcelona and spoke to Glovo spokesperson Magalí Gurman.

This edition not a collection of articles on the platform economy, but a blog and podcast episode around Glovo’s ‘Couriers Pledge’: a delivery platform operating in 24 countries. Have fun reading and listening!

The main discussion about delivery via platforms is about whether such a platform should employ the couriers. After all, in many Western countries, things like safety and fair pay are linked to an employment contract. But this is far from common everywhere. Therefore, platforms are looking for ways to offer riders better conditions regardless of their employment status. One such platform company is Glovo. For The Gig Work Podcast from the WageIndicator Foundation, I travelled to Barcelona and sat down with Magalí Gurman, Institutional affairs & government relations at Glovo.

From customer to courier: keeping everyone happy

The creators of delivery service Glovo know all too well how important international differences are. The Spanish company operates in 24 countries in Southern Europe, Eastern Asia and Africa, among others. Through the Glovo app, delivery drivers bring all kinds of products to private customers: from meals to flowers and groceries. Customers order something from a restaurant or retailer via the Glovo app, the delivery drivers pick it up and bring it to the customer as soon as possible.

“So basically we have three sorts of clients: customers, retailers and delivery drivers,” Gurman says. “Each has its own app with corresponding processes, because we are providing a different service for each user.”

Local differences

In addition, operational needs differ by region, she explains. “For example, here in Barcelona it is normal for delivery drivers to ride bikes or scooters. In cold Romania, on the contrary, most couriers deliver by car. What we also see is that delivery times changes from city to city because of different reasons, like the state of the streets, and therefore, people are used to different types of services. Moreover, laws and regulations vary from country to country.”

It is up to Glovo to meet all these needs and preferences, she says. “Our goal is for all users to be satisfied, not just the end customers.”

‘Flexibility does not exclude good working conditions’

To better meet the needs of couriers, Glovo introduced a ‘Couriers Pledge’ in 2021. With this, the company aims to improve the conditions for all couriers, regardless of their working status. “Two years ago, there was much less regulation around platform workers than now,” she says. “We wanted to lead the way and show that the flexibility of the platform economy does not preclude good working conditions.”

Because there are big international differences in legislation and preferences, Glovo introduces the Pledge on a country-by-country basis each time. This starts with research into local laws and regulations and an exploration of the possibilities within the organisation. How much money and people does the platform have available in this region? Meanwhile, 21 of the 24 countries each have their own Pledge.

“With this, we want to set a standard example for the platform economy sector,” says Gurman.     

Safety, community, equality and income

The Pledge rests on four pillars: safety, community, equality and fair earnings. Once a rider has delivered a certain number of products within a certain period of time, they can access these benefits. For example, couriers get safety training, helmets and parental leave. The exact conditions vary from country to country and depend on the couriers’ preferences, among other things. 

Under the ‘community’ pillar, Glovo organises events to connect couriers with each other. “Sometimes we discover needs we could never have thought of ourselves. For instance, it turned out that relatively many couriers in Kenya had eye problems. We work with a company that does eye checks and we reimburse glasses if needed.”

The current Pledge is the foundation, says Gurman. “We want to be more and more responsive to what couriers need to deliver in a healthy and safe way. Satisfied couriers are also good for business. That’s why we keep researching, talking and adapting.”

Independent help

Glovo also works with two independent organisations around fair working conditions and pay for platform workers:  Fairwork and the WageIndicator Foundation. “Organisations like these keep us on our toes,” says Gurman. “An organisation like Fairwork critically assesses our practices and sets our conditions against those of other platforms. This is how Fairwork helps us improve operations. Our working conditions in Africa are currently the best rated in that region.”

WageIndicator Foundation helps Glovo with the third pillar of the Pledge: payment. “This is the most important issue for couriers and at the same time the most complicated,” says Gurman. “WageIndicator helps us with data on living wages and minimum wages for comparable work. Rates vary from country to country and are constantly changing.”

Most Glovo couriers work as freelancers. Political parties and trade unions in several countries advocate giving these workers employment contracts. “An employment contract does not necessarily lead to better working conditions,” says Gurman. “What an employment contract means for working conditions varies greatly from country to country. The Pledge applies to every courier, regardless of the form of contract.”

Labour contract is not always an improvement

Gurman is right: a labour contract is by no means always better for a courier. For example, if delivery workers are employed through a middleman, they often get the most minimal protection legally possible.

Moreover, it is not always what working people want, figures from the Leeds Index of Platform Labour Protest show. In only 20% of all demonstrations do workers ask for a labour contract, analysis of nearly 2,000 protests shows. This varies greatly by country. For instance, ’employment status’ is an issue in 37.1% of protests in North America, because it brings more security there. In Africa, it was an issue in only 4.7% of protests, because a contract offers hardly any benefits there. 

Analysis

I like how Glovo is trying to improve conditions for couriers. Especially when you know that working as a delivery driver is done in an informal way in many countries. So before, couriers had no protection at all.

 For Glovo, it is a balancing act between what is allowed and what is possible, including financially. And it is true that an employment contract is not worth the same everywhere, but in certain (European) countries it does offer more protection. Could the Pledge be an addition there? There are also several important employment conditions not included in the Pledge. Consider transparency, explainability and accountability of automatic decision-making processes and worker representation. According to Gurman, this is now “not always legally possible”, but that is a little too easily dismissed. In my opinion, there is quite a solution to this. Where there is a will, there is a way.

So the Pledge is not yet perfect, but it is a good step in the right direction and a great example of how to take into account different wishes and circumstances in various countries in which a platform operates.

British tech expert and Uber driver James Farrar has been championing labour rights in the platform economy for eight years. Often successfully. Platform expert Martijn Arets spoke to him on behalf of the WageIndicator Foundation in The Gig Work Podcast about justice, data and the future of platform work. “Our battle has only just begun,” he said.

In the platform economy, the impact of algorithms and AI on the way we work is no longer future music, but daily practice. In platforms for very short jobs, that impact is greatest and the power relationship between the platform and workers is most unbalanced. I see it, for example, with food delivery, taxi apps and clickwork – websites on which people worldwide earn money from home with small online tasks that train artificial intelligence. Workers just have to do the task and have no say in the content, access and reward. The platform’s algorithm decides, but it is far from transparent and fair.

The Gig Work Podcast is available at Apple Podcasts, Springcast, Google, Spotify, Stitcher, Pandora, and Pocket Casts.

These algorithms are difficult for many to understand; they seem like ‘magic black boxes’. Platform operators like to keep this mystique alive. Indeed, it obscures the fact that algorithms do not just make choices, but that they are based on the wishes of platform operators. Knowledge is power and as long as workers do not understand how the platform works, they cannot change it either.

Former Uber driver James Farrar wants to equalise the balance of power by giving platform workers more knowledge and access to data. He does this by uniting platform workers in the Worker Info Exchange and the App Drivers and Couriers Union, a union for platform workers. By analysing drivers’ data, he won several influential court cases. For The Gig Work Podcast from the WageIndicator Foundation, I took the train to London and spoke to Farrar about his fight for justice in the platform economy.

Sharp drop in fare

Farrar drove for Uber as a driver between 2015 and 2016 and experienced how the company violated labour rights. Taxi drivers were structurally underpaid, he says. “I noticed those two years that driver pay dropped sharply. At the same time, the number of drivers increased. So that meant fewer rides and a lower price per ride.”

When he just started driving via Uber, the company did pay well. “We also got hefty bonuses when we brought in new drivers,” he says. “I remember people receiving 30 to 40 thousand pounds through those referrals alone.”

But all those extra drivers depressed fare prices. “Uber said: if the price goes down, demand goes up and then you have more work,” Farrar says. “In other words, you just have to drive more. That, of course, is completely insane. Yet we drivers believed the story and started driving more rides per day at a lower and lower tariff.”

Not an employee and not a self-employed person

Nevertheless, the first lawsuit was not directly about tariffs. Farrar was attacked in his taxi and reported it. The police asked his passenger’s name. “I had no idea, after all, the ride was booked through the app,” he says. “The cop could hardly believe that. So I called Uber: who was my passenger? They wouldn’t say, even to the police.”

After 12 weeks of silence from Uber, Farrar hired a lawyer. “Who said: ‘Uber doesn’t owe you anything because they don’t recognise you as an employee. But in my opinion, you are a worker. That’s a category between self-employed and employee, where the employer does have obligations to the worker.”

A worker’s rights in the United Kingdom (UK) include minimum wage. To claim his worker status, Farrar took Uber to court. The company came back with an army of lawyers and data. The company showed infographics about all the work Farrar had done. These included, for example, the number of rides accepted and rejected, tariffs and how many hours he had worked.

Fool around with figures

“I didn’t have any access to that data,” he says. “That was very difficult, because Uber’s lawyers could claim about my performance all they wanted. The company twisted the facts and emphasised specific data.”

According to the platform, Farrar turned down half of the rides offered. “But that said nothing about my productivity,” he says. “I was completing one-and-a-half rides per hour, more than what Uber’s guidelines expect a driver to do. And even then they said I wasn’t deserving of minimum wage.”

Farrar knows well how data and technology works. He has a background in tech and worked for German software company SAP. Since the new European privacy law (GDPR), companies have to share data with users if they request it. Before that, there was a law in the UK that was quite similar to the GDPR. It also contained the right to access your own data.

The beginnings of a union

“I demanded this right and asked fellow drivers to do the same,” Farrar says. “Uber showed that an employer cannot be trusted with data. If one party has all the knowledge, it can tell the rest what it wants.”

In retrospect, that was the beginning of the Worker Info Exchange (WIE), he says. WIE is a non-profit organisation that helps platform workers gain insight into the data collected about them as they work. “Actually, even then we were an informal union with a common goal: equal knowledge of data for all the platform operator and users.”

Later, Farrar started a full-scale union for platform workers: the App Drivers & Couriers Union (ADCU). In the podcast, he talks more about what that entailed.

Case won

In 2021, the UK Labour Court ruled that Uber drivers were indeed workers. This meant they were entitled to certain employment rights, such as minimum wage and paid holidays. In total, Uber settled over 70,000 drivers. Uber does not disclose how much it paid out in total, but according to The Guardian, the company had set aside some £465 million.

The case sets a precedent for the entire UK gig economy. “The judge made it clear that an employer cannot make up its own mind that it is working with freelancers instead of employees or workers. Not what is in the contract applies, but how the parties work together in practice is leading.”

As an added bonus, Uber was also found responsible for VAT payments on taxi rides in the UK. The platform settled for the astronomical sum of £600 million.

The battle is not over yet

According to the court, opp-taxi drivers have workers’ rights from the moment they log in until they log out. Uber disagreed and does not count waiting time between rides. “Uber still does that, nobody is taking action against that own interpretation,” Farrar says. “So I’m not out of court yet.”

Farrar did not receive his back salary and holiday pay until 2023: £20,000. This is because he did not want to sign a non-disclosure agreement. “My settlement is public and makes it clear that Uber broke the law. That is a major victory for me, but the fight has only just begun.”

Robo-firing

A second important case revolves around so-called robo firing: being automatically fired by the algorithm. Under the guise of safety, Uber introduced automatic facial recognition based on a system from Microsoft during the corona crisis. There were quite a few problems with that. Taxi drivers were incorrectly not recognised and therefore immediately lost their access to the app and their taxi licence.

Research shows that the system is generally 97% accurate, but for people of colour the percentage is significantly lower. For dark-skinned women, it is only 70%. Farrar: “There are a lot of people of colour working for Uber. With those percentages, you end up with tens of thousands of people a year being falsely accused.” Suddenly losing your access to work has hefty consequences, as the fixed costs of the car continue as usual.

The WEI stood up for these affected people and won most cases, often simply because Uber could not or would not share data. “It is startling how little knowledge Uber’s European staff have,” says Farrar. “For example, the head of regulation was excluded as a witness in court because he could not answer basic questions about firing via algorithms. For that, we had to video call someone in Silicon Valley.”

Conclusion and analysis

Farrar’s interview shows that equal access to data is essential. It equalised the power relationship between platforms and workers, which is important because, unfortunately, taxi and delivery platforms still have to be held accountable through lawsuits. The fact that platforms as ‘private regulator’ can easily adjust the rules of the game does not help either: after a ruling, jurisdiction just starts again from scratch. New regulation is badly needed, but Farrar’s court cases also show that existing regulation can work well. At least if they are enforced, but that unfortunately happens too rarely.

The work of Farrar and his colleagues is therefore important, not just for the platform economy. Technology is in fact affecting the labour market as a whole. Indeed, algorithms are playing an increasingly important role in the organisation of labour and the control of work.

Finally, I worry whether legislators and enforcers can keep up with developments. In any case, it is important that they are not distracted by tall stories of utopian scenarios by tech entrepreneurs, but focus on what is actually happening right now. Ask questions, be realistic and curious. Because the impact of technology and platforms is not in the cloud, but on the streets.

Want to know more? In the podcast, I talk to Farrar further about:

– Founding a union for platform workers

– The rise of the platform economy in the UK

– The future and issues surrounding algorithmic management