The market-making matching of supply and demand – not any kind of sharing – defines online platforms. Extension of markets into new spheres may provide new opportunities, but it also undermines the key institution through which our societies organize protection of workers and set their pay: the employment relationship. Re-organisation of activities that traditionally relied on employment into self-employment is, perhaps, the most radical transformative potential of platforms.

So far, however, successful platforms, such as Uber, have rather reorganized sectors that had already relied on some form of self-employment. However, platforms bring greater pressure on pay and working conditions generally as they increase competition by lowering barriers to entry, sometimes also by facilitating regulatory avoidance.

There is now a considerable body of evidence on the extent of platform work as well as working conditions within it (see, for example, here, here, or here). Most workers appear to use platforms to top up their regular income, but there is also a sizable minority, probably exceeding well over a million of workers in the EU, that rely on platforms as their main source of income. Only a small fraction of platform workers attain the local minimum wage. Most transactions on platforms are neither taxed nor covered by social insurance.

Low pay and, somewhat paradoxically, a lack of control over their working time are by far the top grievances of platform workers. Lack of insurance is a major concern for those delivering food. Accidents, including serious injuries, are common among cyclists.

Unfortunately, the regulatory response has neglected the issue of pay and working conditions. At best, we hear about how important it is to ensure portability of insurance systems. However, portability cannot bring security, particularly if the problem is low pay.

Given the precarious position of platform workers, additional measures should be considered to address the risks related to platform-mediated work. First, while many platform workers are likely to be eventually recognized as employees, the specific nature of work on platforms calls for updating existing labour legislation. Indeed, platform workers represent a category of workers that require special protection, similar to the regulatory provisions for part-time, fixed-term, and agency work that exist also at EU level.

Second, EU competition law should not prevent platform workers from self-organizing, as might be the case under current provisions. At the moment, organizing of self-employed persons is in principle not permissible under Article 101 TFEU. However, EU case law has excluded “false self-employed” from the applicability of Article 101 (Case C‑413/13). The decisions by national courts of whether the platform-workers should be considered an employee, self-employed or false self-employed determine the scope to regulate salaries and working conditions via collective agreements but they will have to evaluate the cases in the light of EU competition rules.

Instead, collective agreements should be possible to extend to wider categories of worker than ‘employee’, with a view to including platform workers. At this moments, legislation in many member states does not include possibility for self-employed persons to conclude a collective agreement or to be covered by one.

Third, workers who do not qualify as employees should be protected through regulations on self-employment, or through a platform-specific protection.

Equally, it is important to distinguish a variety of platforms, with different impacts on labour markets as well as opportunities and limits for regulatory responses. Matching platforms that set pay and contract conditions, such as Uber or Deliveroo, are most compatible with protection that approximates to, or fully complies, with standard worker protection. In fact, Uber pays a guaranteed minimum wage per hour in a number of markets. In Belgium, Deliveroo workers benefit from an agreement, negotiated by the agency SMart, that includes a minimum hourly pay rate, minimum working time, insurance against injury at work, and social insurance. However, a regulatory framework is required to extend such provisions to wider groups of workers.

Platforms that reorganize local markets are also easiest to regulate as both customers and suppliers come under one jurisdiction. The oligopolistic tendency which comes with the network effects also makes it easier for the regulator to target the handful of dominant platforms – such has been the experience of Airbnb and, in some cases, Uber. Here, platforms in fact provide an opportunity to regularize undeclared activities, as their model allows for an efficient monitoring of micro-transactions as well as for their incorporation into insurance and tax systems.

On the other side of the spectrum are platforms , such as CrowdFlower and Upwork, that facilitate the remote provision of services, thus potentially leading to the offshoring of work from local labour markets, often across borders. This is one reason why an EU-wide framework is needed, but additional solutions need to be sought for outsourcing of services to low-income countries.

Finally, platforms, such as Upwork or Mechanical Turk, that allocate work through various types auctions or “contests” – which remunerates only the winners while competing bidders do a job without getting paid at all – call for new institutions of protection to address contest-specific issues, such as a lack of grievance mechanism in case of non-payment, or unfair practices in specifying work content. Again, decent pay is the main challenge.

The rise of platforms thus brings a number of challenges and it is not obvious how decent pay and working conditions can be ensured in all contexts. In any case, we need an active response from the policy makers. It does not help that the approach from the European Commission, as well as in many member states, has focused on removing regulatory barriers to this “collaborative economy” and ignored the threat to pay and working conditions. Moreover, in practice, the question is not whether platform work can be left unregulated or not, as we have already seen platforms developing their own codes of conduct. The question is whether the new regulatory environment will reflect the narrow interest of some business(es), or a balance of interest among all stakeholders.