Spain has introduced new legal proposals on 'gig economy' work, which would introduce a presumption that delivery platform riders and drivers are workers, placing the burden on the platform to show that they are not. Gemma Fabregat, Of Counsel at Ius Laboris Spanish firm Sagardoy argues here that these changes may not be as far-reaching as they first appear.

In the wake of a Supreme Court ruling in September 2020 that riders for the food delivery service Glovo were employees, the Spanish government has announced new legislation on platform workers' rights and status.  

The proposed regulation (the Royal Decree-Law on the employment status of individuals working in distribution via digital platforms) specifically refers to employment status in delivery via digital platforms. It will give rise to a new regulation added to the Spanish Statute of Workers Rights and, based on the wording of the proposal, appears basically to clarify the presumption of worker status currently set out in article 8.1 of the Statute of Workers Rights. 

To the extent that the proposed regulation establishes a presumption of ordinary 'worker'  status for platform riders, there is scope to offer evidence to the contrary.  

It appears that the aim of the regulation is to clarify certain aspects of the way the presumption of 'worker' status operates in real life. Readers should note the limited scope of the proposal: it does not affect all platform-based work but only the delivery of consumer products or goods by employers exercising entrepreneurial powers to organise and control (directly, indirectly or implicitly) of the service or working conditions via a digital platform using an algorithm.  

The regulation clarifies the law and does not establish worker status. This is confirmed by the express statement that the clarification does not affect article 1.3. of the Statute of Workers Rights. Article 1.3 rules out worker status in relationships that are not subject to the general conditions of article 1.1, that is, 'independence' ('ajenidad' in Spanish, the idea that the service provider does not own the means of production or the benefits generated by his or her activity)  and 'subordination' (the ability to control and direct a worker's activities).   

It must therefore be understood that the aim is to use Spanish law (adjusted for EU law) in this field. This means that where subordination and independence are both present, there will be a presumption of worker status.   

However, this does not change very much, since if both conditions are met, worker status will obviously never have been a presumption but a fact. In contrast, if neither condition is met, there will be self-employment. The problem arises when there is doubt about whether there is both independence and subordination.  

It is here that the regulation may introduce two new factors: 

  • Firstly, it appears that platform-based work is normally dependent since the market relationship with the client and supplier is via the platform: it is not owned by the rider.  
  • Secondly, it appears that subordination can be created through dependence on a computer programme that uses algorithms to control the service or working conditions to such an extent that it is clear that organisation, management and control are in the hands of the platform via the algorithm (logically, if they are done in any less modern but equally clear way, this will have the same impact, in this case directly via article 8.1 of the Statute of Workers Rights). 

Does this mean that all distribution work done as described above through a platform and involving any form of control via an algorithm creates worker status? I do not believe this would be justified. This would go further than the draft indicates and would clash with the exclusion of independent work from the scope of the Statute of Workers Rights (additional article 1 of the Statute of Workers Rights that cannot be amended or repealed, and article 1.1. of the Statute of Workers Rights).  

In effect, there cannot be an absolute presumption of worker status cannot apply unless there is proof that the work is controlled by the algorithm (or, as explained above, by any other system) to the extent that the power to manage, organise and control the work and/or working conditions lies with the platform. Only then will there be subordination and justification for worker status. 

If we accept this, a question obviously arises as to how we establish subordination.  The concept is graduated and nuanced. There can be factors that depend on combination with others that blur or exclude subordination, meaning that there are many intermediate scenarios between total subordination (employment per se) and total non-subordination (self-employment per se).  The courts must decide in these cases, based on the facts, whether subordination predominates and consequently whether worker status exists.  

This must be the case, since preventing self-employment from being treated in the same way as a small businessman or a small sole trader could violate constitutional entrepreneurial freedom. This freedom includes the right to choose a profession or position, which must include the ability to carry out a profession or position in any lawful manner.  

Dependent worker status cannot be forcibly applied to people who meet all the conditions for self-employment, who are therefore entitled to carry out their own economic activities and who wish to do so. What the regulation can do, and this is what requires interpretation, is clarify when work is carried out on a subordinate or non-subordinate basis. 

Taking the regulation literally, we are therefore brought back to the classic question of what makes work subordinate or non-subordinate. Generally, this will mean considering the intentions of the parties (so long as there is no evidence of fraudulent purpose), though the relationship will analysed on the factual circumstances and not simply as the parties state them to be. It will also mean considering the extent of the material resources provided by the worker and imputed to him or her, and the details of the type of work.  

Domestic and European court decisions provide indicative guidance on this, taking into account factors such as the following: 

  • The habitual nature of the work: this is not a determining factor but one that should be assessed: for example, whether there is a duty to make regular contact, justify absences, etc.  
  • Whether the platform-based service is exclusive in nature: it is not essential that it is not but if it is, it probably offers a real indication of whether the worker has a small distribution or transport business serving a number of clients.  
  • Whether working hours are directly or indirectly fixed or freely decided by the worker, who suffers no penalty for such decisions. This does not rule out the 'employer' setting small limits or criteria (as established by the European Court of Justice in B v Yodel, C-692/19). 
  • Routes and the freedom to choose them and the absence of control over them, meaning that any geolocation services can only be used to check service delivery or to identify the closest delivery man for the benefit of the client, as in the case of radio-taxi systems where there is no doubt that these drivers are self-employed workers.
  • The ability to turn down work without a direct or indirect penalty (as in decision STS 805/2020 of 25 September 2020 in which the court shows the importance of these penalties). 
  • The personal nature of service provision and the ability to replace the provider.  
  • The existence or absence of grounds for contract termination or cancellation, such as breach of labour law, or grounds based on similar controls (rating by management, clients, etc.). This does not include the right to exclude anyone against whom clients have made a justified complaint from service.  
  • The existence or otherwise of actual service orders. This does not prejudice the ability to give general instructions on conduct (such as technical instructions) that do not clash with the services provided by self-employed workers or reasonable instructions that are also compatible with self-employment and do not intrinsically affect non-subordination. 
  • Payment or non-payment of waiting times, downtime and compensation for breakages, faults, etc.  
  • Other factors. 

By assessing all these factors, a decision can be made as to whether a platform uses an algorithm or its own decisions to direct and control work. It does not establish whether worker status is created. 

Does the above mean that the new regulation adds nothing? Obviously not. The new regulation has two very clear effects:

1. It states that work done via this kind of distribution platform is normally notindependent.The fact that a platform exists does not make it so. But what is really relevant to the existence or otherwise of an employment relationship is not so much the ownership of the means of production (including the digital platform), but subordination.   

2. It states that an algorithm governing the provision of services will create a presumption of subordination that, combined with the above and unless contradicted, will lead to an assumption of worker status. 

The new regulation consequently shifts the burden of proving that the algorithm (or management and control system) does not hinder the worker's freedom to organise all or most of the essential decisions governing the work provided and the terms under which it is provided.  

So far as the regulation is concerned, the existence of an algorithm is sufficient proof that this hindrance to the worker's freedom exists, since this seems the most logical function of the algorithm. The 'denier' (the 'employer') carries the burden of proving that this is not the case and that the algorithm does not create organisational subordination, at least in all essential aspects. This is because the algorithm is able to condition or control certain, but probably not all, factors.  

This means the worker could retain sufficient organisational control to show that the parties are able freely and lawfully to organise their relationship on a self-employed basis. While the algorithm provides indicative proof to justify the presumption of worker status, it does not of itself constitute full evidence. 

It is precisely for this reason that a new subsection of the Statute of Workers Rights (article 64.4 (d)) establishes a duty to inform workers' representatives about computer systems that can impact working conditions, access to and retention of employment and any profile processing. This information can of course be used as evidence for the above and must therefore be understood to refer only to information on the rules used by the algorithm when making decisions on working conditions, access to and retention of employment and the production of profiles. It will not include technical details that are not required for this purpose and could affect industrial secrecy. 

Conclusion

  • The proposed Regulation contains no declaration or absolute presumption that platform-based delivery work necessarily creates worker status. 
  • The Grounds section of the proposal alludes to the need for, and importance of, examining the real nature of the relationship when clarifying the nature of the services provided and the actual and asymmetric organisation of obligations under the contract. This view is supported by the fact that the regulation introduces a presumption that the existence of an algorithm and a platform of itself creates subordination and worker status.  
  • More forceful proof is therefore required since the new regulation appears to clarify article 8.1 of the Statute of Workers Rights and the presumption of worker status. This is a strong presumption but it can be disproved by factual evidence, and all contributing factors can therefore be examined on a case-by-case basis.  
  • The key element of the regulation is the transfer of the burden of proof to the 'employer'. The 'employer' must now in all cases show that despite the existence of an algorithm that conditions or controls some aspects of the work, the essential organisation and management of work remain in the worker's hands. Despite the existence of a signed contract and a platform or algorithm, the relationship between the parties is not so asymmetrical that it impacts the worker's ability to organise and manage his or her own work.  
  • A stricter interpretation of the regulation could clash with entrepreneurial freedom and the right to choose a profession or position. It is not supported by the ECJ decision in B v Yodel (C-692/19), which includes in these principles the ability to use replacements, accept or reject jobs, set a limit on the number of jobs accepted, provide services to third parties including competitors and set your own working hours, subject to certain parameters. These must be real conditions and the Grounds insist that services must be real, not simply fictitious.  

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.