Platform work and labour protection. Flexibility is not enough.
Regulating for Globalization
Please refer to this post as:, ‘Platform work and labour protection. Flexibility is not enough.’, Regulating for Globalization, 23/05/2018, http://regulatingforglobalization.com/2018/05/23/platform-work-labour-protection-flexibility-not-enough/
In April 2018, the Labour Tribunal of Turin, Italy, rejected a claim from six platform workers of the food-delivery company, Foodora, seeking to be reclassified as employees. In reaching his decision, the judge relied extensively on the fact that these workers were allegedly free to decide when to work and to disregard previously agreed shifts, returning a verdict that the six workers were self-employed.
Should flexibility of working time be sufficient grounds to exclude workers from labour protection associated with an employment relationship? If during a worker’s shift the person is subject to extensive managerial powers and have no say in determining their working conditions, should they still be deprived of fundamental labour protection because they can decide their shifts? I would argue that current tests on employment status do not adequately take these nuances into account, with the end result that a growing number of workers, particular platform and casual workers, are being deprived of basic labour rights.
The decision in Turin was not the only example. Over the last few months, courts and public bodies in various parts of the world delivered decisions on the employment status of platform workers. While some represented victories for platform workers, a number of these decisions rejected the claims of platform workers to be protected under employment law.
In November 2017, for instance, the Paris Court of Appeal found that a Deliveroo rider could not be reclassified as an employee because he was free to select his “shifts” and choose when to work, and refusing a shift did not trigger any sanction from the company. A few months later, in January, the Paris Conseil des prud’hommes, a lower judicial body in France, declared that an Uber driver could not be considered an employee for the same reason: he had the flexibility to choose on which days and hours he worked.
In the Unites States, a federal judge followed the same line of reasoning in the case Razak v Uber when he decided that Uber drivers are independent contractors because they “work when they want to and are free to nap, run personal errands, or take smoke breaks between trips”.
In the UK, the Central Arbitration Committee, ruled that Deliveroo riders cannot be classified as workers entitled to statutory collective labour rights because their contract includes a “substitution clause”, allowing them to ask someone to replace them during their shifts. This, according to the Committee, is sufficient to exclude that their contract is for “personal” services, making them self-employed. The Committee did not investigate whether this right to be replaced was genuine. On the contrary, it ruled: “even if [Deliveroo] did it in order to defeat this claim and in order to prevent the Riders from being classified as workers, then that too was permissible”. The High Court upheld this decision on appeal.
We must not overestimate the actual flexibility of platform work
Other bodies, however, challenged the idea that workers of ride-hailing and delivery platforms are genuine fully-fledged self-employed persons.
In Belgium, the Administrative Commission for the Governance of the Employment Relationship decided that Deliveroo riders qualify as employees under Belgian Law. The Commission noted that riders are not able to manage their working activity independently and that they are not actually free to organize their working time, since they are “required to reserve, and submit to the agreement of Deliveroo, more than one week in advance, the time slots during which [they] should be available to the ‘platform’ (without, however, having the guarantee of actually having orders to execute during these periods)”.
In Spain, the Labour inspectorates of Valencia and Madrid held that workers of Deliveroo and Glovo, another delivery platform, work in conditions of subordination to the platform, something that is not compatible with the purported self-employed status of riders. The inspectors found that the platforms strictly control the amount of time workers take to complete their delivery and use the smartphone’s app to encourage them to work faster.
Moreover, judging over the employment status of Uber drivers, the London Employment Tribunal decided that they qualify as “workers”, an intermediate status between employment and self-employment giving access to labour rights such as the minimum wage, holiday pay and non-discrimination protection. The Tribunal dismissed the company’s claim that drivers are self-employed, noting that Uber “imposes numerous conditions on drivers (such as the limited choice of acceptable vehicles)”, “instructs drivers as to how to do their work and, in numerous ways, controls them in the performance of their duties” and “subjects drivers through the rating system to what amounts to a performance management/disciplinary procedure”. This decision was upheld in appeal.
These decisions show how complex the issue of qualifying platform work is under employment law. The ones accepting the workers’ claims point towards working activities that do not correspond to the idea of genuine self-employment – the lack of independence when completing the task or in setting its price, the instructions given by the platforms and the control they exert over the time spent for every task and over the quality of the work done, also through the customers’ rating.
The ones rejecting these claims, however, point to elements that do not correspond to the commonplace idea of employment, such as the flexibility of deciding if and when to work and the ability to ask someone to replace them in doing the job. In many legal systems, these elements are traditionally associated with self-employment. And while it is impossible to discuss here the soundness of each decision under the relevant national laws, it is interesting to notice, that the elements that were deemed to weigh in favour of self-employment do not regard the execution of the tasks. Flexibility in working time and the ability to be replaced, instead, concern when and by whom the work is going to be executed.
When platform workers do work, their activity is not truly independent, as it suits to self-employment; it is monitored second by second with GPS and digital tools, prices are fixed by the platforms, there is a rating system that also takes account of the customer’s level of satisfaction, and algorithm-driven assessments are made to understand whether the service was successful or not. If the algorithm returns a poor performance, the worker can be sanctioned, expelled from the platform, given less favourable shifts, or assigned less work.
But until the criteria under which the performance of workers are assessed are made public and transparent, should this flexibility be taken granted? It is not clear, for instance, how the algorithms evaluate workers’ irregular work schedules or non-compliance with previously agreed shift when assigning new shifts and tasks. How can we be sure that workers are not penalised if their work patterns do not match the algorithms’ settings?
Under new policies adopted in November 2017 by Deliveroo in the UK, “riders on hourly rates can no longer work the same ongoing shift patterns to guarantee an income. Instead, they have to compete for available hours at the start of each week with those Deliveroo classes as ‘high performing’ and who have priority.” When such high competition for work exists and when it is not clear how shifts or tasks are distributed by the platform, can schedules really be deemed flexible? And if workers do not suffer any penalty for skipping a shift, why would they bother to find a replacement, when substitution clauses allow doing so?
Flexibility should not exclude people from protection
We should also be asking whether the criteria for classifying employment status are in need of a revision. Some of these criteria were delineated when work was much steadier and stable in time. Are they still applicable for an era when working time is ever more volatile and technology allows platforms to organize and strictly monitor their workforce in a “just-in-time” fashion, mitigating the need to rely on the fixed shifts of workers? Does it make sense to deny labour protection to workers with flexible schedules when more and more traditional employees in the clerical professions can set their own schedules? It is, in fact, grossly inaccurate to assume that employment and flexibility cannot go together.
From a policy perspective, it seems absurd to deny access to fundamental rights such as freedom of association in trade unions, collective bargaining and the right not to be discriminated against to these workers because of their purported flexible working time or ability to be replaced. Nor is it reasonable to exclude from social protection, occupational health and safety law or minimum wages a growing segment of the working population who, during their actual spells of work, are subject to invasive control from platforms and who do not have any say on their working conditions, including the compensation of any of their tasks.
It is sometimes said that labour protection should be updated to face the changes that work is undergoing. Much before that, however, it seems that it is the criteria that deny access to this protection that deserve a revision.