A basic principle of good employment law, you would think, should be that you can tell an employer when you are caught by it and when you are not. An early fail then for the rules around worker status, in particular following the EAT’s recent decision in Johnson –v- Transopco UK Limited, which appears to bring into express consideration in that question factors over which the employer will rarely have any knowledge, let alone control.
By section 230(3) Employment Rights Act, a worker is (very slightly paraphrased) someone who contracts to do or perform personally any services for another party to that contract whose status is not that of a customer or client or any profession or business carried on by the individual. That requires consideration of two points – first, whether the individual operates a trade or business and second, if he does, whether any given end-user of his services is a customer or client of it. The two are often run together with the implication that if someone operates their own business, anyone he does any work for must necessarily be a customer of it. However, that is an over-simplification. It is entirely possible for an individual to operate a business but separately to supply his services outside it on a basis which would bring him within section 230(3).
The detailed facts of Johnson are not a necessary part of this post. In brief, he is or was a cab driver, part of whose income was earned doing taxi jobs received via a subscription app, Mytaxi, latterly operated by Transopco. For reasons which don’t matter, Johnson was ultimately removed from the app and brought the now traditional gig-economy claim for accrued holiday pay. That required him to show himself a worker of Transopco under Section 230. There he fell down, the Employment Tribunal deciding that he was operating his own business and that Mytaxi was a customer of it, so he could not be a worker.
So far, nothing to see here. However, what drove Johnson’s appeal was how far the ET had felt it appropriate to answer the (i) own-account business; and (ii) customer or client questions by reference not only to the work he had done for Mytaxi but also to that which he did for others. It was Johnson’s contention that all that should have been looked at was the relationship between him and Mytaxi. The Employment Appeal Tribunal rejected this. It said that you could not tell whether Johnson was operating a business by looking at his contractual position with one other entity only. Instead it took the view that the ET had looked at his non-Mytaxi time only to shed more light on the nature of his relationship with it. You had to take into account issues around how much work Johnson did for Mytaxi as opposed to other sources and whether his services for Mytaxi were of the same type as those he provided to other alleged customers of his business. That would inform questions of economic dependence and whether Johnson was providing his services in line with the normal offering or product of his business, or outside it.
Logically this all makes perfect sense until you put yourself in the shoes of the “employer” of an individual who is by written contract with you fully self-employed, but who then turns round when something goes wrong and says that he is entitled to all sorts of rights and protections by way of TUPE consultation, minimum wage, discrimination or holiday pay because he was a worker all along. Then you find that the question of whether your chap is right about that may depend not just on what he does for you and how, but also on what he does for others too. Most gig-economy employers go out of their way not to ask what people working for them do at other times. They consider with some justification that even asking the question may be taken as implying some degree of latent interest or control, all anathema to maintaining the contractual distance necessary to refute a worker status claim.
But suppose they do ask anyway, what can they usefully do with the information they receive? Here, Johnson derived about 15% of his overall income from Mytaxi jobs, not enough for the ET to find any real level of economic dependency. However, if Johnson had worked less hard on sourcing other rides through street hails, etc., that same number of Mytaxi jobs could have been a much higher percentage of his income. Similarly, if instead of doing only taxi work, as the ET found, Johnson more usually operated as a carpenter or gardener, then Mytaxi would suddenly find itself not being a customer or client of that business after all, and so back into worker status all by reason of external factors it could neither know about nor control.
Not only that, but also factors which might wax and wane with the fluctuating fortunes of the individual’s external activities and their impact on the amount of his time which goes on that other work, the proportion of his overall income which comes from it and the diversification or otherwise of that other work. Simplistically, the question could even vary seasonally – in winter, demand for taxis is higher and for garden work much reduced, while in summer the reverse is true. How is the poor employer to know on that basis who is a worker and when? The unions would have a simple answer to this – give worker rights to everyone from the start and the issue goes away. That may or may not be the right thing for wider society given its grossly inflationary effect, but it is certainly not what the law says now.
Here the EAT freely accepted that the very fact-specific nature of the worker status test could lead to different outcomes in different cases for people signed up to exactly the same contractual terms and working in exactly the same way. In a fairly feeble attempt to mitigate this, it did also indicate that these external factors would not make the difference in every case and that some relationships would be clearly worker or non-worker already. However, we are still left with the inescapable conclusion from this case that a gig-economy employer is very much at the whim of its individual associates when it comes to questions of worker status. They can potentially move themselves into and out of that at their own discretion through variations in their external activities with no obligation to tell the end-user of the transition either way.
By that opening measure, therefore, Section 230(3) is not good law. What can be done about that is a separate question and one to which neither statute nor case law has yet provided a satisfactory answer. So if you are an operator in the gig economy, whether through an app or a written contract and a mobile phone, where does this leave you? Should you now check out the other remunerative activities of your would-be associates or not? Probably not. That looks more like a vetting, which is an unattractively employment-y exercise, and more to the point, whatever information you obtain on their joining can be out of date within weeks and whatever your associates’ work activities have been in the past will not affect their status at the moment of their claim anyway.