There is a long-established legal principle that you can only imply an employment relationship in the face of a contract saying something different if it is necessary to do so, i.e. if the found facts of the relationship are not consistent with any other explanation, in particular, worker status or genuine self-employment. Until the Court of Appeal inserted a corrective boot in James –v- London Borough of Greenwich in 2008, the tendency had been for the Employment Tribunals to tot up the aspects of the relationship which looked like employment and then those that didn’t, stick a moistened finger in the wind and then using an unholy and largely unappealable mix of gut-feel and discretion, come down on one side or the other. James took a much stricter view and said that you could only imply an employment relationship if it was necessary to do so, not just because it felt right on balance.
It is however another long-established principle that came off the bench in this month’s EAT decision in Richards –v- Waterfield Homes Limited. This confirmed that the stated (or indeed actual) intention of the parties on that front is persuasive at best but in no sense conclusive, nor enough to justify a finding for or against employment status in the face of bountiful evidence to the contrary. No news there – the surprise is what happened next.
Mr Richards was engaged (to use a neutral term) by Waterfield from 2010 to 2018 as a self-employed contractor paid via HMRC’s Construction Industry Scheme, a particular tax mechanism expressly not available to Schedule E employees. In 2018, Waterfield received advice that its contractor contracts would not withstand HMRC scrutiny and in consequence Richards and his colleagues were all moved to employment contracts. That new contract said that he had continuous service from 2018, but Richards argued that it should be from 2010 when he had first started, i.e. that he had in fact been an employee all along.
The Exeter Employment Tribunal disagreed. The use of the CIS, being closed to Schedule E employees, indicated a clear intention and agreement between the parties that Richards was not an employee over that period, end of story. It was not a sham arrangement. Richards was fully aware of the terms of the CIS scheme as he had been on it previously, and his being in need of a job at the time did not mean that he had had insufficient bargaining power to object if he had wanted. Indeed, the evidence was that he had positively insisted on being paid through the CIS.
Waterfield conceded up front that Richards had been a worker for Working Time Regulations and holiday pay purposes, but the ET decided that he was not an employee for employment protection purposes until 2018. Therefore, he lacked the two years’ service necessary to bring an unfair dismissal claim in relation to the termination of his employment the following year.
On his appeal, the EAT found in his favour. The immediate mystery is why. It did not mention the necessity test under James, nor explain why (given the clear worker status finding made by the Exeter Tribunal) it was appropriate to conclude off its own bat that the facts were consistent with nothing but an employment contract. It did so without even referring the matter back to the Employment Tribunal to look at those facts again, a particularly robust approach given that one of its criticisms of the ET’s judgment was that it did not rehearse sufficiently the facts which it had relied upon, and therefore that the closing state of the evidence heard by the ET was unclear. Nonetheless, the EAT decided that it was clear enough that Richards was an employee that it could make that decision even in the absence of those facts.
It does not appear to me that it was necessarily correct in this. Richards’ contract pre-2018 had included the right to work also for other businesses (though he never did) and he seemingly owned his own van and tools. Neither is a common feature of an employment contract, but even if it were, the fact that a particular term could legitimately appear in a contract of employment does not mean that it could only appear in such a contract. Even the fact that nothing of Richards’ daily work had changed when he became an employee in 2018 means only that the prior arrangements were compatible with employment status, not that there was no other possible explanation for them. The EAT’s decision does not give any details of which specific terms of Richards’ pre-2018 contract it considered to be inconsistent with anything but a full employment relationship. There was in particular nothing as to why it was necessary to go beyond the worker status finding made by the original Tribunal and make Richards a full employee.
It is to be hoped that this decision does not represent the end of the James necessity test for the implication of a contract of employment. A decision that employment status can arise from facts which are merely not inconsistent with that status (rather than those which are only consistent with it) would badly blur the line between employee and worker status. It would potentially open the flood-gates to tens, perhaps hundreds, of thousands of claims by workers both within and outside the gig economy that they are in fact employees. Unfortunately the brevity of the EAT’s own reasoning here gives no obvious basis for argument to the contrary. Therefore it might be that an urgent appeal or another more senior decision on the point would be welcome, ideally to reinstate the James test, but as a minimum to provide clarity on this very sensitive debate.