A service provision change TUPE transfer occurs when activities carried out by one provider are taken over by a new provider as long as there was, immediately before the change, an organised grouping of employees, the principal purpose of which was to carry out the relevant activities for the client. 

When one provider is replaced by another provider the position is quite simple. But the position can be more complex where a single provider is replaced by multiple providers. In principle, TUPE may still apply, but not where, following the change, the services are fragmented and randomly allocated among new providers.

This was the subject matter of the recent EAT decision in Carewatch Care Services Ltd v Henry.

Sevacare ended its contract with the London Borough of Haringey to provide care for Borough Residents. London Care, Carewatch and other care providers took over. The 17 claimants in the case were employed by Sevacare as Homecare Support Assistants providing care to adults in their homes. The care provided by Sevacare was under contracted packages of care to 168 service users. Care to the service users was delivered by care workers who were employed on zero hours’ contracts. They were asked to take delivery of specified care for a service user, allocated to those service users and placed on the rota maintained by Sevacare. Workers were commonly allocated to particular service users to ensure continuity of care, trust and efficiency of care delivery. Sevacare adopted a regionalised approach so that wherever possible carers worked within one zone and were allocated clients within that zone. Sevacare’s work in Haringey was almost exclusively made up of servicing clients funded by the Council, although there were a small number of private clients who engaged Sevacare on a private basis. Sevacare rarely allocated business to the Haringey team of carers outside Haringey Borough. Sevacare gave notice to terminate to terminate the arrangement. The clients were reallocated to other providers largely on the basis of postcodes. Due diligence was undertaken to establish which carers were allocated to which clients according to the rotas that had been prepared for a six week period prior to the handover. In some cases all of the carers went to the same provider but in other cases there was a split between one or more providers. The Employment Judge concluded there had been a TUPE transfer. 

London Care and Carewatch advanced three main grounds of appeal: (1) the EJ should have found that the relevant activity was so fragmented as to preclude any finding of a service provision change, (2) the EJ erred in concluding that the activities were carried out pre-transfer by an organised grouping of employees which had as its principal purpose the carrying out of the activities concerned on behalf of the client and (3) the EJ erred in concluding that each claimant was assigned to such an organised grouping. 

The EAT found that there was merit in the first two grounds of appeal. First, the EJ should have considered the possibility of fragmentation, which would negative the possibility of a service provision change. The EJ’s finding was that the activity transferred was “the provision of adult homecare to individual service users in accordance with care plans”. But according to the EAT:

“That being so the EJ should have considered whether there was fragmentation of the activity amongst the new providers. There is no evidence that one contractor took on the majority of the work; and in relation to a number of employees it is difficult to establish where the employment should transfer given that various service users went to different contractors. Whilst the Sevacare work generally was organised on a regional basis, post termination the Council-funded work was divided on the basis of both capacity and postcode. It does not appear from the judgment that proper consideration was given to these various factors when the EJ considered fragmentation which, as is agreed, should have been at the stage when she determined whether or not the relevant activities carried out by the original contractor were fundamentally the same post-transfer.”

Nor was the EJ right on the subject of whether there was an organised grouping of employees in Sevacare’s employment. The EJ considered that there was an organised grouping because the principal purpose of the activity was delivering care to service users for whom the Council was responsible. But the fundamental flaw in this approach was that the EJ had confined her consideration to the purpose of such a grouping without first considering whether any grouping existed in the first place and, if so whether it had intentionally been formed. 

Because of these two flaws it was unnecessary to consider the issue of assignment, but the EAT did say as a matter of principle that:

“…when considering whether there was an organised grouping of employees the question is whether “before the change there existed an organised grouping of employees whose principal purpose was the carrying out of the activities for the client”…it follows that the assignment must be to an organised grouping of employees that exists before the change.”

The case was remitted to another employment tribunal for these issues properly to be considered.

The case in particular reminds us there may be a problem in finding a service provision change where an outgoing provider’s work is randomly allocated amongst a panel of new or existing providers. In this regard the case reminds us of the EAT’s decision in Clearsprings Management Limited v Ankers UKEAT/0054/08. This case involved the National Asylum Seekers Service, the function of which was to provide accommodation for asylum seekers. Contracts were awarded to contractors to provide this service. In the North West there were four such service providers, including Clearsprings. On the expiry of the contracts the service was re-let. Three contractors (but not Clearsprings) were appointed. The asylum seekers looked after by Clearsprings were randomly allocated to the incoming contractors. The EAT held that the service was so fragmented on its random reallocation amongst multiple providers that the service provision rules were not engaged.