ISS Facility Services – v – Govaerts was a European Court of Justice case in March 2020 concerning what happened where the work being done by a group of employees was split up and all sold or contracted off in different directions.

Traditionally the UK view has been that the employee would go with whichever bit of the contract he was most assigned to.  If he worked across them all such that none took the majority of his time, then he would not be caught by TUPE and would stay behind to be redeployed or made redundant by the transferor at its cost.  This had the attraction of simplicity but it could undeniably lead to some anomalous outcomes.

Take the example of a business activity of ten people servicing one client 70% and another 30%.  If the client work is sold, contracted out or lost to two replacement contractors, what happens?  Consider the options:

(i) If seven of the employees are assigned to work full-time on the bigger client and three on the smaller, then they all go with their relevant client.

(ii) However, if all ten were assigned to work 70% on one client and 30% on the other, then under the traditional UK view, all of them would probably go with the bigger client and none with the smaller. The transferee of the bigger client would need seven people but inherit ten, while the company servicing the smaller one would require three employees but get none.

(iii) And if there were no assignments by the employer of its staff to one client or the other but at the point of the transfer those just happened to be the proportions in which they spent their time, there would be no organised grouping of employees attached to either client (as required by the service provision change rules in TUPE) and so they would most likely all stay behind with the transferor. Then the two transferors who needed staff wouldn’t get any and the transferor who didn’t need those staff any longer would be left with all of them.

In the circumstances it is not surprising that pragmatic allocations of affected staff between outgoing clients or business activities have sometimes been agreed to ensure that as many employees as possible end up with someone who can use them.

However, Govaerts disagreed with that traditional position.  It said that the contract of employment of an employee distinct bits of whose role went off to several different transferees could also be divided into equivalent “mini-contracts” with more than one transferee at the same time, each referable to the specific tasks going to that transferee.  Obviously this could easily become a practical and administrative nightmare, the basis of the original UK position, and still more so if the two transferees were competitors – then the employee might at the same time be obliged to work for them by TUPE and be forbidden to do so by contract. It would also not be OK for that splitting up of the contract to worsen the employee’s working conditions or otherwise undermine the fundamental protections of TUPE. Govaerts confirmed that in such cases, TUPE would still bite and so it would be the transferees who would be responsible for the costs of termination, not the transferor.

This is a splendid triumph of legal theory over practical reality, perhaps what might happen if you take a legal proposition in the ECJ bar one night as an academic exercise and then stretch it beyond its natural breaking-point. “Hard cases make bad law” says the old legal adage, and here is proof positive of that. We have discussed this case internally at some length and been unable to see how it will work in practice, it dawning on us only slowly that perhaps the mistake is to assume that it has to.  The approach taken by many UK employers engaged in business transfers and outsourcings since the decision came out consequently seems to have been to ignore the case in the hope that if no one talked about it, Govaerts might just go away.

Sadly that has proven not to be correct, and the Scottish EAT in McTear and Others – v – Bennett and Others has just shed a light on what the case might mean in the UK.  The detailed facts are complex but not that relevant.  Essentially a local authority moved from a single outsourced fitter of kitchens to its housing stock (Amey) to two (McTear and Mitie). Both provided the same services and Amey’s former territory was divided less than helpfully into north North Lanarkshire and south North Lanarkshire. Amey had a gallant stab at dividing up its relevant workforce between McTear and Mitie based on where they had carried out their most recent jobs for it, but it was pretty rough and ready.  It also assumed that all the Amey employees had to go to one or the other contractor, even where their activities had been split sometimes quite finely across the two regions.

However, not all the Amey employees found homes with one of the new contractors, nor did McTear and Mitie respect their prior contract terms in all cases. About 20 of them claimed unfair dismissal.  To determine who they should properly be suing, the Employment Tribunal sought to decide first where each of them should have gone if TUPE had been applied, and it relied upon Amey’s finger-in-the-wind assessment to make sure that they all ended up with one of the two transferees.

Mitie and McTear appealed on the grounds that between them they had been saddled with all the former employees whereas if each  had been looked at individually, it was possible that at least some of them might have remained behind as Amey’s problem, not theirs.

After that appeal was lodged but before it was heard, along came Govaerts.  Although that case had concerned a business transfer, there was no serious argument raised in McTear that it did not also cover service provision changes as here, so therefore it did. The Scottish EAT found that this meant that the ET should not have assumed that all Amey’s employees had to go to one of McTear or Mitie, either in full or at all.

That meant in turn that the “forced allocation” approach taken by Amey in good faith to protect its employees could not trump a proper analysis of whether the activities of each affected employee were split cleanly and clearly enough that he could now legitimately be said to be subject to distinct contracts with each transferee.  The matter was therefore sent back to the Employment Tribunal to decide on an individual analysis which employee should have ended up where, including the possibility that some would end up not transferring at all. We don’t know the answer to that yet, but the damage is done anyway.

The principal legal difference which this case makes in the UK is to reduce the number of people in a business transfer or service provision change who won’t be caught by TUPE.  It will no longer be an obstacle to their being transferred that they can’t point to the clear majority of it going to one particular transferee.  If they are then transferred, they become the responsibility of the transferees in the same proportions as those companies inherited that employee’s former role. It also seems to override the TUPE service provision change requirement that the employee be wholly or mainly “assigned” by the employer to the part of the business being moved on. Govaerts makes no direct reference to assignment, probably because it is an ECJ case and the SPC rules are a domestic UK insertion not featured in the EU Acquired Rights Directive which that case was ultimately applying. Instead it applies to transfer the contract of employment “in proportion to the tasks transferred”. Hopefully the ET will provide some view on which concept takes priority, express assignment or simple weight of tasks at the point of transfer.

The principal practical difference which Govaerts makes is of course that in few cases only is there any realistic possibility of this working at all, as a minimum without a wholly improbable degree of liaison and co-operation between incoming transferees.  Bear in mind that those transferees have no interest in helping each other nor any contractual obligation to do so.  And the more the transferees, the bigger the problems.  Take the simple example of just two, as in McTear, and assume that an employee’s former activities are found to be divisible 80:20 between them.  Therefore the employee effectively ends up contracting the equivalent  of one day a week for one company and four days a weeks for the other.  But which one day?  What if both companies want him on the same day?  What if (as here) they operate in different places? Who is counting his WTR hours?  On what days does he take holidays?  What if the individual transferees cannot accommodate someone working part time because of their shift arrangements?  Or one can and the other can’t? What happens if the two are direct competitors? How does the new employer take steps to resolve the 20% of a grievance or disciplinary matter which is suddenly deposited in its lap, especially if the 80% employer decides not to bother?

In the great majority of cases it therefore seems inevitable that there will be a worsening of his working conditions in serving two or more masters, and the employee will end up losing part or perhaps all of his previous role which a more pragmatic or blunt-instrument approach might have allowed him to keep with one transferee or the other. To claim full compensation he will then have to pursue as many different transferees as took over any part of his previous responsibilities, almost however small.

The scope for satellite litigation between and at the cost of the transferees seems significant.  What proportion of the employee’s job did each take over?  How do you measure that, and over what period? Is it by hours? By value? By importance? And if it takes an ET case a year to get to hearing, even leaving aside any appeals, what is the position of the employee in the meantime, potentially employed to an extent he can’t determine by companies he doesn’t know with claims against each which he can’t quantify. The only realistic outcome is that he will do what most of the claimants did in McTear and just sue absolutely anyone he can find – both all the incoming and outgoing contractors and the local authority end-user —  and then stand back and let the ET sort out the resulting cat-fight between respondents.

Both transferees and transferors will also have to keep these cases in mind whenever there is a business sale or service provision change under TUPE . Effectively transferors to more than one transferee will be doing multiple transfers all at the same time, each with its own separate obligations by way of Employee Liability Information, information and consultation and  obtaining measures information. Transferees acquiring only small parts of a business or activity may now find themselves inheriting bits of rights and obligations under employment contracts they weren’t expecting and can’t do anything useful with, and having to pay to clear them up. They may also wish to challenge more robustly the transferor’s assertion that the employees are substantially dedicated to the part of the business they are acquiring. Previously as transferee, once you accepted that the employees were at least 70% committed to your bit, you would generally have to roll over and take them on. By contrast, there may now be an active benefit to a would-be transferee in pushing back on every percent of claimed assignment to those tasks. If another transferee took on as little as 10% of the activity, that could still represent a very material saving in severance costs.

And there is no logical reason why there should even need to be multiple transferees – on the face of it, though it was not discussed in McTear, why could a single buyer or contractor taking on, say, 80% of the employee’s job with the rest clearly remaining where it was, not look back to the transferor for a 20% contribution should it all go toes-up on the grounds above? It will be correspondingly important for any prospective transferee to have a fresh look at its standard due diligence enquiries and warranties/indemnities in business sale and outsourcing agreements with these cases in mind.