FREE CHAPTER from ‘A Practical Guide to TUPE and Employee Transfers – Core Legal Principles, Tricky Issues and How It Really Works in Practice’ by Dominic Holmes


The first of the two types of “relevant transfer” governed by TUPE is that for which the ARD was originally introduced – and which has formed the basis for transfer of undertakings legislation across the EU for over 40 years.

I will call this the “transfer of an economic entity”, to distinguish it from the more recent UK-specific statutory concept of “service provision change”[1].

Key principles

Regulation 3(1)(a) states that TUPE applies to:

“a transfer of an undertaking, business or part of an undertaking or business situated immediately before the transfer in the United Kingdom to another person where there is a transfer of an economic entity which retains its identity.”

We must look at the constituent elements of that definition in turn, to determine whether TUPE applies in any given scenario. There are five questions to consider:

  1. Is there an economic entity?
  2. Does the economic entity transfer to another person?
  3. Does the economic entity retain its identity in the hands of the transferee?
  4. Is the economic entity situated in the UK immediately before the transfer?
  5. When does the transfer take place?

Economic entity

Regulation 3(2) provides a simple definition of “economic entity” as follows:

“an organised grouping of resources which has the objective of pursuing an economic activity, whether or not that activity is central or ancillary”

This wording mirrors precisely the definition in Article 1(1)(b) of the ARD.

It should be contrasted with the concept of an “organised grouping of employees“, which is used in the context of service provision changes. The distinction is deliberate, and its importance will become apparent.

Further, Regulation 3(4)(a) confirms that TUPE applies to:

“public and private undertakings engaged in economic activities whether or not they are operating for gain”

The scope is wide and it will usually be easy to satisfy this element of the test.

It will cover publicly and privately-owned commercial enterprises of all sizes (regardless of the corporate, partnership or other vehicle that is used), from multi-national plcs with thousands of staff across multiple sites to an individual who employs a single employee in their personal capacity. It can include the NHS, government departments and public sector bodies, educational establishments, temp agencies, charities and not-for-profit organisations.

Any organisation that carries out any type of economic activity will potentially count as an economic entity, even if that activity is not the main purpose of the organisation and regardless of whether it is intended to be profit-making. This means a support function within a larger business (such as marketing, IT or facilities management) can constitute its own economic entity, even though the principal objective of the undertaking as a whole is focused elsewhere.

However, the economic entity must have at least some structure and autonomy, even if it does not have significant assets. This is an important principle.

As we shall see, the approach to determining whether the entity “retains its identity” will differ, depending on how it is found to be constituted in the first place. An “asset heavy” entity which does not transfer any significant assets is unlikely to fall within TUPE, whereas this criterion is much less relevant for “labour intensive” entities, which essentially comprise one or more people doing work with few assets.

In this way, we can see that a cleaning business employing one individual (the frequently cited person with a mop and bucket[2]) can be just as much an “economic entity” as a large manufacturing business with factory premises, machinery, stock, goodwill and intellectual property rights in the design of its products. The underlying purpose of TUPE and the ARD to safeguard employee rights is paramount, regardless of the size of the employer or the number of employees potentially affected.

A useful summary of the case law and relevant principles can be found in EAT’s decision in Cheesman and others v. R. Brewer Contracts Limited[3]. Although decided over 20 years ago in the context of TUPE 1981, the landscape in relation to the transfer of an economic entity (and whether it retains its identity) has not changed significantly. Cheesman is one of the leading domestic authorities in this area.

As Regulation 3(1)(a) confirms, there can also be a transfer of part of an undertaking.

This occurred in Fairhurst Ward Abbotts Ltd v. Botes Building Ltd and ors[4], where a maintenance contract was re-tendered by a local council and split into two separate lots based on geography, with each lot awarded to a different bidder. The Court of Appeal held that TUPE applied in this case. Even though resources were not organised by geography pre-transfer:

“it is sufficient if a part of the larger stable economic entity becomes identified for the first time as a separate economic entity on the occasion of the transfer separating a part from the whole.”[5]

Of course, each case is fact-specific and the decision may have been different if the contract had been split up in a different way, or dissipated across a larger number of new providers.

Fairhurst is arguably inconsistent with the subsequent ECJ decision in Amatori and ors v. Telecom Italia SpA and anor[6], which stated that the ARD would not apply if the entity did not have “sufficient functional autonomy” before the transfer (although this did not preclude Member States from legislating otherwise).

This issue of fragmentation has also become more complicated following the recent CJEU decision in ISS Facility Services NV v. Govaerts and anor[7], to which we return in Chapter 10.

Even if there is an economic entity, the question of whether it retains its identity must still be determined. However, the fact that it may subsequently not remain a distinct entity post-transfer (for example, where it is absorbed into the transferee’s wider operations) does not preclude TUPE from applying[8].

Transfer to another person

In the very first operative provision of the ARD, Article 1(1)(a) states:

“This Directive shall apply to any transfer of an undertaking business, or part of an undertaking of business to another employer as a result of a legal transfer or merger.(emphasis added).

The reference to a “legal transfer or merger” is vague – but perhaps necessarily so, as it needs to cater for different types of transaction under the domestic laws of each EU Member State.

However, the ARD is very clear that a transfer must involve a change of employer. In similar terms, TUPE refers to “a transfer… to another person”[9].

European case law has repeatedly and consistently addressed this point. The clearest exposition is in Landsorganisationen i Danmark v. Ny Mølle Krø[10]:

“The Directive is therefore applicable where, following a legal transfer or merger, there is a change in the legal or natural person who is responsible for carrying on the business and who by virtue of that fact incurs the obligations of an employer vis-à-vis employees of the undertaking.”

TUPE can apply to transfers between two businesses that are entirely unrelated or to an intra-group transfer between two associated companies, with a common ultimate parent. It is prudent to consider whether a simple proposal to move employees to a different entity within the same corporate group might trigger a TUPE transfer, therefore requiring more than a simple letter to employees confirming the change of employer.

As a general principle, the purchase of some or all of the share capital in a company will not trigger a TUPE transfer. Although the ownership of the employing entity may change, the employer itself remains the same.

There is, of course, scope for an exception – for example, where the purchaser assumes de facto control of the employment-related activities previously done by the employer (such as paying staff and managing pensions). The most commonly cited example of this is Millam v. The Print Factory (London) 1991 Limited[11], where the integration of the target company’s business into that of the purchaser was deemed to fall within TUPE[12].

Such cases are likely to be quite rare in practice. Compare Millam to the subsequent High Court decision in ICAP Management Services Limited v. Berry and anor[13], where a share transaction did not result in a TUPE transfer.

A separate decision to move assets or employees in the context of a wider share transaction may still fall within TUPE, even if it does not happen immediately pre- or post-acquisition. This could happen, for example, where staff are moved by the seller into a “Newco” (which is the target company of the share transaction) or where the buyer decides to hive across the target company’s assets to another group entity, as part of an internal reorganisation.

Retaining identity

For TUPE to apply to the transfer of an economic entity, Regulation 3(1)(a) stipulates that it must “retain its identity” in the hands of the transferee. But what does this mean in practice?

The leading ECJ authority is Spijkers v. Gebroeders Benedik Abbatoir[14]. The Spijkers “multi-factoral test” has withstood the passage of time for over 35 years and remains the first point of reference for practitioners advising clients on the application of TUPE. It is worth citing in full:

“12. …a transfer of an undertaking, business or part of a business does not occur merely because its assets are disposed of. Instead it is necessary to consider, in a case such as the present, whether the business was disposed of as a going concern, as would be indicated, inter alia, by the fact that its operation was actually continued or resumed by the new employer, with the same or similar activities.

  1. In order to determine whether those conditions are met, it is necessary to consider all the facts characterizing the transaction in question, including the type of undertaking or business, whether or not the business’s tangible assets, such as buildings and movable property, are transferred, the value of its intangible assets at the time of the transfer, whether or not the majority of its employees are taken over by the new employer, whether or not its customers are transferred and the degree of similarity between the activities carried on before and after the transfer and the period if any for which those activities are suspended. It should be noted, however, that all those circumstances are merely single factors in the overall assessment which must be made and cannot therefore be considered in isolation.”

There has been a substantial body of European and domestic case law since Spijkers which has had to grapple with applying these principles to a multitude of factual scenarios.

It is a highly fact-sensitive exercise, and an exhaustive analysis of these judgments is beyond the scope of this book. Different parties to a transaction will inevitably emphasise or downplay certain factors to suit their own narrative, depending on whether they want a transfer to occur.

For this reason, it is easy to see why determining whether there has been a transfer of an economic entity is one of the most mercurial aspects of TUPE. Save in the most clear-cut of situations, it can be difficult to advise on this issue with complete certainty.

However, there are some guiding principles which provide the outlines of a clear path through what can seem a confusing morass:

  1. Don’t stray too far from Spijkers

The courts have stressed time and again that the Spijkers test is the only one that really matters. All relevant Spijkers factors should be considered in any particular set of circumstances and given appropriate weight, according to the context.

Outside of this, attempts to equate any given factual matrix with a case authority decided on what are perceived to be analogous facts can often be unhelpful.

  1. Establish the type of entity

The weight to be attached to each element of the Spijkers test will differ, depending on the activity carried on[15] – hence the importance of establishing what the economic entity is, in the first place.

A distinction has evolved between the approaches to “asset heavy” entities (where the relevant economic activity requires the use of significant tangible or intangible assets) and “labour intensive” entities (which require people to do the work but perhaps not much else). Manufacturing would typically be “asset heavy” whereas certain types of professional services would not.

  1. Asset heavy undertakings: Consider what assets will
    actually be used?

Where the assets are owned by the transferor and not sold or transferred to the transferee, this does not necessarily preclude a TUPE transfer[16]. However, it is more difficult in such cases to demonstrate that the undertaking has retained its identity.

If the assets used by the transferor are owned by a third-party customer and are then subsequently made available to and used by the transferee, there can still be a TUPE transfer[17]. But these are not hard and fast rules.

To illustrate where the boundary sits, a useful comparison can be made between two European authorities concerning the transfer of public bus services to a new provider. In both cases, the principal tangible assets required to deliver the services (that is, the buses) were not transferred – but there were different outcomes.

In Oy Liikenne AB v. Lisojärvi and Juntunen[18], the transfer of seven local bus routes in Helsinki fell outside the scope of the ARD. Although most of the employees were taken on by the new provider, the buses were not. It leased two buses from the old provider for a few months (pending the delivery of 22 new buses) and purchased some driver uniforms, but that was it. The ECJ concluded that:

“in a sector such as scheduled public transport by bus, where the tangible assets contribute significantly to the performance of the activity, the absence of a transfer to a significant extent from the old to the new contractor of such assets, which are necessary for the proper functioning of the entity, must lead to the conclusion that the entity does not retain its identity.”[19]

Conversely, in Grafe and Pohle v. Südbrandenburger Nahverkehrs GmbH[20], the CJEU ruled that there was a transfer of an undertaking. The tender process required that buses should not exceed a certain age and had to comply with a given environmental standard (which the existing provider’s buses did not). Accordingly, the assets could not have been used by the successful bidder in any event and the decision not to take them on did not prevent the ARD from applying.

This factual nuance allowed the CJEU to distinguish this case from the approach taken in Oy Liikenne. Potentially good news for the two claimant bus drivers (both of whom had over 35 years’ continuous service, before being dismissed by the outgoing provider) – although the matter was sent back to the referring German court for final determination.

  1. Labour intensive undertakings: It’s not all about the people

Regulation 3(6)(b) states that a transfer may take place whether nor not any property is transferred to the transferee by the transferor. This simply confirms the position in Spijkers and subsequent case law, namely that the absence of a transfer of assets will not be determinative of itself.

Where there are no significant tangible or intangible assets involved in performing the activity that constitutes the “economic entity”, the retention of its identity cannot logically depend on the transfer of such assets[21]. Therefore, the other factors identified in Spijkers will assume greater weight. Specifically, in Süzen, the ECJ ruled:

“it must be recognized that such a [labour intensive] entity is capable of maintaining its identity after it has been transferred where the new employer does not merely pursue the activity in question but also takes over a major part, in terms of their numbers and skills, of the employees specifically assigned by his predecessor to that task.”[22]

This principle has been applied by our domestic appellate courts (including the Court of Appeal) in several subsequent cases[23].

On its face, it appears to give significant control to the transferee in determining whether or not TUPE applies, assuming that: (a) there is not otherwise a “service provision change” transfer (see Chapter 3); and (b) the transferor does not choose to redeploy its employees pre-transfer.

For example, the transferee can choose not to employ existing staff, but take on new staff on such terms (and in such numbers) as it wishes, with the transferor footing the bill for any redundancy costs. Or it can inherit the current staff and utilise their experience in performing the activities in question, albeit constrained by the fact that their existing terms and continued employment will be protected by TUPE.

However, in practice, this should be approached with care. The judicial treatment of Süzen in the UK suggests that the importance of a transferee’s decision not to take on staff should not be overstated. For example, in RCO Support Services Ltd v. Unison[24], the Court of Appeal ruled that the transfer of a labour-intensive entity may still fall within TUPE even where the major part of the workforce (in terms of numbers and skills) is not taken on, especially if other Spijkers factors (such as the transfer of customers and similarity of activities) are in play.

It might be argued that a conscious decision to avoid TUPE applying by not taking on staff might, in fact, have the opposite effect. However, as always, there is a danger of being distracted from the multi-factoral Spijkers test which, although not exhaustive, makes no reference to deliberate avoidance measures.

Commercially, the transferee’s choice may be dictated by customer preference, if it is participating in a tender process to take on services. This will, no doubt, be factored into the overall cost of those services. In that regard, the terms of any existing contract between the customer and an incumbent service provider may give a putative transferee access to employee information, to enable it to assess what it might inherit before making its decision and bidding accordingly.

Situated in the UK

Although the economic entity must be situated in the UK, it is possible that some or all of the employees who are in scope to transfer are based overseas. Under Regulation 3(4)(c), TUPE can apply to:

“a transfer of an undertaking, business or part of an undertaking or business (which may also be a service provision change) where persons employed in the undertaking business or part transferred ordinarily work outside the United Kingdom.”

The approach in any given case will depend on the type of undertaking.

For an “asset heavy” entity, the focus will be on the location of the relevant tangible and intangible assets. It is these which will comprise the economic entity and, therefore, determine where it is situated immediately before the transfer. A manufacturer selling its UK-based factory, equipment and stock will likely fall within TUPE, even if there are regional sales employees in other countries.

There will be some areas of uncertainty. For example, assets may be scattered across jurisdictions – perhaps dotted across Europe to support an increasingly remote or mobile workforce, working from home or on the road. Physical assets may be in the UK but intellectual property rights may be registered elsewhere – so it may be necessary to assess the relative importance of the proportion of tangible assets against the value of the underlying intellectual property rights, to identify the relevant undertaking.

For a “labour intensive” entity, the location of the employees will be more important. If a major part of the workforce (in terms of numbers or skills) is based in the UK, TUPE is more likely to apply.

In both cases, if the “economic entity” is in the UK, the transfer may have an impact on employees in other countries. I will return in Chapter 19 to look more broadly at the impact of cross-border transfers – including the circumstances in which an employee assigned to a transferring economic entity but based outside the UK might be able to claim rights under TUPE.

Finally, it is worth noting that the considerations which apply to a service provision change transfer are different. The relevant test in those circumstances is that there must be an “organised grouping of employees situated in Great Britain”[25]. It is an important distinction, particularly if a transfer could potentially comprise both the transfer of an economic entity (under Regulation 3(1)(a)) and a service provision change (under Regulation 3(1)(b)).

Timing of the transfer

In many cases, it will be self-evident when the transfer takes place.

For a business sale, it is likely that the economic entity will transfer on the date the relevant assets are sold to the transferee.

Similarly, if the transfer relates to re-tendered services, it will usually be the date on which the incumbent provider ceases operations and the new provider takes over.

Sometimes, to provide certainty, an asset purchase agreement (or contract for services) will stipulate the precise “effective time” of transfer – for example, 00.01 on the completion date (or services commencement date). This will be helpful in the vast majority of cases, but not conclusive.

There may be circumstances where the position is less clear-cut. For example, the assets comprising the economic entity may be passed to the buyer gradually, or through a succession of transactions. This might potentially result in two (or more) separate TUPE transfers, if each part that is transferred can be considered a stable economic entity in its own right.

If not, Regulation 3(6)(a) ensures that the parties cannot circumvent TUPE by a “drip-feed” of assets or employees, by confirming that a relevant transfer “may be effected by a series of two or more separate transactions”. The question of precisely when the contracts of employment transfer is one of fact.

When appointing a new service provider, there may be a long transition period. There may even be a separate transition services agreement, detailing how the existing activities will be phased out and ensuring a smooth handover as the new services are ramped up. This raises an important question – can there be a gradual transfer of employees over a period of time?

The emphatic answer is no. In Celtec Ltd v. Astley[26], the ECJ stated that the transfer date is:

“the date on which responsibility as employer for carrying on the business of the unit transferred moves from the transferor to the transferee. That date is a particular point in time which cannot be postponed to another date at the will of the transferor or transferee.”

Just as the parties cannot bypass TUPE by agreeing whether it applies, they cannot engineer when it applies. If we return to the statutory wording, we need to consider at what point the relevant undertaking “retains its identity” in the hands of the transferee.

For an asset-heavy entity, it will likely be when most of the core assets required to carry out the relevant activity are transferred. Without these, it is difficult to see how the transferee can become responsible for carrying on the business of the unit transferred (as outlined in Celtec).

For a labour-intensive undertaking, it may be when a substantial proportion of the employees (in terms of numbers or skills) begin working for the new employer (and effectively drag the others who are assigned to the work over with them). Again, without the employees to do the work in an otherwise asset-light business, the transferee may not be able to carry out the relevant activity that is being transferred.

However, these issues need to be considered in their full context. A protracted delay in the principal assets or employees coming across might ultimately be relevant in whether the economic entity retains its identity at all. A temporary cessation of activities will not, in itself, prevent TUPE from applying[27]. However, if activities are suspended altogether for a lengthy period and the post-transfer activities look quite different when resumed, any delay may carry greater weight.

Let’s use a practical example to illustrate this. An Italian restaurant of 15 years’ standing is sold to a new owner and then closed for refurbishment. The employees are made redundant by the seller. The buyer takes on vacant possession, but the sale includes kitchen equipment.

The buyer rips out the kitchen and undertakes a major refit and rebranding exercise. The restaurant re-opens for business four months later as a café bar, specialising in cocktails and Spanish tapas. In those circumstances, it is hard to see how TUPE might apply. Some assets have come across, but several have not – the seller has taken away the tables, chairs and other assets required for serving food and drink to guests. The activities have been put on hold for an extended period and when they resume, there is little similarity with what went on before.

Compare this to a different scenario where the buyer redecorates front-of-house, but re-opens as an Italian restaurant with the same name, telephone number and website, a couple of weeks later. This is much more likely to fall within TUPE – there is value in the intangible assets (which will likely attract back loyal customers of the old business), the suspension of activities has been limited and the undertaking remains similar when it re-opens. It has, quite clearly, retained its identity, albeit under new ownership.

As the case law demonstrates, there are several other ways in which the facts in this simple example can be altered, to illustrate the various shades of grey. It neatly encapsulates the interplay of factors that will be relevant to whether there is a transfer of an economic entity under Regulation 3(1)(a). As ever, the correct analysis starts and ends with Spijkers.


[1]   See Chapter 3

[2]   See, for example, Schmidt v. Spar und Leihkasse der Frühren Ämter Bordesholm, Kiel und Croshagen [1994] IRLR 302 (ECJ)

[3]   [2001] IRLR 144 (EAT), paragraph 10. It has since been approved by the Court of Appeal in McCarrick v. Hunter [2013] IRLR 26 (CA).

[4]   [2004] IRLR 304 (CA)

[5]   Fairhurst, paragraph 32

[6]   C-458/12, [2014] IRLR 400 (ECJ)

[7]   C-344/18, [2020] IRLR 639 (CJEU)

[8]   Klarenberg v. Ferrotron Technologies GmbH C-466/07, [2009] IRLR 301 (ECJ)

[9]   Regulation 3(1)(a)

[10]  C-287/86, [1989] IRLR 37, paragraph 12

[11]  [2007] IRLR 526 (CA)

[12]  See also Jackson Lloyd Ltd and anor v. Smith and ors UKEAT/0127/13 (EAT)

[13]  [2017] IRLR 811 (QBD)

[14]  C-24/85, [1986] CMLR 296 (ECJ)

[15]  See Süzen v. Zehnacker Gebäudereinigung GmbH Krankenhausservice C-13/95, [1997] IRLR 255 (ECJ)

[16]  Cheesman, paragraph 11(vii), citing Allen and ors v. Amalgamated Construction Co Ltd [2000] IRLR 119 (ECJ), paragraph 30

[17]  See for example, Abler v Sodexho MM Catering GmbH C-340/01, [2004] IRLR 168 (ECJ)

[18]  C-172/99, [2001] IRLR 171 (ECJ)

[19]  Oy Liikenne, paragraph 42

[20]  C-298/18, [2020] IRLR 399

[21]  Cheesman, paragraph 11(vi), citing (amongst others) Allen, paragraph 28

[22]  Süzen, paragraph 21

[23]  See, for example, Betts v. Brintel Helicopters Ltd and anor [1997] IRLR 361 (CA)

[24]  [2002] IRLR 401 (CA)

[25]  Regulation 3(3)(a)(i)

[26]  C-478/03, [2005] IRLR 647 (ECJ)

[27]  P.Bork International A/S (in liquidation) v. Foreningen af Arbejdsledere i Danmark C-101/87, [1989] IRLR 41 (ECJ)