Extending Statutory Protection to Action Short of Dismissal for Participating in Industrial Action: The Supreme Court on the Right to Strike in Secretary of State for Business and Trade v Mercer

Abstract

1. INTRODUCTION

Secretary of State for Business and Trade v Mercer is immediately a landmark judgment for the Supreme Court finding that a provision in the Trade Union Labour Relations Consolidation Act (‘TULRCA’) 1992 violated the right to strike,¹ under freedom of assembly and association, in article 11 European Convention of Human Rights (‘ECHR’).² Only a few years earlier, according to Court of Appeal jurisprudence, such a finding seemingly would have been unthinkable. Lady Simler, writing for a unanimous Court in Mercer, found that the ‘lacuna’ in section 146 Trade Union Labour Relations Consolidation Act (‘TULRCA’) 1992 as is currently construed, which means there is no legal protection against detriment short of dismissal for taking part in industrial action, falls short of the requirements of article 11 ECHR. While the Supreme Court agreed with the Court of Appeal that a Convention-compliant re-interpretation of section 146 TULRCA would not be possible under section 3 Human Rights Act (‘HRA’) 1998; unlike the Court of Appeal, the Supreme Court granted a declaration of incompatibility under section 4 HRA. To underline how rare declarations of incompatibility under the HRA are in practice, this was the first declaration of incompatibility, regardless of the substantive area of law, in some 3 years.³ The onus now falls on Parliament and the executive to legislate, if they so choose, to remedy the violation of article 11 as found.⁴

The substantive part of this note proceeds in three parts. It will be suggested that Mercer is significant for two main reasons: first and foremost, the Supreme Court’s recognition of, and finding of a violation of, the right to strike and, second, practically in relation to the workings of the HRA 1998 as regards the scope of, and interplay between sections 3 and 4. Finally and in the third part, while the Supreme Court, and before that the Court of Appeal, drew a distinction between public authority workers and private sector workers when interpreting Strasbourg jurisprudence in this respect, it will be argued here that it is important that if and when the new Parliament and Government legislate to remedy this violation, any distinction between public and private sector workers should definitely not be carried forward into any remedying legislation.

2. FACTS

Ms Mercer was a union representative when a support worker for the respondent, Alternative Future Group (‘AFG’) Ltd: ‘a health and social care charity providing a range of care services across the North West of England’.⁵ Her trade union, UNISON, had organised industrial action in the form of a series of strikes in relation to the withdrawing of ‘top up’ payments for sleep-in shifts at AFG’s care homes.⁶

Ms Mercer was involved in ‘planning and organising’ these strikes.⁷ Once the strikes had commenced, Ms Mercer was suspended. As summarised by President Choudhury in the Employment Appeal Tribunal (‘EAT’) in this case: ‘she was told that this was because she had abandoned her shift (presumably to take part in the strike) on two occasions without permission and that she had spoken to the press without prior authorisation’.⁸ The suspension was lifted, but workplace disciplinary action proceeded, and Ms Mercer was given a first written warning for leaving her shift, albeit that sanction was overturned on appeal. A subsequent internal grievance lodged by Ms Mercer was rejected. Ms Mercer was paid her basic rate of pay during the period of suspension; however, in the wording of Lady Simler, she ‘received nothing for the overtime she would normally have worked’.⁹

The litigation in Mercer stemmed from a preliminary hearing as to the scope of section 146 TULRCA 1992,¹⁰ therefore all the facts were only ‘assumed’ (and possibly ‘disputed’).¹¹ In each of the judgments, it is emphasised that the facts were not yet established and it was not possible to pronounce on, for example, ‘AFG’s motives in suspending and pursuing disciplinary proceedings against the appellant, nor about the reasonableness or proportionality of its actions’.¹²

A notable feature of this litigation was that Ms Mercer was also a union representative, and as above, active in ‘planning and organising’ the industrial action. Another former President of the EAT, Lady Simler, authored the judgment for the Supreme Court in Mercer. It is notable and refreshing that the Supreme Court’s judgment in Mercer referred to interpretations by the International Labour Organisation (‘ILO’), albeit not on the point about to be mentioned. For the ILO, it is a particularly serious breach of freedom of association to subject trade union organisers to sanctions for ‘leading a legitimate strike’.¹³ Indeed, the same might be stated for the European Court of Human Rights, with the Strasbourg Court highlighting the special ‘trade union context’ when sanctions were applied to a trade union representative in Straume v Latvia.¹⁴ Lady Simler referred at the beginning of her judgment to Ms Mercer’s role as a union representative, however Ms Mercer’s special status as a union representative was not referred to again later in the judgment.¹⁵ As will be discussed below, the Court of Appeal and Supreme Court in Mercer differentiated as to the level of protection required under article 11 ECHR for public authority and private sector workers with regards to employer-imposed sanctions for taking part in industrial action. There is, nevertheless, surely a strong argument, to use the later wording of the judgment, that even a ‘minimal’ sanction should have been sufficient to constitute a violation of article 11 for a trade union organiser in respect of a ‘lawful’ strike, whether they work in the public or private sector.

3. DECISIONS

Section 146 TULRCA 1992 provides protection for ‘detriment on grounds related to union membership or activities’. Only part of section 146 was pertinent in Mercer: detriment connected to ‘taking part in the activities of an independent trade union’, the latter which must be at an ‘appropriate time’. An ‘appropriate time’ is defined in section 146(2). Lady Simler in the Supreme Court in Mercer took the opportunity to observe that this provision was ‘originally derived from article 1 of ILO Convention No 98 1949 on the Application of the Principles of the Right to Organise and to Bargain Collectively’.¹⁶

Section 146, as relevant, reads as follows:

• ‘(1) A worker has the right not to be subjected to any detriment as an individual by any act, or any deliberate failure to act, by his employer if the act or failure takes place for the sole or main purpose of –

  • (a) ...

  • (b) preventing or deterring him from taking part in the activities of an independent trade union at an appropriate time, or penalising him for doing so…’

As also noted by Lady Simler, ‘the remedy for a breach of section 146 is a complaint to an employment tribunal under section 147. If the tribunal finds that the complaint is well-founded, it makes a declaration to that effect and may make an award of compensation (including for injury to feelings) under section 149 of TULRCA’.¹⁷ While the Supreme Court briefly alluded to the concept of causation,¹⁸ it should also be noted that for a claim to be successful under section 146, the relevant standard is that it must be the employer’s ‘sole or main’ purpose to prevent, deter or penalise a worker from taking part in the activities of an independent trade union, with case law in this respect setting a high threshold, distinguishing—sometimes strictly—between ‘purpose’ (required) and ‘effect’ (not enough).¹⁹

In each of the four stages of this litigation (employment tribunal, EAT, Court of Appeal and finally Supreme Court), it was agreed that section 146 TULRCA, as currently construed, does not cover detriment, short of dismissal, for taking part in industrial action. To use as an example the facts of Mercer: if Ms Mercer had been suspended for attending a trade union meeting, this detriment would be covered by section 146 (presuming the sole or main purpose test were to be met).²⁰ Even if the meeting were about forthcoming industrial action, this would most probably be covered, even without referring to previous authorities.²¹ The industrial action at that point presumably has not commenced, with reference to ‘taking part in industrial action’. With reference to attending a union meeting, while the relevant cases tend to involve union representatives and Deakin and Morris’ Labour Law refer to the concept of ‘authorisation’,²² attending as a member should also suffice.²³ In comparison, when Ms Mercer was suspended for participating in industrial action, as section 146 is currently construed, this was found not to be covered. (Dismissal is a different situation again, and will be discussed below.) There is a block to section 146 so far as taking part in industrial action is concerned. This is even though, in the words of a different case: ‘in ordinary language, and the discourse of industrial relations, industrial action called by a trade union is a trade union activity’.²⁴ The various decisions in Mercer cited two main reasons why, despite the ease with which industrial action could also fit within section 146, industrial action is not included.²⁵

First, there is the concept of an ‘appropriate time’. ‘Normally’ industrial action will take place within working hours, at a time when the employer has not ‘consented’ to the union activity, although as noted in the judgments, there are limited exceptions.²⁶

Second, there is the reason highlighted in Drew v St Edmundsbury Borough Council: an authority relied upon relatively heavily in Mercer.²⁷ Dismissal for taking part in industrial action is regulated in a separate section of now TULRCA 1992, at Part V. If an employee is dismissed for taking part in industrial action and that is considered taking part in the activities of an independent trade union for the purposes of ‘sibling’ provision section 152 TULRCA, this would mean that the dismissed employee would now fall under two, potentially conflicting provisions.²⁸ Section 152 TULRCA 1992 in this respect will likely be more generous than the unfair dismissal provisions in Part V TULRCA on industrial action.²⁹ The employee would benefit from automatically unfair dismissal protection without needing to show either the industrial action was protected (section 238A) or, as was the extent of legal protection at the time of Drew, the dismissals were selective (section 238). Accordingly, the reasoning in Drew went that dismissal connected to industrial action is regulated in a separate and special part of TULRCA (moreover, with specific limitations attached) and hence the limit to now section 152 TULRCA 1992. In Mercer, the reasoning from Drew was also interpreted to apply to section 146 as otherwise the same wording (‘taking part in the activities of an independent trade union’) would be interpreted differently in the same Part of TULRCA.³⁰

Each of the different judges in Mercer then agreed that section 146 does not, as construed, cover detriment on the grounds of participating in industrial action. Each of the judgments also agreed that it is, or likely would be, a breach of article 11 ECHR if no legal protection whatsoever is provided for detriment on the grounds of taking part in industrial action. Although it was briefly argued otherwise, the different judges also agreed that there are currently no other viable legal mechanisms to protect against this type of detriment, whether under statute or at common law.³¹

For the employment tribunal, Court of Appeal and Supreme Court, it would not be possible to interpret section 146 using section 3 HRA 1998 to remedy this gap. In comparison, for the EAT only, it would be possible to use the ‘strong interpretative obligation’ required under section 3 HRA, to rectify section 146.³² Choudhury P in the Employment Appeal Tribunal suggested reading in a new subsection, using section 3 HRA, to the definition of at ‘an appropriate time’ at section 146(2):

• ‘(c) a time within working hours when he is taking part in industrial action’³³

The Court of Appeal and Supreme Court disagreed with the suggested addition. Lady Simler cited two reasons. First, it would go against the ‘grain’ of the legislation to utilise section 3 HRA in these circumstances.³⁴ The second argument, more powerfully, by both the Court of Appeal and Supreme Court, was that to utilise section 3 in these circumstances, would be to cross the impermissible line from interpreting into legislating.³⁵ There were said to be difficult policy decisions involved, so that it would be more appropriate for Parliament to rectify this gap than judges.³⁶ For example, with reference to Choudhury P’s suggested formulation, the latter would potentially provide wider protection than under Part V TULRCA as there was no reference to ‘lawful’ or ‘protected’ industrial action. Indeed, by the time of the Court of Appeal’s hearing, counsel for Ms Mercer now agreed that narrower wording should be utilised, to link any wording to the concept of ‘protected’ industrial action at section 238A TULRCA:

• ‘(c) in respect of a detriment short of dismissal, a time within working hours when he is taking part in protected industrial action within the meaning of section 238A(1)’³⁷

Instead, the Supreme Court agreed with the Court of Appeal and initial employment tribunal that it was not appropriate to use section 3 HRA. For example, even with the reformulated suggested additional section 146(2)(c): would protection against detriment also end presumptively after 12 weeks, similar to section 238A?³⁸ Even leaving aside specific questions about the scope of section 238A TULRCA, other questions remained as to whether all possible detriments would qualify for this purpose, with discussion of the hypothetical of being passed over for a promotion or a discretionary bonus being refused.³⁹

Where the Supreme Court disagreed with the Court of Appeal was on the subsequent consideration of section 4 HRA, on a declaration of incompatibility. The Court of Appeal had concluded that, even though there was a breach of article 11 ECHR by virtue of this gap whereby detriment for taking part in industrial action does not have legal protection; even though section 3 HRA 1998 could not be used in the circumstances; nevertheless a declaration of incompatibility was also not to be granted. This was because there was a ‘lacuna in the law rather than a specific statutory provision which was incompatible’, plus ‘the extent of the incompatibility is unclear and the legislative choices are far from being binary questions’.⁴⁰

By contrast, at the Supreme Court, a declaration of incompatibility was granted, although Lady Simler was keen to emphasise that the declaration of incompatibility did not invalidate or effect current legislation or indeed even mandate Parliament and the executive to take remedial action.⁴¹ A declaration of incompatibility under section 4 HRA would rather be an instruction, albeit a strong instruction, to Parliament to look at this area:⁴² ‘to decide whether to legislate and, if so, the scope and nature of such protection’.⁴³ As such the following declaration under section 4 HRA was granted:

that section 146 of TULRCA is incompatible with article 11, insofar as it fails to provide any protection against sanctions, short of dismissal, intended to deter or penalise trade union members from taking part in lawful strike action organised by their trade union. ⁴⁴

4. THE ‘LACUNA’ IN MERCER

Various statutory provisions have already been mentioned: as well as section 146 TULRCA 1992, also section 152 on dismissal ‘on grounds related to union membership and activities’, plus sections 238 and 238 TULRCA 1992. An excerpt from section 146(1)(b) was provided above. It is useful not only to reflect further on how these different and complicated statutory provisions fit together, but also to amplify on what exactly the ‘lacuna’ in Mercer was. The wording of ‘lacuna’ was used frequently in the various judgments in Mercer.

In Part III TULRCA 1992, there is a package of measures providing protection against union-related discrimination: section 137 on refusal of employment on grounds related to union membership, section 146 on detriment, sibling provision section 152 on dismissal (only where the individual is an employee), sections 145A and section 145B on inducements (in response to Wilson v UK).⁴⁵ Indeed, the ‘package’ of measures already present potentially contributes to making it even more striking that there is not currently a provision providing for detriment against participation in industrial action. Somewhat confusingly, the scope of each provision is not always the same: for example, section 137 only mentions union ‘membership’ rather than also union activities and services; the personal scope of these provisions also differ. Then there is Part V TULRCA which contains the specific unfair dismissal provisions which relate to dismissals for taking part in industrial action: section 237 when industrial action is unofficial (essentially removing unfair dismissal jurisdiction altogether), section 238 providing protection against selective dismissal when industrial action is ‘non-unofficial’ and,⁴⁶ finally, section 238A providing that dismissal of an individual striker will be automatically unfair at least for the first 12 weeks of industrial action if the industrial action is ‘protected’ (or, in Lady Simler’s preferred wording, ‘lawful’).⁴⁷ Lady Simler in Mercer helpfully listed the many various procedural and substantive requirements for industrial action in the UK to be ‘lawful’: these are requirements that the union must meet when organising and carrying out industrial action.⁴⁸ In this way, the individual striker’s level of unfair dismissal protection under TULRCA is contingent upon their trade union’s actions. However, if industrial action is not ‘protected’, so long as the union has not disavowed the industrial action or if there is no union involvement, weaker residuary section 238 protection will still be available.⁴⁹

The ‘lacuna’ as identified in Mercer was the following: due to the overlap with these provisions in Part V TULRCA, section 152 on dismissal could not be interpreted to cover taking part in industrial action; if section 146 is to be interpreted conterminously with section 152 as in the same Part of TULRCA; with this interpretation, effectively there is a missing provision or missing subsection. If section 146 cannot be interpreted to cover participating in industrial action, there needs to be a new subsection or section in Part III or—if industrial action is to regulated exclusively in Part V—an equivalent in Part V on detriment. The result is that an employee may be protected either by sections 238A or 238 if they are dismissed for taking part in industrial action, however they have no redress if they suffer detriment short of dismissal such as, in Mercer itself, suspension or, as another possible example, demotion.⁵⁰ In addition, a so-called ‘limb (b) worker’ ‘(that is to say, a worker who is not an employee but who nonetheless enjoys protection within the wider definition of “worker”)’ will have no protection whatsoever for any detriment or dismissal for taking part in industrial action as a non-employee would be compelled to bring a dismissal claim instead as a detriment claim under section 146.⁵¹ The Supreme Court in Mercer dismissed existing alternative options for legal protection, such as a claim under the Blacklisting Regulations 2010, either as unrealistic or inapplicable.⁵² As stated by Lady Simler, the effect of this gap is: ‘seen in this way, section 146 both encourages and legitimises unfair and unreasonable conduct by employers’.⁵³

One criticism might be that this lacuna was not necessarily inevitable when a set of facts such as in Mercer arose. All the different courts in Mercer readily accepted that the reasoning in Drew would be applicable also to section 146.⁵⁴ While no counsel challenged Drew, tentatively it might be suggested whether this could have been queried more.⁵⁵ The issue in Drew was that, if decided otherwise, the claimant could have chosen whether to bring a claim under now section 152 or under Part V TULRCA, when Part V TULRCA has additional requirements (such as showing industrial action is protected under section 238A or showing that there have been selective dismissals under section 238). When it comes to detriment, there is however no other provision whereby a claim could be brought. As noted above, the different sections under Part III already have differing coverage: there may not necessarily be an insuperable reason why sections 146 and 152 could not be interpreted differently in this respect. Although it is a valid observation that the same wording of ‘taking part in the activities’ is being used in both sections, at least in practice the same wording of ‘membership’ is potentially being interpreted differently in the same Part of TULRCA.⁵⁶ Against this however, even if Drew were not determinative, this does not fully answer the concern about the statutory definition of at an ‘appropriate time’ under section 146(2). Another consideration would be that if section 146 and Part V were to be entirely decoupled, this could lead to, as identified by the different courts, the ‘surprising’ result that a ‘worker’ would be more strongly protected in the sense that the twelve weeks time limit under section 238A applicable to dismissed ‘employees’ would not be applicable to workers in the same circumstances.⁵⁷ A different response—as noted shortly below, in a related context—would be to question the 12-week limit under section 238A.

The Supreme Court granted a declaration of incompatibility in Mercer. While obviously responding is a matter for Parliament, there are immediately two ways in which the lacuna might be filled. If Parliament agrees with the Court of Appeal and Supreme Court that there are indeed multiple policy decisions involved, an entirely new statutory provisions or provisions might be added, more readily to deal with these multiple policy questions; as opposed to Choudhury P’s suggestion and Mercer’s suggestion more simply to add a subsection to section 146(2). On the other hand, if Parliament were to follow Mercer’s suggestion and link any new protection to section 238A: as noted by counsel for Mercer, section 238A itself answers some of these policy questions: chiefly, in deciding what type of industrial action would be covered.⁵⁸ In addition, while the Court of Appeal was concerned that the existing concept of detriment without more would not be sufficient to deal with two hypotheticals posed (refusal of a discretionary bonus and passing over an employee or worker for promotion), this would seem largely unconvincing.⁵⁹ Employment tribunals are well versed in identifying what is and what is not a relevant detriment from multiple other statutory references to detriment in the employment sphere, not least under the Equality Act 2010.⁶⁰ It was also agreed by all that withholding of pay during a strike would not count as a detriment,⁶¹ although the more difficult question, predictably, would be disproportionate pay deductions. The latter could, however, still be a difficult question for the courts even if there were instead to be new legislation.⁶² There is, however, possibly a different concern if this route were to be taken.

While section 238A likely may not be problematic under article 11 ECHR, this not to say it is without criticism. As above, an individual striker’s protection under section 238A is contingent on their union satisfying multiple procedural and substantive requirements. When evaluating compliance with article 6(4) European Social Charter, the European Committee of Social Rights (‘CESCR’) has repeatedly criticised the scope of section 238A.⁶³ The CESCR criticises, among others, the arbitrary 12 weeks presumptive limit.⁶⁴ As an additional observation here: while section 238A lists 12 weeks, as a comparison, the Trade Union Act 2016 introduced an expiry to industrial action ballots as presumptively 6 months.⁶⁵

5. RECOGNITION OF, AND VIOLATION OF, THE RIGHT TO STRIKE

Mercer is immediately a landmark judgment for the Supreme Court finding that a provision in TULRCA violated the right to strike under article 11 ECHR. In comparison to the Court of Appeal’s finding that there ‘may’ have been a violation,⁶⁶ more definitive language was used in the Supreme Court: ‘does put the United Kingdom in breach’.⁶⁷ The Court’s judgment in Mercer follows on from the conclusion, albeit in obiter, only a few months earlier by the Supreme Court in the Deliveroo case, albeit in obiter, that Strasbourg jurisprudence on article 11 does not include a right to compulsory collective bargaining.⁶⁸

It was only in 2009 that the Court of Appeal infamously dismissed the right to strike in the UK as nothing more than a slogan or legal metaphor.⁶⁹ It deserves to be highlighted how far the appellate courts have travelled, from an admittedly low starting point in Metrobus Ltd v Unite the Union to the Supreme Court in 2024 not only recognising that the right to strike is included within article 11 ECHR but, moreover, also finding a violation of article 11 with regards to the right to strike in domestic law. While there is a valid difference between Mercer as involving the treatment of an individual striker in comparison to RMT as a ‘collective’ strike case with a wider margin of appreciation, still the UK Supreme Court potentially in this respect went further than the European Court of Human Rights itself in RMT v UK by also finding a violation of article 11 ECHR rather than simply finding engagement of article 11 but no breach.⁷⁰ It will be recalled that in RMT v UK, while the Strasbourg Court took the important step of declaring the right to strike as ‘clearly protected’ within article 11,⁷¹ at the same time, a complete prohibition on secondary action in the UK, at the extreme end of the spectrum among European States was deemed to be proportionate and not to be a violation of article 11 ECHR.⁷² In comparison, in Mercer, the Supreme Court was willing to find a violation of article 11 and also to make a rare declaration of incompatibility in respect of this violation.

Indeed, it is perhaps a reflection of how firmly the right to strike is now settled within article 11 jurisprudence that there was no part in the Supreme Court’s judgment explaining the establishment of the right to strike within Strasbourg jurisprudence. The right to strike is, of course, implicit rather than explicit within article 11. In Mercer: RMT was discussed, however not for this point, but rather for whether the ‘right to strike, including taking secondary strike action, should properly be seen as core or essential elements of trade union freedom’.⁷³ That the Supreme Court now referred to the right to strike as recognised, and also ‘protected’, in domestic law effectively in passing is notable in itself.⁷⁴

While the different judgments in Mercer show the massive direction of judicial travel to not only recognising, but also finding a violation of the right to strike with regards to the lacuna in section 146 TULRCA, Mercer simultaneously, however, potentially also highlights the need, ideally, for further clarity in Strasbourg jurisprudence on the right to strike.

The Supreme Court in Mercer concluded the right to strike does not, at this moment in time, constitute an ‘essential’ element of article 11,⁷⁵ and that the language of core and essential is ‘interchangeable’.⁷⁶ For Lady Simler, this meant that the right to strike, as non-essential and thereby non-core, leaves the State with a wide margin of appreciation when deciding how to regulate.⁷⁷ In comparison, for counsel for Ms Mercer, ‘core’ and ‘essential’ in RMT were not interchangeable.⁷⁸ The core to accessory spectrum applies to each element of article 11 (whether the right to strike or whether the right to collective bargaining and so forth). The argument went as follows. In Mercer’s case, this was direct or primary industrial action, which would fall at the core end of the spectrum (in comparison to secondary action, after RMT), hence a narrower margin of appreciation would apply.⁷⁹ That it is possible to interpret the Strasbourg jurisprudence, if only at the more detailed level so differently, shows the need, ideally, for further clarity. For example, to build on Bogg and Ewing’s original criticism after RMT v UK, if introducing a core to accessory spectrum, it is necessary also to introduce some rationale to explain how elements are assigned within that spectrum.⁸⁰ In Mercer, further clarity is needed as to whether ‘core’ and ‘essential’ are interchangeable or not, how the core to accessory spectrum applies as to these facts (if the distinction is between primary and secondary action, or between type of employer, or both), and on Lady Simler’s distinction under article 11 between the State as employer and as ‘regulator’.

6. DECLARATION OF INCOMPATIBILITY

It is rare in practice for a declaration of incompatibility to be granted under the HRA at section 4.⁸¹ This was the first declaration of incompatibility in approximately 3 years, in relation to any area of law.⁸² In terms of previous declarations of incompatibility broadly or narrowly connected to the employment field, this is potentially the fifth declaration since the advent of the HRA.⁸³ As repeated in Mercer, declarations of incompatibility are ‘described as a measure of last resort’.⁸⁴ This is ‘because a Convention-compliant interpretation under section 3 of the HRA is seen as the primary remedial measure’.⁸⁵

On the one hand, it seems quite strong that the Supreme Court issued a declaration of incompatibility.⁸⁶ The so-called measure of last resort was reached. For example, declarations of incompatibility have been described as conflictual. A declaration of incompatibility is essentially courts stating that they cannot work with and within legislation, despite the instruction in Ghaidan v Godin-Mendoza to try always to work with legislation.⁸⁷

On the other hand, in human rights scholarship, interestingly and perhaps surprisingly, there is also an argument that judges choosing a section 4 declaration is possibly weaker than a section 3 interpretation.⁸⁸ With a section 4 declaration, this does nothing to help the immediate claimant: the offending legislation, in this instance lacuna and all, remains as is.⁸⁹ While Parliament may choose to act in response to a section 4 declaration (and this is discretionary, at least in theory), any new legislation will likely be prospective and will not help the current claimant.⁹⁰

When judges are deciding between sections 3 and 4 HRA, for example, Kavanaugh identifies two sets of ‘guidelines’ which potentially factor in favour of a section 4 declaration: the ‘fundamental features limit’ (as discussed above, with reference to the ‘grain’ of the legislation argument) and ‘legislative deliberation limit’.⁹¹

On the second, one of the factors consistently emphasised in the scholarship is whether there are difficult policy-based choices to be made or if a whole new legislative scheme needs to be written.⁹² Neither would seem necessarily to be present in Mercer. On the former, for example: as suggested above, Mercer’s suggestion to add words to section 146 TULRCA, using section 3 HRA to tie any new protection to section 238A, might answer many of the supposed difficult policy questions, in addition to existing extensive jurisprudence on the concept of ‘detriment’ and potentially additionally the limiting role of the ‘sole or main purpose’ test.⁹³ Even if not following Mercer’s suggestion, to remedy the lacuna would likely take only one or two new provisions or at most a small handful of provisions, depending on drafting (eg, if contrary to the above, if the Court of Appeal’s concerns might lead to a new provision being added on the meaning of detriment in this context). Another factor sometimes mentioned in human rights scholarship is that a section 4 declaration might be preferred if it is known that new legislation in the same area is already on the horizon,⁹⁴ however Lady Simler explicitly noted that no new legislation on this topic was forthcoming at that time (before the change in Government).⁹⁵

In comparison however to the Supreme Court, the Court of Appeal in Mercer deemed there was a violation of article 11 but decided not to use either section 3 or 4 HRA, which seemed in itself quite an alarming result: for the Court of Appeal, a likely violation of a Convention right, however simultaneously no response under the HRA.⁹⁶ In this respect, if section 3 HRA truly could not be utilised, it is positive that the Supreme Court overcame judicial ‘reluctance’ to utilise section 4.⁹⁷ It is worth recalling: a declaration under section 4 is discretionary (‘may’, in comparison to ‘must’ at section 2 HRA). By contrast, after the Supreme Court’s decision, new legislation most likely will be forthcoming for future claimants in the position of Ms Mercer, although to obtain financial remedy herself, Ms Mercer will be compelled to take her case to Strasbourg.⁹⁸

7. DISTINCTION BETWEEN PUBLIC AND PRIVATE SECTOR WORKERS?

Both the Court of Appeal and Supreme Court in Mercer interpreted Strasbourg jurisprudence on article 11 ECHR in relation to detriment for taking part in industrial action to distinguish between the State as public authority employer and private sector employers. Lady Simler discussed the following Strasbourg authorities: Ezelin v France,⁹⁹Karaçay v Turkey,¹⁰⁰Danilenkov v Russia,¹⁰¹Ognevenko v Russia¹⁰² and Tek Gıda İş Sendikası v Turkey.¹⁰³

While a ‘minimal’ sanction was described as sufficient to breach article 11 for an employee working for a public authority, in comparison Strasbourg authorities were interpreted to require only ‘some’ protection if the worker or employee was in the private sector.¹⁰⁴ For example, for Lady Simler: ‘the Strasbourg court has also recognised a distinction between the state acting as public sector employer and penalising employees directly for participating in a strike (Ezelin, Karaçay and Ognevenko are examples and to the extent that a margin of appreciation was accorded to the state employer in question, it was a narrow margin); and cases where the state as regulator is said to owe a’ weaker ‘positive duty to legislate to protect privately employed workers’ article 11 rights’ (adjective added).¹⁰⁵

It is here that confusion is liable to arise. The Supreme Court was not drawing necessarily a distinction between public and private sector workers, but rather between claims against the State directly as employer versus when claims are brought against the State as ‘regulator’. Two points can be made in response.

First, it is not necessarily clear that Strasbourg jurisprudence does so sharply differentiate between these two types of scenario. For example, in the seminal case of Good Shepherd, the Grand Chamber memorably stated the ‘boundaries of the State’s positive and negative obligations under Article 11 of the Convention do not lend themselves to precise definition’.¹⁰⁶ ‘Similar principles’ were stated to apply in both situations: ‘the criteria to be applied do not differ in substance’.¹⁰⁷ Lady Simler relied on a particular paragraph for these purposes from RMT v UK.¹⁰⁸ Nor is it clear, however, that the paragraph in question supports the analysis drawn from it. If the Strasbourg Court in that paragraph in RMT was drawing a distinction between ‘state and private sector employers’, there would surely at the end of the quotation additionally be a reference to private sector workers rather than more generally to the ‘secondary aspects of trade union activity’.¹⁰⁹ While it is refreshing that Lady Simler dealt with Strasbourg authorities in some detail and provided quotations from each (eg, particularly in comparison to the earlier authority of Metrobus), there are also statements from the same authorities listed above indicating to the contrary.¹¹⁰ While ideally there could be more clarity in the Strasbourg jurisprudence also on this point: rather than a sharp distinction, potentially a spectrum more comes to mind.¹¹¹ The Strasbourg Court in Tek Gıda v Turkey repeated the reasoning above from Good Shepherd,¹¹² and in Danilenkov v Russia, when considering whether or not there was ‘direct intervention by the State’:

It considers that it is not necessary to rule on this issue, since the responsibility of the Russian Federation would, in any case, be engaged if the matters complained of resulted from a failure on its part to secure to the applicants under domestic law the rights set forth in Article 11 of the Convention. ¹¹³

While the Supreme Court distinguished the former two decisions (Tek Gıda and Danilenkov) as not about the right to strike and rather about the union’s ‘core’ right to be heard,¹¹⁴ this does not answer the question about how the ‘similar’ ‘balancing exercise’ applies when the State is employer as opposed to when the State is regulator.¹¹⁵ Further, on Mercer’s distinction between State as employer or regulator, it might also be noted that the Strasbourg Court memorably awarded stringent protection under article 11 in Wilson v UK: finding a violation whilst not even considering article 11(2), in the private sector context when there was a comparatively light detriment of withholding a pay increase.¹¹⁶ No doubt it would be responded that Wilson v UK, similar to the aforementioned Tek Gıda and Danilenkov, was also instead about the union’s ‘core’ right to be heard.¹¹⁷

Second and more importantly practically, going back to broader human rights scholarship, in that scholarship, it is sometimes stated that courts under section 4 HRA can shape the subsequent remedial response by Parliament by giving a steer in the judgment awarding the declaration.¹¹⁸ The potential concern here is if Parliament or the executive takes the judgment in Mercer as a steer to distinguish between public and private sector workers in any remedial response, if public authority worker would foreseeably be interpreted as public sector worker. Indeed, the Court of Appeal in Mercer at one point referred more broadly to ‘public sector employees’,¹¹⁹ although potentially directly employed by the State is a narrower concept.¹²⁰ How one defines the distinction is not, however, the concern. To be clear, the Supreme Court was not stating that legislation can or should distinguish between public and private sector workers; rather, when stating what the violation of article 11 was, the Supreme Court interpreted Strasbourg jurisprudence to deem stricter protection necessary for public authority workers. If using the judgment as a blueprint, that in turn might lead to a legislative response distinguishing between types of worker. There would be three very strong reasons for concern if this would be the route that was followed. In addition to those three strong reasons: as above, this may not be a correct interpretation of Strasbourg jurisprudence and, while some passages in Mercer suggest that the State’s ‘regulatory’ role applies only to workers in the private sector, arguably this role should apply more broadly to assess the level of legal protection on a given matter for all individuals impacted.¹²¹

First and foremost, it would be surely counterintuitive to use a human rights judgment to introduce a new, limiting, type of distinction in legislation. Most obviously, human rights scholarship tends to advocate a broader approach to personal scope. Cases where human rights are successfully invoked tend also to take a broader approach to personal scope. Even where the claim is unsuccessful as in Deliveroo, for example, Underhill LJ in the Court of Appeal suggested that the Strasbourg Court might ‘be more generous in its definition of the limits to ‘permitted substitution’,¹²² although the Supreme Court in that case would appear implicitly to have disagreed with this.¹²³ For instance, domestically, in recent cases, section 3 HRA has memorably been used to expand the eligibility of particular statutory employment and labour rights to otherwise excluded persons: in Gilham v Ministry of Justice,¹²⁴ in Vining v London Borough of Wandsworth,¹²⁵ in NUPFC v Certification Officer.¹²⁶ As another example, elsewhere in Strasbourg jurisprudence: under article 6 ECHR, the more recent case law on civil servants has changed to include a presumption that article 6 will also apply to civil servants, save where a limited exception is met.¹²⁷ This is the opposite trend of direction: article 6 is being used to overcome a similar type of distinction.

Second, the Supreme Court in Mercer took time to recount the history of now sections 146 and 152 TULRCA.¹²⁸ Never in that history has legislation on detriment or dismissal for taking part in the activities of an independent trade union distinguished between public and private sector workers. It would be a complete departure from that history if any new remedying legislation were to distinguish between public and private sector workers. The Strasbourg Court deemed continuity in statutory provisions to be important in RMT v UK.¹²⁹ Third, while admittedly more recent legislation on industrial action—in terms of the procedural requirements for a ballot to be valid (section 4 Trade Union Act 2016) and on minimum service levels—does distinguish ‘important public services’ and specific public services in minimum service levels legislation (Strikes (Minimum Service Levels) Act 2023),¹³⁰ after the General Election in July 2024, the incoming new Government has pledged to repeal both these Acts.¹³¹ Hence if any remedying legislation to section 146 were to include a distinction between public and private sector workers, not only would this be a historical departure for these provisions; presuming the 2016 and 2023 Acts are repealed, any such new remedying legislation would be the total outlier.

8. CONCLUSION

Mercer is immediately a powerful and landmark judgment for the finding that there was a violation of the right to strike under article 11 ECHR when there is no legal protection against detriment short of dismissal for taking part in industrial action. The Supreme Court felt unable to render a convention-compliant interpretation of section 146 TULRCA and instead made a declaration of incompatibility under section 4 HRA. The onus now passes to Parliament and the executive, if they so choose, to legislate to remedy to fill the ‘lacuna’ as found. The lacuna was that, as currently construed, while legislation protects employees from dismissal for taking part in industrial action, there is no protection against action short of dismissal in these circumstances. While it is positive that the Supreme Court delved into and engaged with the detail of Strasbourg jurisprudence on article 11, the judgment in Mercer should not, however, be taken as a blueprint for any possible new remedying legislation in the following respect. Any new legislation definitely should not take the Supreme Court’s interpretation of Strasbourg authorities as a steer to distinguish between public and private sector workers.

Footnotes

Author notes