New Judgment: Commissioners for Her Majesty’s Revenue and Customs v Coal Staff Superannuation Scheme Trustees Ltd [2022] UKSC 10

On appeal from: [2019] EWCA Civ 1610

The Respondent is the corporate trustee of a tax-exempt United Kingdom pension fund. It held a large portfolio of UK and overseas shares. To generate revenue, it engaged in a practice known as stock lending. This involves a shareholder (the lender) transferring ownership of shares to another party (the borrower) on terms that the borrower will (i) return equivalent shares to the lender at the end of the lending period and (ii) pay an amount to the lender equivalent to the dividends paid on the shares during that period. These payments are known as a “manufactured dividend” (“MD”) if the shares are held in a UK company. If the shares are in a non-UK company, they are known as a “manufactured overseas dividend” (“MOD”).

This appeal concerns the alleged differential treatment for tax purposes of MDs and MODs received by tax-exempt taxpayers such as the Respondent. That differential treatment is said by the Respondent to be unlawful as a matter of European law because it constitutes a restriction on the free movement of capital, contrary to Article 63 of the Treaty on the Functioning of the European Union (“TFEU”).

The Respondent pays no UK tax on dividends it receives from shares in both UK and overseas companies. In relation to overseas companies however, the dividend may be subject to a withholding tax (“WHT”) charged by the country in which the company is based. To prevent double taxation of UK taxpayers, the UK grants a tax credit to the shareholder lender known as withholding tax credit. That tax credit is of no effect in relation to a tax-exempt shareholder such as the Respondent however because it has no relevant tax liability to which the credit can be applied.

Under EU law, this phenomenon of dividends being taxed both in the country where the company is based and in the country where the shareholder is based (known as “juridical double taxation”) is accepted as an unfortunate fact of life which does not generally fall foul of Article 63 TFEU.

Being contractual payments rather than actual dividends, MODs are not liable to WHT. To prevent differential treatment for tax purposes between real dividends and manufactured dividends, however, the UK implemented a regime whereby MODs were subjected to a deemed withholding tax (the “MOD WHT”) payable at source by the borrower of the shares. To compensate for this deduction, the share lender was granted a corresponding tax credit equivalent to that which would have been available to avoid juridical double taxation had it received an actual overseas dividend rather than a MOD.

In the case of a tax–exempt share lender such as the Respondent, however, the effect of the MOD WHT was to reduce its net income for which it was compensated only with tax credits for which it had no use. Unlike in relation to actual dividends, that did not amount to juridical double taxation, because it arose solely as a result of the UK scheme for taxing MODs which was a policy choice. The Respondent therefore complained that the MOD WHT regime amounted to a restriction contrary to Article 63 because, given that no MOD WHT was deducted in relation to MDs, it disincentivised tax–exempt entities from acquiring and stock lending shares in overseas companies relative to shares in UK companies.

The Respondent claimed that, as a result, it was entitled to repayment of the MOD WHT as tax deducted on its own income. Its own claim amounted to some £8.8 million, but the practice of stock lending by tax–exempt entities was widespread while the UK MOD WHT regime was in operation. This case therefore serves as a test case for more than £600 million in MOD WHT which, subject to the outcome of this appeal, the Appellant may have to repay.

At first instance before the First–tier Tribunal, the Appellant succeeded on the basis that there was no restriction which offended Article 63 TFEU. The Upper Tribunal overturned that decision, finding in favour of the Respondent. The Court of Appeal arrived at the same outcome as the Upper Tribunal though by different reasoning.


HELD – The Supreme Court unanimously allows the appeal.


In order to determine whether or not the MOD tax regime contravened Article 63 TFEU, the appropriate question to ask was whether, “but for” the MOD tax regime, investors would be sufficiently better off from engaging in stock lending such that its presence was a disincentive beyond that already created by juridical double taxation to them acquiring overseas shares, as opposed to UK shares.

The Court of Appeal concluded that the MOD tax regime did create such a disincentive because, without the MOD WHT being deducted prior to the MOD being paid to the lender, lenders could negotiate for a larger MOD which reflected the benefit to the borrower of dealing with the borrowed shares in a way which reduced the WHT which the borrower actually paid by using dividend arbitrage (i.e. moving the ownership of the borrowed shares to entities in different countries so as to optimise the rate of WHT payable on them).

Lord Briggs and Lord Sales consider that to be a flawed market economic analysis. The benefit to the stock lender of any additional benefits generated by the borrower’s use of the shares (for example, dividend arbitrage) is to be found in the size of the lending fee, not of the MOD. In a sophisticated market, it is to be assumed that the lending fee already had a relationship to the expected benefits accruing to the borrower from the shares.

Whether there was any relevant dissuasive effect constituting a breach of Article 63 TFEU therefore depended on whether the obligation on the borrower, here Authorised UK Intermediaries (“AUKIs”), to account to the Appellant for the MOD WHT was likely to have reduced the amount which the borrower would otherwise have been prepared to pay the lender as a lending fee for the opportunity to, for example, engage in dividend arbitrage. That depended, in turn, on whether the MOD tax regime would result in the AUKI actually having to pay anything to the Appellant by way of MOD WHT deductions.

The Appellant did not, in fact, receive any cash payments attributable to MOD WHT. The reason for this was that AUKIs typically had more than enough withholding tax credits to soak up, by way of set–off, all the MOD WHT payable by them. To suggest that the MOD tax regime would potentially have the effect of discouraging investment in foreign shares relative to UK shares by tax–exempt investors was therefore no more than pure speculation which was insufficient to make out a breach of Article 63 TFEU. The AUKIs were not, in reality, paying anything more to the Appellant due to the MOD tax regime than they would have done otherwise.

Even if there is a dissuasive effect, Lord Briggs and Lord Sales conclude that the appeal should nonetheless succeed based on the remedy sought by the Respondent.

The Respondent’s complaint is that the UK tax regime should have provided it with usable tax credits in relation to the liability of a borrower to pay the MOD WHT, on the basis that it was a UK tax deducted at source from the share lender. Lord Briggs and Lord Sales reject this. The purpose of the MOD regime and contractual arrangements between lenders and borrowers was to mimic the position which would have existed had the lender retained the shares. From the lender’s perspective, the tax credit it received under the MOD tax regime was exactly the same in its effect as the tax credit in respect of WHT which it would suffer if it retained the shares.

The tax credits which the Respondent claims should have been paid to it were therefore ultimately in respect of WHT levied and retained by foreign tax authorities. The UK tax authorities did not receive those sums, but had decided as a matter of policy to provide access to credits in relation to tax levied by foreign states. Although the UK used its domestic tax regime to achieve this, that did not mean as a matter of substance that the foreign WHT was to be treated as if it had been collected and retained by the UK tax authorities.

Further, although the dissuasive effect regarding investment in foreign shares identified by the Court of Appeal was comparatively small, the relief sought by the Appellant was a sum equal to the full amount of the credits. The remedy claimed was therefore wholly disproportionate when compared with the breach. Having reviewed the relevant principles of the European Court of Justice, Lord Briggs and Lord Sales conclude that a remedy to be provided in respect of a breach of Article 63 is required to be proportionate to the relevant violation of EU law. Where the issue is not the restitution of money received by the state but the provision by the state of a financial benefit (i.e. the payment of credits in relation to taxes paid to another state as in this case), the remedy has to be tailored to the wrong committed in breach of EU law.

On the facts of the present case, if there was a breach of Article 63 it was only on the basis that the MOD WHT regime omitted to allow for the grant of a tax credit payment to the extent that a tax–exempt lender like the Respondent could show that it had been unable to benefit from the possibility of sharing in the financial gains from dividend arbitrage by an AUKI. That is the extent of the relief to which the Respondent is entitled, if the remedy is to be kept proportionate to the wrong suffered. However, the Respondent has never sought to claim for such a loss nor presented any evidence to show that in fact it suffered any such loss.


Judgment (PDF)
Press summary (HTML version)
Judgment on The National Archives (HTML version)
Judgment on BAILII (HTML version)

Watch hearing

26 October 2021
Morning session
Afternoon session

27 October 2021
Morning session
Afternoon session

New Judgment: R (on the application of Coughlan) v Minister for the Cabinet Office [2022] UKSC 11

On appeal from [2020] EWCA 723

This appeal concerns a challenge brought by the Appellant to orders made by the Respondent in respect of Braintree District Council and nine other local authorities (“the Pilot Orders”). These Pilot Orders authorised schemes to temporarily change the rules set out in secondary legislation governing local elections. These schemes, which were implemented in ten local authority areas in respect of the local government elections in May 2019, each introduced a new requirement for some form of voter identification for those local elections.

The Appellant is a resident of the Braintree area who challenged the lawfulness of the Pilot Orders. His challenge was dismissed on the merits in the High Court. The Court of Appeal granted permission to appeal this decision in view of the “important constitutional function served by local government elections.” The Appellant’s appeal was dismissed on the merits by the Court of Appeal.

The primary issue in the appeal was whether the Pilot Orders were made ultra vires, that is outside the legal powers of the Minister for the Cabinet Office, because the pilot schemes they sought to establish were not schemes within the meaning of section 10(2)(a) of the Representation of the People Act 2000.

The second issue in the appeal was whether the pilot schemes were authorised for a lawful purpose under section 10(1) of the Representation of the People Act 2000, consistent with the policy and objects of that Act.

The appeal does not concern the merits or otherwise of the decision to introduce the pilot schemes, nor the merits of voter identification schemes in general, but only whether the decision to introduce the pilot schemes was lawful. The Runnymede Trust, Operation Black Vote, Voice4Change England, LGBT Foundation and Stonewall were given permission to intervene by written submissions.


HELD – The Supreme Court dismissed the appeal.


In respect of the primary issue, the Court found that the Pilot Orders were not made ultra vires. Section 10 of the Representation of the People Act 2000 is titled “Pilot schemes for local elections in England and Wales”. Section 10(1) enables the Minister for the Cabinet Office by secondary legislation “to make such provision for and in connection with the implementation of a scheme as he considers appropriate”. However, that power to make secondary legislation is limited to a scheme within the meaning of section 10(2). Section 10(2)(a) provides for schemes as regards “… how voting at the elections is to take place”.

Having regard to the relevant principles of statutory interpretation, the legislative framework for local government elections, and the content of the pilot schemes in question, the Court finds that the pilot schemes were schemes within the meaning of section 10(2)(a) of the Act, and in particular, that they were schemes as regards “… how voting at the elections is to take place”.

In respect of the second issue, the Court found that the pilot schemes were authorised for a lawful purpose under section 10(1) of the Representation of the People 2000 Act. The Court finds that the purpose of section 10 is to facilitate pilot schemes to enable the gathering of information to assist in the modernisation of electoral procedures in the public interest. The Pilot Orders were made to promote that object, and accordingly, were authorised for a lawful purpose.

Case Comment: Her Majesty’s Attorney General v Crosland [2021] UKSC 58

In this post, Leigh McLevy, trainee solicitor at CMS, comments on the decision in Her Majesty’s Attorney General v Crosland, a case which concerns an embargoed judgment and contempt of court.


Mr Crosland appealed against a decision of the Supreme Court in which he was ordered to pay a fine of £5,000 to HM Paymaster General, and costs of a further £15,000, for contempt of court. The court at first instance (“First Instance Panel”) was satisfied that Mr Crosland committed contempt of court by disclosing the outcome of the court’s judgment in R (on the application of Friends of the Earth) v Heathrow Airport Ltd [2020] UKSC 52 (“Heathrow Judgment”) whilst still in draft and subject to embargo.

First Instance Panel Decision

Mr Crosland, an unregistered barrister, had been involved in the proceedings giving rise to the Heathrow judgment. The Heathrow case concerned the lawfulness of the Airports National Policy Statement (“ANPS”), particularly on whether the Secretary of State had failed to have proper regard to the Paris Agreement or to explain how the ANPS was compatible with the UK’s emissions targets. The Heathrow Judgment was circulated to the parties’ representatives on 9 December 2020 and it was to be handed down on the morning of 16 December 2020. Mr Crosland formed the view that there were inaccuracies in the draft and believed that the ANPS was misleading as it was presented as being compatible with the UK’s international obligations on climate change and failed to take the Paris Agreement targets into account.

On the morning of 15 December 2020, Mr Crosland sent an email to the Press Association in which he disclosed the outcome in Heathrow Judgment and outlined what he saw was the inaccuracies in the judgment. Additionally, he issued a statement on Plan B Earth’s Twitter account. Mr Crosland acted in the knowledge that the judgment was under embargo until the following morning and that his actions may be in contempt of court. It was Mr Crosland’s position that it was necessary to “sound the alarm early” and raise awareness of the issue before the judgment was released due to the possible impact on the climate.

On 10 May 2021 the First Instance Panel of Supreme Court Justices issued the decision on liability and penalty, finding Mr Crosland to be in contempt of court and ordered to pay a fine of £5,000. A second decision on costs was delivered on 14 June 2021 in which Mr Crosland was ordered to pay £15,000 of the £22,504 of costs incurred.

Appeal to the Supreme Court

On 16 July 2021, Mr Crosland filed a notice of appeal on four grounds to which the Attorney General filed a notice of objection. The Attorney General submitted that there was no right of appeal from a decision of the Supreme Court and raised as a preliminary issue whether the court (“the Appeal Panel”) had jurisdiction to entertain the appeal.

Preliminary Issue – Jurisdiction

Mr Crosland sought to appeal against the order, on the basis that he had not been in contempt at all, and against the order for costs. Permission to appeal was granted by the same panel which heard the contempt application, and it was to be heard before a panel of five judges, none of whom had been involved either in the proceedings in the Heathrow Judgment or in the contempt application. The Appeal Panel decided that they had jurisdiction to hear the appeal. The basis for the jurisdiction is that where a decision such as contempt of court which can attract a custodial sentence, s 13 of the Administration of Justice Act 1960 can be read to include appeals from the Supreme Court to a separate panel of the same court in such limited circumstances. The court noted that “it is open to pave the way for a procedural route by which a substantive right may be exercised.”

Grounds 1 and 2

Mr Crosland appealed on the grounds that 1) the First Instance Panel erred in its approach to considering the relevance of Mr Crosland’s beliefs and motivations namely whether his breach of the embargo was a proportionate response to the suppression of evidence about the dangers of the Heathrow airport expansion and 2) the First Instance Panel failed to mention and therefore wrongly disregarded a letter written by leading scientists (“the Scientists Letter”) which demonstrated, amongst other things, the efficacy of Mr Crosland’s tactic of breaching the embargo. Both grounds were rejected on appeal. The Appeal Panel were not satisfied that Mr Crosland’s submission that his breach of the embargo drew facts to the attention of the public in relation to the impact on climate change would have otherwise passed unnoticed as he would have been free to comment and criticise the judgment the very next day. Further, the Scientists Letter did not suggest that its signatories only became aware of the Heathrow Judgment and its contents as a result of Mr Crosland’s breach or that they would not have participated in the public discussion that followed its release.

Ground 3
The third ground of appeal was that the First Instance Panel was not an independent and impartial tribunal as required by Article 6(1) ECHR and the Human Rights Act 1998. Mr Crosland submitted that the First Instance Panel was not independent and impartial and that the fair-minded observer would conclude that there was a real possibility that the Panel was biased. This was rejected by the Appeal Panel. Two main reasons were provided – 1) the decision to bring proceedings for contempt was taken by the Attorney General, to whom the matter had been referred by the President of the Supreme Court and was not taken by the court itself and 2) the First Instance Panel did not include any of the judges who sat on the Heathrow Judgment appeal.

Ground 4

Mr Crosland submitted that pursuant to the Attorney General’s obligations of disclosure under Article 6 ECHR and s 3 of the Criminal Procedure Investigations Act 1997, Mr Crosland should have been given information about an alleged breach of the embargo on court judgments in the case of Begum v Special Immigration Appeals Commission [2020] EWCA Civ 918. It was held that this submission had no force because it was irrelevant to the decision as to whether Mr Crosland was in contempt of court, or the punishment for contempt, to consider what the Attorney General may or may not have done in a different case.

Ground 5

This submission was on the basis that the court’s ruling on costs was oppressed and unjust. Mr Crosland submitted that the cost order of £15,000 should not be greatly at variance with any fine imposed and that as he has a modest income, the overall financial penalty of £20,000 for an act of conscience is oppressive, arbitrary, and disproportionate. This ground was rejected by the Appeal Panel. The award of costs is a matter for the discretion of the court making the order and the appeal court should only interfere if there has been an error of legal principle, which the Appeal Panel did not find.


Overall, Mr Crosland’s appeal was dismissed by the Appeal Panel of the Supreme Court on all grounds.


This case demonstrates that it is, in limited circumstances, possible for the Supreme Court to act as an independent and impartial tribunal to entertain appeals on decisions made by the Supreme Court itself when exercising its jurisdiction on contempt of court and where the Appeal Panel is made up of judges who did not sit on the First Instance Panel.

Further, the judgment shows that contempt of court is taken very seriously and is difficult to defend, even where there may be a public interest argument or a moral reason to disclose the information and high penalties and cost orders may be imposed.

Case Preview: Stanford International Bank Ltd (In Liquidation) v HSBC Bank Plc

In this post, Emilija Lazarevic, a trainee solicitor in the litigation team at CMS, previews the decision awaited from the UK Supreme Court in the matter of Stanford International Bank Ltd (In Liquidation) v HSBC Bank Plc.

On 19 January 2022, the Supreme Court heard the appeal in Stanford International Bank Ltd (In Liquidation) v HSBC Bank Plc. The appeal turns on whether a company in liquidation can be considered to have suffered loss where, while it is still trading, its bank pays money out of the company’s accounts to discharge debts owed by the company. It is likely that this case will further set out the limits of the Quincecare duty, following a spate of recent high-profile cases in this area.

Factual Background

In April 2009 the claimant went into insolvent liquidation after being used as a vehicle for a Ponzi scheme fraud. It held accounts with the defendant.

The claimant’s liquidators brought a claim for a loss amounting to £116.1 million. A sum of £118.5 million had been paid out of the claimant’s accounts between August 2008, when the liquidators considered that the defendant should have frozen the accounts, and February 2009, when the defendant did freeze the accounts. The claimant contended that had the accounts been frozen by August 2008, the monies would have remained there rather than being paid away to holders of Certificates of Deposit, which were issued by the claimant.

Damages were sought by the claimant for breach of an implied duty owed by the bank to its customers, as explained in Barclays Bank plc v Quincecare Ltd [1992] 4 All E.R. 363. In this case it was established that “a banker must refrain from executing an order if and for as long as the banker is ‘put on inquiry’ in the sense that he has reasonable grounds (although not necessarily proof) for believing that the order is an attempt to misappropriate the funds of the company”.

Procedural History

The defendant sought a summary judgment and/or striking out from the High Court in respect of the claim, of which there were two elements. The first element related to the defendant’s alleged breach of implied duty, as considered in Quincecare. The second element concerned the defendant’s alleged dishonest assistance.

At first instance, the court granted the application in part. It was held that it was not appropriate to strike out or grant summary judgment in respect of the Quincecare breach. On the alleged facts, if the company’s accounts were frozen when they should have been, the claimant would have possessed an extra £118 million in assets, which would have allowed the liquidators to pursue the company’s own claims and distribute any winnings to the creditors. The liquidators were deprived of such opportunities owing to the defendant’s alleged breach, and the court considered this to be a real loss to the company.

The plea relating to dishonest assistance was struck out on the basis that dishonesty had not been pleaded against any particular individual working for the defendant. The defendant relied on aggregate information held by different individuals, and as those individuals could not be said to be dishonest, no case could be made against the defendant. The claimant argued that corporate recklessness could amount to dishonesty, but this was not accepted by the court.

The defendant bank appealed against the High Court’s decision not to strike out the loss claim, and the claimant appealed against the decision to strike out its claim of dishonest assistance.

The defendant’s appeal was allowed. The Court of Appeal held that, although if the accounts had been frozen at the appropriate time the claimant would have been able to pay its creditors once the insolvency process began, the defendant’s duty had been to the claimant alone and not the creditors (citing Singularis Holdings Ltd (in Liquidation) v Daiwa Capital Markets Europe Ltd [2018] EWCA Civ 84).

Furthermore, it was held that the previous judge had mistakenly confused the claimant company’s position before and after the inception of insolvency proceedings. In its view, the true distinction is between a company that is trading and a company in respect of which a winding up process has commenced, not between a solvent trading company and an insolvent trading company. If an insolvent company is trading, money paid to a creditor would reduce the assets of the company, but it would also reduce the company’s liabilities. In other words, the company’s overall net asset position would remain unchanged. On that basis, having more cash available upon the insolvency would be a benefit to the creditors rather than to the company itself while it was trading. As a result, the claimant was deemed to have suffered no loss and so the claim could not succeed.

The claimant’s appeal was dismissed. It was held that the previous judge was correct in stating that dishonesty could not properly be alleged by adding the knowledge of one innocent person to another (citing Armstrong v Strain [1952] 1 K.B 232 and Greenridge Luton One Ltd v Kempton Investments Ltd [2016] EWHC 91 (Ch)).


The Court of Appeal has limited the scope of the Quincecare duty by confirming that it cannot extend to a customer’s creditors. It has also established that negligent or reckless practices by banks will not be sufficient to fulfil the meaning of “dishonesty”, which is required for a dishonest assistance claim to succeed.

If the UK Supreme Court ruling favours the Court of Appeal judgment, this may deter future claimants from pursuing similar claims against banks that have been unknowingly involved in processing Ponzi scheme payments. Such a ruling would be welcomed by financial institutions that operate client accounts and process payments in accordance with client instruction.

This week in the UKSC – w/c 25th April 2022

This Week in the Supreme Court – w/c 25th April 2022

Hearings in the Supreme Court are now shown live on the Court’s website.

On Wednesday 27th April, the Court will hand down two judgments at 09:45 in Courtroom One.

The first is Commissioners for Her Majesty’s Revenue and Customs v Coal Staff Superannuation Scheme Trustees Ltd and will consider the UK tax treatment of “manufactured overseas dividends”. The citation will be [2022] UKSC 10 and the judgment being appealed is [2019] EWCA Civ 1610.
The second judgment being handed down is R (on the application of Coughlan) v Minister for the Cabinet Office [2022] UKSC 11, on appeal from [2020] EWCA 723. The Court will consider whether the voter identification pilot schemes that were implemented in the May 2019 local government elections were unlawful.

On Thursday 28th April 2022, the Court will hear the case of Hastings v Finsbury Orthopaedics Ltd and another (Scotland). The question the Court is asked to consider is did the lower courts in law by not holding that the hip replacement product used for the Appellant’s hip replacement was defective within the terms of the Consumer Protection Act 1987? This is on appeal from [2021] CSIH 6.

From Thursday 28th to Friday 29th April, the Court will hear an assessment of costs hearing in R (on the application of Haworth) v Commissioners for Her Majesty’s Revenue and Customs [2021] UKSC 25.

The following Supreme Court judgments remain outstanding: (As of 27/4/22)

The Law Debenture Trust Corporation plc v Ukraine (Represented by the Minister of Finance of Ukraine acting upon the instructions of the Cabinet of Ministers of Ukraine) Nos. 2 and 3, heard 9-12 December 2019
BTI 2014 LLC v Sequana SA and Ors, heard 4 May 2021
East of England Ambulance Service NHS Trust v Flowers and Ors, heard 22 June 2021
Basfar v Wong, heard 13th-14th October
Secretary of State for the Home Department v SC (Jamaica), heard 19th October
Harpur Trust v Brazel, heard 9th November 2021
Guest and another v Guest heard 3rd December 2021
Fearn and others v Board of Trustees of the Tate Gallery heard 7th December 2021
Stanford International Bank Ltd (in liquidation) v HSBC Bank PLC, heard 19th January 2022
Commissioners for Her Majesty’s Revenue and Customs v NCL Investments Ltd and another, heard 25th January 2022
R v Maughan, heard 27th January 2022
Cornerstone Telecommunications Infrastructure Ltd v Compton Beauchamp Estates Ltd, Cornerstone Telecommunications Infrastructure Ltd v Ashloch Ltd and another and On Tower UK Ltd (formerly known as Arqiva Services Ltd) v AP Wireless II (UK) Ltd, heard 1st February 2022
DCM (Optical Holdings) Ltd v Commissioners for Her Majesty’s Revenue and Customs (Scotland), heard 8th February 2022
Competition and Markets Authority v Pfizer Inc and Flynn Pharma Ltd, heard 22nd February 2022
Hastings v Finsbury Orthopaedics Ltd and another (Scotland), heard 28th April 2022

Lord Kerr Essay Competition

Lord Brian Francis Kerr, Baron Kerr of Tonaghmore, PC was a Northern Irish barrister and a senior judge who was Lord Chief Justice of Northern Ireland and then a Justice of the Supreme Court of the United Kingdom. At the time of his retirement, he was the longest-serving Supreme Court Justice and the last original member of the Court.

Following his passing, the Human Rights Lawyers Association (HRLA) received a gift donation from his family to create an opportunity that celebrates the formidable legal career that Lord Kerr had through his influential judicial understanding of human rights and the law. From this gift, the HRLA and the Young Lawyers Committee (YLC) have created an essay competition that focuses on a contemporary human rights issue.

The UKSC Blog has partnered with the HRLA and, as an additional prize, the winning essay will be published on the UKSC Blog.

The essay question is:

“Is eco-anxiety capable of amounting to torture or to inhuman or degrading treatment or punishment, within the meaning of Article 3 ECHR?”

Further details and full rules can be found at The deadline for entries is Friday 20 May 2022 – good luck!

Case Comment: Croydon London Borough Council v Kalonga [2022] UKSC 7

In this post, Jen Knibbs, an associate at CMS, comments on Croydon LBC v Kalonga [2022] UKSC 7, an important decision on whether a landlord can terminate a flexible tenancy agreement before the expiry of the fixed term if the tenancy agreement does not expressly provide for re-entry or forfeiture.


Ms Chipo Kalonga (“Ms Kalonga”) was the tenant of a property in Croydon under a flexible tenancy for a fixed term of five years from 25 May 2015 to 24 May 2020 (the “Tenancy Agreement“). Croydon London Borough Council was her landlord (“Croydon“).

A flexible tenancy is a type of secure council tenancy which lasts for a fixed period of time, usually at least 5 years.

On 2 August 2017, Croydon served a notice on Ms Kalonga, which relied on grounds 1 and 2 of schedule 2 of the Housing Act 1985, seeking to terminate the Tenancy Agreement due to rent arrears and anti-social behaviour. The notice was in the standard form.

On 29 August 2017, Croydon issued a claim in the County Court at Central London seeking possession of the property. The claim was made on grounds 1 and 2 of schedule 2 of the Housing Act 1985. No claim was made on the ground of forfeiture.

In response, Ms Kalonga served a defence and counterclaim arguing, amongst other things, that the Tenancy Agreement did not include a forfeiture clause; Croydon had failed to serve a valid notice; and the claim was defective as a claim for possession of a flexible tenancy during the fixed term. Ms Kalonga defended the claim on the basis that it was only possible to determine a flexible tenancy, as is the case with fixed term secure tenancy, by terminating it in accordance with the forfeiture-like procedure in section 82(3) Housing Act 1985.

On 9 May 2019, the preliminary issue of “the correct manner in which to determine a secure flexible tenancy during the fixed term (including whether, and if so how, and principles relating to forfeiture apply)” was transferred to the High Court. The remainder of the claim and counterclaim was stayed pending the determination of this preliminary issue.

High Court

The High Court determined that the Tenancy Agreement did not contain a forfeiture clause and that without one, Croydon had no right to end the Tenancy Agreement before the end of the fixed term. Croydon appealed.

Court of Appeal

The Court of Appeal dismissed the appeal and agreed with the ruling of the High Court.

The Court of Appeal held that the words “subject to termination by the landlord” did, at first glance, allow a landlord to terminate the tenancy by any lawful means. However, considering the legislative intention behind section 32 of the Housing Act 1980 (the precursor to section 82 of the Housing Act 1985), a fixed term tenancy could only be brought to an end by obtaining an order for termination pursuant to a forfeiture clause, at which point a periodic tenancy would commence. Croydon appealed to the Supreme Court.

Supreme Court

There were two issues for the Supreme Court to consider:

Whether the existence of a provision for forfeiture in the Tenancy Agreement and its exercise by obtaining a termination order in lieu of forfeiture under section 82(3) of the Housing Act 1985 is the sole way in which a secure fixed term tenancy can be brought to a premature end; and
Whether the Tenancy Agreement contained a provision for forfeiture, under which Ms Kalonga could terminate the least because of some fault on the part of Croydon.

In consideration of the first issue, it was held that there are only two ways a landlord can recover possession during the fixed term. The first way is where a tenancy agreement contains a break clause; the second where a tenancy agreement contains a forfeiture clause. In short, a landlord is able to seek possession against a tenant with a secure fixed term tenancy where there is an exercisable break or forfeiture clause.

In consideration of the second issue, whether a particular clause amounted to a forfeiture clause was a matter of substance, not form. The Tenancy Agreement allowed Croydon to seek an order for possession from the Court “at any time” if Ms Kalonga breached the terms of the Tenancy Agreement. The Supreme Court held that was to be considered a forfeiture clause if you apply the test established in Clays Lane Housing Co-operative Ltd v Patrick [1985] 17 HLR 188, and the lower courts had been wrong to conclude otherwise.

The Supreme Court unanimously concluded that the Tenancy Agreement did contain a forfeiture clause. However, as Croydon had stated that it was not relying on a right of forfeiture, the appeal was allowed in part.


There is thought to be somewhere in the region of 30,000 fixed term secure tenancies in England, therefore this is a significant case setting out the position in terminating a flexible tenancy.

Whilst the outcome only applies to flexible tenancies granted by local authorities, it is important to note that private registered fixed term tenancy agreements do need to contain a provision to determine the tenancy by way of provision for re-entry to ensure compliance with section 7(6)(b) of Housing Act 1988.


Case Comment: R (on the application of Z) v Hackney LBC [2020] UKSC 40

In this post, Mathew Purchase QC of Matrix Chambers shares his views on the Supreme Court’s decision in R (on the application of Z) v Hackney LBC [2020] UKSC 40. The Supreme Court unanimously dismissed an appeal about the application of anti-discrimination law to charities, where they are established to provide benefits (in this case, social housing) for particular groups which are the subject of their charitable objectives. The relevant anti-discrimination laws are contained in the Equality Act 2010 and Council Directive 2000/43/EC of 29 June 2000 (the “Race Directive”).



The narrow result of this appeal is that, on the facts, it was proportionate and lawful for a charity to restrict the allocation of its housing stock to Orthodox Jewish families. However, in reaching that conclusion, Lord Sales, giving the leading judgment, made a number of points of wider importance.



The Agusas Israel Housing Association (‘AIHA’) is a charity whose objectives include providing housing ‘primarily for the benefit of the Orthodox Jewish Community’. In practice, the demand from the Orthodox Jewish community is so high that, under current market conditions, AIHA will only consider applications from that community. As a result, Hackney London Borough Council will refer families on its waiting list to AIHA only if they are Orthodox Jewish.

The claimant is a mother with four children, two of whom were very young and two of whom had autism spectrum disorders. She needed re-housing into a larger property. AIHA properties became available but they were allocated to members of the Orthodox Jewish community, which she was not. She accordingly had to wait longer than she otherwise would have for a suitable home.

That approach constitutes direct discrimination on the ground of religious belief, contrary to section 29 of the Equality Act 2010 (‘the 2010 Act’) read with section 13. However, section 158 of the 2010 Act permits such discrimination if it is ‘a proportionate means of achieving the aim of… enabling or encouraging persons who share [a] protected characteristic to overcome… a disadvantage [or] meeting… needs that are from the needs of persons who do not share [the protected characteristic]’. Further, section 193 of the 2010 Act permits a charity to provide benefits which would otherwise be discriminatory if they are ‘a proportionate means of achieving a legitimate aim’ (section 193(2)(a)) or are ‘for the purpose of preventing or compensating for a disadvantage linked to the protected characteristic’ (section 193(2)(b)).

The Divisional Court held that it was proportionate to restrict access to the Orthodox Jewish community in order to overcome disadvantages they faced in securing appropriate housing and to meet their particular needs. They had a particular need for large houses because of their large family sizes, living as a community was central to their way of life, they experienced high levels of poverty, and as a group they faced prejudice and discrimination when seeking accommodation in the private sector. Accordingly, AIHA (and by extension) Hackney’s approach was lawful under sections 158 and 193 of the 2010 Act.

The main issues in the appeal were (i) whether the Divisional Court was wrong to hold that those defences were satisfied on the facts and (ii) whether they had to be read down or disapplied in order to comply with EU law or under section 3 of the Human Right Act 1998 (‘the 1998 Act’).



The Supreme Court upheld the findings of the Divisional Court. Those findings are instructive but have little wider legal importance. However, the claimant developed a number of arguments which did raise issues of wider public importance. In particular, it was argued that:

The section 158 and 193 defences should be interpreted generally in light of EU law. It is impermissible under EU law to engage in ‘positive action’ which would otherwise constitute discrimination unless: it is aimed at achieving equality of opportunity rather than outcome, it is limited to the different treatment of persons who are in an equivalent position apart from their protected characteristic, and it allows for exceptions on individual facts.
The claimant’s treatment was on the ground of race as well as on the ground of religious belief, such that the Race Directive 2000/43/EC applied. Accordingly, EU law applied directly so as to restrict the scope of those defences. (There is a directive preventing discrimination on the ground of religious belief in employment and occupation, but not in the provision of goods and services.)
The claimant’s treatment also fell within Articles 8 and 14 of the European Convention on Human Rights (‘the Convention’) so that it was impermissible to allow positive action without consideration of its proportionality on the facts of the particular case (meaning that section 193(2)(b) of the 2010 Act was too wide and had to be interpreted so as to incorporate such a requirement).
Whatever the claimant’s position, in other cases the Race Directive would be engaged and so the legislation had to be interpreted or disapplied so as to be compatible with it.

In rejecting those arguments, the Supreme Court made some important statements of principle. These included the following.

First, the approach to ‘proportionality’ under section 158 and 193 is the conventional one. It is legitimate to pursue equality of outcome as well as equality of opportunity and it may be proportionate to apply a ‘bright line rule’ which focusses on the position of identifiable groups rather than on the particular circumstances of individuals. The claimant’s reliance on the narrower approach mandated by case law under the Equal Treatment Directive 76/207/EEC was misplaced. That Directive applied to discrimination on grounds of sex in employment and occupation – which is covered by section 159 of the 2010 act – whereas other EU legislation allowed for much wider exceptions for positive action. (Note also that the Equal Treatment Directive has itself been replaced by Directive 2006/54/EC, which contains a wider exception for positive action than its predecessor.) [58-72, 79-82, 85-86]
Secondly, provided the first instance court has applied the correct legal approach, an appellate court should interfere only if its decision was ‘wrong’ in the sense of containing a significant error of principle or an identifiable flaw in reasoning, such as an error of logic or a failure to take account of a material factor. This test is derived from the Civil Procedure Rules, but should also be applied by the Supreme Court. [56, 74]
Thirdly, on the available evidence, the discrimination in the present case was on the ground of religious belief and not on the ground of race. Sometimes, a person’s status as a Jew may be a racial status based on ethnic origin: see, for example, R (E) v Governing Body of JFS [2010] 2 AC 728 (children who were recognised as Jewish as a result of matrilineal descent) and Mandla v Dowell Lee [1983] 2 AC 548 (distinct groups with, as a minimum, a long shared history and cultural tradition of their own). However, the criterion applied by AIHA was based solely on religious observance as an Orthodox Jew. The Supreme Court left open the possibility that, in a different case, evidence could establish that this group was a racial group in the Mandla However, no such evidence had been adduced in the present case, the point not even having been raised at first instance. [89-94]
Fourthly, Parliament’s decision to provide in section 193(2)(b) of the 2010 Act that it was permissible for a charity to take ostensibly discriminatory action ‘for the purpose of preventing or compensating for a disadvantage linked to [a] protected characteristic’ without any additional requirement of proportionality did not breach the Convention. Parliament’s decision to apply a general rule of this nature was itself proportionate in pursuit of the aim of allowing more charitable resources to be directed towards those who need them. [92-110]
Fifthly, the ‘margin of discretion’ to be accorded to Parliament in this context was wide because (a) ‘Parliament… had its attention drawn to the competing interests and for the need for the regime it enacts to strike a balance which is fair and proportionate and has plainly legislated with a view to satisfying that requirement’ and (b) the test of whether the approach is ‘manifestly without reasonable foundation’, which applies to ‘general measures of economic or social strategy’, also applies to decisions concerning the ability of charities to allocate funds to meet a range of needs. [108-109]
Sixthly, section 193(2)(b) of the 2010 Act could not be interpreted under section 3 of the 1998 Act or under the Marleasing principle to include a proportionality test in any event. Parliament had purposely decided not to include a proportionality test in that subsection, in contrast to the general test set out in subsection (a). If a proportionality test were included, it would render subsection (b) otiose as all such cases would fall within section (a) as well. (The Court did not address the disapplication of domestic legislation to comply with EU law but, as the following point explains, this did not arise.) [111-112]
Finally, on the facts, there was no breach of EU law in the claimant’s case. And, although it is possible that, in other cases, the application of section 193(2)(b) might raise issues under the Race Directive: ‘The proper approach to construction is that legislation should be read and given effect in a particular case according to its ordinary meaning, unless the person who is affected by it can show that this would be incompatible with their Convention rights under the HRA or some provision of EU law as applied to their case. Only then do the special interpretive obligations under section 3(1) of the HRA or under the Marleasing principle come into play to authorise the court to search for a conforming interpretation at variance with the ordinary meaning of the legislation. This means that the same legislative provision might be given a different interpretation in different cases, depending on whether Convention rights or EU law are applicable in the case or not’. [113-114]

Case Comment: Bott & Co Solicitors v Ryanair DAC [2022] UKSC 8

In this post, Mark Chapman, Lauren Cousins and Jessica Eaton, all associates at CMS, comment on the decision of the UK Supreme Court in Bott & Co Solicitors v Ryanair DAC [2022] UKSC 8.  On 16 March 2022, the Supreme Court, by majority, allowed the appeal and ruled for Bott & Co Solicitors. The decision concerns the extent of the solicitor’s equitable lien.


The case brought by Bott & Co Solicitors (“Bott”) against Ryanair DAC (“Ryanair”) concerns the extent of the solicitor’s equitable lien, a remedy that has been recognised by the courts for over two hundred years. A helpful history of the lien is set out in the introductory paragraphs of the dissenting judgment of Lord Leggatt and Lady Rose:

“In its traditional form, it entitled a solicitor who assists a client to recover money (or other property) through litigation to recoup the costs of doing so out of the money recovered. Any proceeds of a judgment or settlement will normally be paid to the solicitor’s firm, which can then deduct its costs before accounting to the client for the balance. But if the opposing party pays the money directly to the solicitor’s client despite knowing or being on notice of the solicitor’s interest in the debt, and the client then fails to pay the solicitor’s costs, the court may order the opposing party to pay those costs to the solicitor – in addition to the payment already made to the solicitor’s client.”

For much of the early history of the lien, it was thought that it arose only where there were court (or arbitration) proceedings in existence when the money was recovered. However, following the Supreme Court’s decision in Gavin Edmondson Solicitors Ltd v Haven Insurance Co Ltd [2018] UKSC 21 (“Gavin Edmondson Solicitors SC”), it is clear that the lien can now arise even in circumstances where no formal proceedings have been raised.

The present appeal concerned whether Bott were entitled to the lien to recover the costs incurred for claiming compensation for flight delays from Ryanair on behalf of customers.

The facts that gave rise to the present proceedings are simple. Some years ago, Bott developed an online tool, accessible by their website, whereby delayed passengers could enter their flight information to check whether they were eligible for compensation from Ryanair (and other airlines) in terms of EU Regulation 261 (“Regulation 261”). Regulation 261 applies where a passenger is leaving from or arriving in an EU member state with an EU airline and empowers customers to claim compensation where their flight is delayed, except in certain exceptional circumstances. If Bott’s software confirmed eligibility, and the individual confirmed their desire to instruct Bott on a “no win, no fee” basis to pursue the compensation claim, Bott would then send a standard form letter to the airline to ask whether the claim would be admitted or resisted. If the claim was admitted, Bott would instruct the airline to pay the compensation due into its client account whereupon Bott would deduct its fee and pass on the difference to their client/the delayed passenger.

In 2016, perhaps in response to Bott’s software, Ryanair set up its own online compensation portal whereby it began to compensate Bott’s client’s directly, thereby avoiding payment of Bott’s fee. Bott’s evidence before the Court was that they only had a 30% success rate in recovering their small fee (on average about £95) from their clients who had received their compensation direct from Ryanair. Given the sums involved, it wasn’t financially viable for Bott to pursue their clients in court for their unpaid fees. Bott then brought proceedings against Ryanair claiming a lien over the sums payable by Ryanair to Bott’s clients.

Decisions of the lower courts

The first instance decision handed down by Mr Edward Murray, sitting as a judge in the Chancery Division, rejected Bott’s claim to a lien over the compensation paid directly to Bott’s clients by Ryanair. Mr Murray held he was bound by the decision in Meguerditchian v Lightbound [1917] 2 KB 298 and the Court of Appeal decision in Gavin Edmondson Solicitors Ltd v Haven Insurance Co Ltd [2015] EWCA Civ 1230 that held a lien does not arise where compensation is obtained without the need to issue proceedings. Shortly after Mr Murray’s decision, the Supreme Court decided Gavin Edmondson Solicitors SC and found that it was not necessary for formal proceedings to have been issued for the lien to arise.

On appeal, Bott asked the Court of Appeal to consider whether Mr Murray was wrong to reject Bott’s claim and argued that Gavin Edmondson Solicitors SC now answered the question of lien in its favour.

The Court of Appeal unanimously dismissed Bott’s appeal, Lewison LJ gave the leading judgment, with which Simon and Lindblom LJJ agreed. Lewison LJ held that for the lien to arise Ryanair had to dispute the claim for compensation under Regulation 261 and the act of writing a letter of claim or assisting with an online form was not sufficient for Bott to establish the lien.

Lewison LJ recognised the decision in Gavin Edmondson Solicitors SC, however, he considered that the services being provided by the solicitor “must still be recognisable as litigation services, promoting access to justice”. Lewison LJ considered, by way of reference to the definition of “litigation services” in s 119 of the Courts and Legal Services Act 1990 (meaning any services which it would be reasonable to expect a person who is exercising, or contemplating exercising, a right to conduct litigation in relation to any proceedings, or contemplated proceedings, to provide) and “contentious business” in s 87 of the Solicitors Act 1974 (meaning business done, whether as solicitor or advocate, in or for the purposes of proceedings begun before a court or before an arbitrator. . ., not being business which falls within the definition of non–contentious or common form probate business contained in s 128 of the Senior Courts Act 1981) that for the term “litigation” to apply there had to be a dispute, which could also include arbitration, proceedings under a pre-action protocol, and potentially even in relation to the costs incurred by way of alternative dispute resolution. Therefore, the process undertaken by Bott in claiming for compensation under Regulation 261, being largely administrative and formulaic, did not constitute litigation, and would not allow for an equitable lien until and unless Ryanair disputed the claim for compensation.

Bott appealed to the Supreme Court.

Summary of Supreme Court’s findings

By majority, the Supreme Court allowed Bott’s appeal. The Court held that for the lien to arise, the solicitor must provide services (within the scope of the retainer with its client) in relation to the making of a client’s claim (with or without legal proceedings) which significantly contribute to the successful recovery of a fund by a client.

In reaching their decision, the Court found; (i) there must still be limits on when an equitable lien can be exercised to avoid it arising in non-contentious matters (such as a conveyance); (ii) the Court of Appeal was wrong to ‘fix the boundary’ (i.e. as to when the lien does and does not apply) by reference to statutory definitions; (iii) the key to ‘fixing the boundary’ is to consider the underlying purpose of promoting access to justice and; (iv) the reason for the extension of the solicitor’s lien beyond matters where proceedings have been issued is due to the (relatively recent) development of various forms of ADR – otherwise solicitors would be incentivised to issue claim forms at an early stage to make sure they get paid. That would be contrary to the rationale behind the pre-action protocol and the idea that court proceedings should be a last resort.


The test set out above is clearly a wide one and will make it difficult for typical defendants, such as Ryanair, to attempt to cut solicitors out of the process by paying compensation to claimants or potential claimants direct. However, it is important to note that the lien will only apply where the claim has not been admitted by a prospective defendant. In other words, where a prospective defendant has made public their intention to comply with the legal obligation (in this case to compensate delayed customers), then it is unlikely that the lien would be effective as the solicitor would probably find it difficult to prove that any subsequent assistance made a “significant contribution to the successful recovery of a fund by a client”.  

Interestingly, Lord Briggs (who joined the majority), noted that

“the simple test proposed by Lady Arden and Lord Burrows may occasionally involve the recognition of the equitable lien in wider circumstances than is strictly justified by its animating access to justice principle. It may also confer upon solicitors a proprietary security for payment for services with no very sophisticated legal content which other providers of the same services do not enjoy, simply because they are not solicitors. The facts of this case illustrate both those concerns. Bott went on providing their flight delay service to Ryanair’s passengers long after Ryanair introduced its own apparently equally simple way of obtaining prompt payment of compensation in full. Other claims-handlers entered the market for low-cost claims services with products similar to Bott’s scheme who, because they were not solicitors, lacked the protection of the lien.”

Ultimately, however, Lord Briggs acknowledged that while there might be room to extend the lien to other legal service providers, that was a development best left to Parliament. It will be interesting to see whether this lien is extended to non-solicitor service providers by legislation in due course.

Case Preview: Harpur Trust v Brazel

In this post, Hattie Ryland, an Associate at CMS, previews the decision awaited from the UK Supreme Court in the matter of Harpur Trust v Brazel. 

On 9 November 2021, the Supreme Court heard the appeal in Harpur Trust v Brazel. The forthcoming decision is expected to provide some much-needed clarity on how employers should approach calculating annual leave entitlement and pay for workers who work irregular hours, including those workers on zero hours contracts.

Factual background

The respondent, Ms Brazel, is a clarinet and saxophone teacher. She is employed by the appellant (the “Trust”), which runs Bedford Girls School, as a “visiting music teacher”. She is employed by the Trust on a zero-hours contract, meaning that she does not have a set number of working hours; the hours she works in a given school term depend on the number of pupils requiring tuition in her instruments. She is paid monthly on the basis of an agreed hourly rate applied to the hours worked in the previous month.

As a worker within the meaning of the Working Time Regulations 1998 (“WTR”), Ms Brazel is entitled to 5.6 weeks’ paid annual leave. Since the school holidays are longer than that, the Trust did not designate any particular parts of them as statutory leave. Instead, it made three payments in lieu of holiday in respect of each term, in December, April and August.

In 2011, the Trust altered the manner in which it calculated those holiday payments. It purported to follow a method recommended in then available ACAS and Government guidance for calculating the holiday entitlement of casual workers. The guidance suggested that for workers with casual or irregular hours, holiday entitlement could be calculated as 12.07% of hours worked. This was on the basis that 5.6 weeks is equivalent to 12.07% of hours worked over a year (5.6 weeks’ holiday, divided by 46.4 weeks (being 52 weeks – 5.6 weeks) multiplied by 100 = 12.07%) (the “12.07% Method”). The employer will then pay the normal hourly rate for that holiday, as and when it is taken.

The Trust adopted the 12.07% Method, which had the effect of Ms Brazel receiving a smaller sum in respect of her holiday pay than she had previously been receiving from the Trust. She brought proceedings for unlawful deductions from her wages by underpayment of her entitlement to holiday pay.

The legal issue explained

The issue in this case before the Supreme Court is whether the 12.07% Method is compliant with the 5.6 weeks’ paid holiday entitlement under the WTR.

Under the WTR annual leave entitlement is based on a proportion of the number of weeks in the holiday year that the individual has been engaged under their contract. Where the worker has been engaged under a contract for the full 52 weeks, their entitlement will be to 5.6 weeks’ annual leave. This is the case regardless of how many (if any) hours are worked in a particular week; all that matters is that the contract between the employer and the worker remains in place. The worker is then entitled to their “normal pay” for that holiday. For workers with no fixed hours, this will be calculated on the basis of their average pay over the previous 52 paid weeks before the date they take their holiday (Employment Rights Act 1996, s 224) (“ERA”) (at the relevant time, the reference period was the previous 12 paid weeks). Weeks in which no remuneration is earned are ignored, and earlier weeks are brought into the calculation to make up the requisite 52 weeks, up to a maximum of 104 weeks before the calculation date.

The result is that the longer the period of the year that the worker is engaged but undertakes no work (and receives no pay), the bigger the disparity between the 12.07% Method and the approach under the WTR. As an extreme example (which was put forward by counsel for the Trust in the proceedings), the apparent result of the WTR is that a worker engaged on a permanent contract, but who works only one week of the year, for which they earned £1,000, would then be entitled to 5.6 weeks (notional) annual leave, for which they would receive £5,600. Under the 12.07% Method, they would receive £120.70.

Appellate history

Ms Brazel brought proceedings in 2015 before the Employment Tribunal for unlawful deductions from her wages by underpayment of her entitlement to holiday pay. The Employment Tribunal decided against her on this issue, holding that the application of the 12.07% Method to either the length of the holiday entitlement or to what it described as Ms Brazel’s: “average pay over the course of the working year of 46.6 weeks” would give her proportionately the same holiday entitlement as a full-year worker.

Ms Brazel then appealed to the Employment Appeal Tribunal (“EAT”), where she was successful. The EAT held that there was no reason to depart from the plain statutory language; the intention of the WTR is that all workers are entitled to 5.6 weeks’ paid holiday and there was no basis for pro-rating that entitlement (irrespective of whether or not this favoured part-year workers over full-time workers).

The Trust brought the issue before the Court of Appeal in 2019.

Key Court of Appeal submissions

The Trust’s fundamental submission was that pro-rating of Ms Brazel’s holiday / holiday pay entitlement to her salary, using the 12.07% Method, was necessary in order to avoid unjust results that cannot have been intended; the holiday pay to which, on her case, she was entitled would be a much higher proportion of her actual earnings than if she worked full-time. On the basis of a 32-week working year it would be 17.5%, while the holiday pay of a full-year worker would only be 12.07% of their earnings. Even greater anomalies would be apparent for employees on permanent contracts who worked for even smaller proportions of the year, as noted above. The WTR right to paid annual leave (or at least the right to four weeks’ paid leave) is derived from the European Working Time Directive (“WTD”). The Trust submitted that relevant Court of Justice for the European Union authorities, in particular Greenfield v The Care Bureau Ltd (C-219/14), established that, as a matter of EU law, entitlement to annual leave accrued in step with the relevant units of work, so that if an employee performed less than a full year’s work, they should get less than a full year’s holiday entitlement / pay.

Ms Brazel’s position was that the Trust’s method of calculating her holiday / holiday pay bore no relation to the calculation required by the WTR and produced a lower figure than that required by law under the WTR. It was common ground that she had no normal working hours within the meaning of the ERA, and, accordingly, section 224 applied in determining the amount of “a week’s pay”. She argued that, by virtue of regulation 16(1) WTR, which incorporated s 224 ERA, the correct approach was to calculate a week’s pay by taking her average weekly remuneration for the 12 weeks prior to the calculation date in which she received pay (the process laid out in s 224 ERA as it stood at that time), and then, by virtue of regulation 13 and regulation 13A WTR, to multiply it by 5.6.

Decision of the Court of Appeal

The Court of Appeal dismissed the appeal and held that, despite Ms Brazel not working the full year, she was still entitled to 5.6 weeks’ holiday and her holiday pay should not be limited to 12.07% of her salary.

The Court of Appeal noted that the WTD does not prescribe any particular mechanism for the assessment of holiday pay entitlement, and article 15 expressly provides that member states may accord workers entitlements which are more favourable than those required by the WTD itself. Therefore, even if the WTR provides for a different model for assessing entitlement to annual leave that results in favourable treatment of part-year workers, this is not in itself an issue, so long as it accords workers the minimum rights required by the WTD.

Addressing the potentially extreme results put forward by the Trust’s counsel, the Court of Appeal accepted these, but noted that general rules sometimes produce such anomalies when applied in untypical cases. It also noted that it would be unusual for a worker who worked so little proportion of the year to be on a permanent contract, as opposed to being engaged on a freelance basis. The Court of Appeal noted that, because Ms Brazel was engaged on a permanent contract, it did not seem unreasonable to treat that as a sufficient basis for fixing the quantum of holiday entitlement (at 5.6 weeks where the contract is in place for the full year), irrespective of the number of hours, days or weeks that the worker may in fact have to perform under the contract.


The Supreme Court’s judgment is eagerly anticipated, because we hope it will provide much need clarity for employers who engage workers on irregular hours.

If the Court of Appeal’s decision is overturned, it will legitimise the practice of calculating holiday entitlement as 12.07% of pay for irregular workers. However, if the Court of Appeal’s decision is upheld, it will confirm the anomaly that those who work only part of the year might end up receiving more pay for their holiday than those who work full time. Moreover, given the prevalence of the 12.07% Method currently, employers will have to evaluate their practices when it comes to calculating holiday pay and change them (if they have not already done so); and may face claims from affected workers in respect of underpaid holiday pay. It may also lead employers to reconsider the use of permanent contracts for casual workers, instead engaging such workers on short-term contracts as and when required.

Scroll to top