United Kingdom Supreme Court

You are here: BAILII >> Databases >> United Kingdom Supreme Court >> Recovery of Medical Costs for Asbestos Diseases (Wales) Bill (Reference By The Counsel General For Wales) [2015] UKSC 3 (9 February 2015)
URL: http://www.bailii.org/uk/cases/UKSC/2015/3.html
Cite as: [2015] HRLR 9, [2015] 1 AC 1016, [2015] UKSC 3, [2015] AC 1016, [2015] 2 WLR 481, [2015] Lloyd's Rep IR 474, [2015] WLR(D) 67, (2015) 143 BMLR 1, 143 BMLR 1, [2015] 2 All ER 899

[New search] [Printable PDF version] [View ICLR summary: [2015] WLR(D) 67] [Buy ICLR report: [2015] 2 WLR 481] [Buy ICLR report: [2015] 1 AC 1016] [Help]

Summary

Hilary Term
[2015] UKSC 3

JUDGMENT

RECOVERY OF MEDICAL COSTS FOR ASBESTOS DISEASES (WALES) BILL: Reference by the Counsel General for Wales

before

Lord Neuberger, President
Lady Hale, Deputy President
Lord Mance
Lord Hodge
Lord Thomas

JUDGMENT GIVEN ON

9 February 2015

Heard on 14 and 15 May 2014

Appellant

Theodore Huckle QC

Richard Gordon QC

(Instructed by Welsh Government Legal Services Department)

Intervener

Michael Fordham QC

Jason Pobjoy

(Instructed by Bircham Dyson Bell)

LORD MANCE (with whom Lord Neuberger and Lord Hodge agree)

Introduction

This reference, made by the Counsel General for Wales, raises for determination whether the Recovery of Medical Costs for Asbestos Diseases (Wales) Bill is within the legislative competence of the National Assembly for Wales (the "Welsh Assembly"). The issues involved are novel and important, and the Counsel General was right to recognise them as such and to make the present reference with a view to resolving them. The reference has been well-presented and argued on both sides.

The Bill contains in section 1 its own "overview". It

"(a) imposes liability on persons by whom or on whose behalf compensation payments are made to or in respect of victims of asbestos-related diseases to pay charges in respect of National Health Service services provided to the victims as a result of the diseases;

(b) makes provision for the certification of the amount of the charges to be paid, for the payment of the charges, for reviews and appeals and about information;

(c) extends insurance cover of liable persons to their liability to pay the charges."

Liability to pay NHS charges arises under section 2 "where a compensation payment is made to or in respect of a person (the 'victim') in consequence of any asbestos-related disease suffered by the victim". It is imposed on the person who is or is alleged to be liable to any extent in respect of such disease and by whom or on whose behalf the compensation payment is made after the Bill comes into force. It is convenient to describe such a person as the "compensator". The liability is to reimburse the Welsh Ministers in respect of any relevant Welsh NHS services provided to the victim as a result of the disease, in an amount set or amounts out in, or determined in accordance with, regulations under section 6(2) and specified in a certificate to be issued by the Welsh Ministers subject to any limit fixed by regulations under section 6(5)(a). The Bill contains extensive provisions requiring sufferers, compensators and others to provide information (section 12), requiring compensators to apply for and the Welsh Ministers to issue certificates specifying the relevant charges arising under section 2 in accordance with regulations and reduced where appropriate to reflect any contributory fault on the part of the sufferer (section 6) as well as regulating other matters, such as the time for payment of charges (section 7), the recovery of charges (section 8), the review of certificates (section 9), appeals against certificates (section 10 and 11) and cases in which compensators make lump sum or periodical payments (section 13).

Section 14 deals with the liability of insurers. It provides:

"(1) Where the liability or alleged liability of the person by whom or on whose behalf a compensation payment is made is, or (if established) would be, covered to any extent by a policy of insurance, the policy is to be treated as covering the person's liability under section 2.

(2) Liability imposed on the insurer by subsection (1) cannot be excluded or restricted.

…

(5) This section applies in relation to policies of insurance issued before (as well as those issued after) the date on which this section comes into force.

(6) References in this section to policies of insurance and their issue include references to contracts of insurance and their making."

Section 15 provides:

"(1) The Welsh Ministers must, in the exercise of their functions under the National Health Service (Wales) Act 2006, have regard to the desirability of securing that an amount equal to that reimbursed by virtue of section 2 is applied, in accordance with that Act, for the purposes of research into, treatment of, or other services relating to, asbestos-related diseases.

(2) The Welsh Ministers must report annually to the National Assembly for Wales on the application of amounts equal to sums reimbursed by virtue of section 2."

The Bill in these circumstances has the following characteristics:

(i) First, by section 2, it imposes a novel statutory or "quasi-tortious" liability towards the Welsh Ministers on compensators (defined as set out in para 3 above).

a. This liability is a liability for pure economic loss which does not exist and has never existed at common law.

b. It does not reflect any liability which the compensator had to the victim, since the victim has no liability to the Welsh Ministers to meet any economic loss the Welsh Ministers may have suffered.

c. The liability exists whether the compensation is paid to the victim with or without admission of liability; the Counsel General in written submissions states that a "key point is that it is a necessary condition of the Bill attaching to insurers that there must be liability established or conceded". But a payment without admission of liability does not in law or even de facto amount to a concession of liability.

d. The liability is based on future compensation payments made in respect of actual or potential wrongs, the operative elements of which were committed many decades ago, though the victims are or will only suffer the consequences and the Welsh National Health Service will only have to bear the hospitalisation costs in the future.

(ii) Second, by section 14, the Bill imposes a new contractual liability on the liability insurers of compensators (typically employers' liability insurers such as those involved in the Trigger litigation: Durham v BAI (Run-Off) Ltd [2012] UKSC 14, [2012] 1 WLR 867) to cover any liability which such compensators have as a result of section 2.

a. It imposes this new liability on any insurer whose policy would to any extent cover the compensator for any liability which the compensator has or would (if established) have towards the victim.

b. It imposes it irrespective of any policy exclusion or restriction.

c. It imposes it in relation to policies issued before as well as after the date section 14 comes into force - and so in relation to policies issued and covering events occurring many decades ago.

d. It does all this although - indeed no doubt because - such liability insurers would not otherwise be likely to have to answer for any charges levied under section 2. This is clear on any reading of the typical employers' liability policy wordings summarised in annex A to my judgment in the Trigger case. In essence, such policy wordings cover employers' liability in damages for claims by actual or former employees suffering injury or disease. They are, furthermore, triggered by the original exposure to asbestos during the course of the insurance, not by the imposition of charges under section 2 as a result of compensation payments made, with or without admission of liability, long after the expiry of the policy period.

(iii) Third, section 15 provides that the Welsh Ministers must, in the exercise of their functions under the National Health Service (Wales) Act 2006 ("the NHS (Wales) Act"), have regard to the desirability of securing that an amount equal to that reimbursed by virtue of section 2 is applied, in accordance with that Act, for the purposes of research into, treatment of, or other services relating to, asbestos-related diseases.

The Bill thus imposes new liabilities on compensators in respect of past conduct and on liability insurers under past insurance contracts. The Counsel General stresses that compensators would only incur such liabilities as a result of their making future compensation payments to or in respect of victims of asbestos-related diseases who suffer future hospitalisation; and that insurers would only incur such liabilities under such contracts upon such compensation payments being made and then only if such contracts would to some extent cover any liability which the compensator might have towards the asbestos-related disease sufferer to make such compensation payments. The Bill is thus not retrospective in the fullest sense, but it does significantly restructure both the consequences of actual or possible negligence or breach of statutory duty committed long ago by compensators, and the terms of and liabilities attaching under insurance policies also underwritten years ago to cover any such negligence or breach of duty.

Unsurprisingly, in view of the identity of the interveners, the Association of British Insurers, the primary focus of submissions before the Supreme Court has been on section 14 of the Bill. But, inevitably, attention has also had to be given to the aim and effect of other provisions of the Bill, particularly section 2, which is directed to compensators.

The question referred to the court subdivides into two more specific issues: whether the Bill, and in particular, but not exclusively, section 14, falls within section 108(4) and (5) of the Government of Wales Act 2006 ("GOWA"), which in turn depends in this case upon whether it relates to "Organisation and funding of national health service" in paragraph 9 of Part 1 of Schedule 7 to GOWA – an issue on which section 15 has a potential bearing; and whether, if it does fall within section 108(4) and/or (5), it is nonetheless outside the Welsh Assembly's competence by virtue of section 108(6), read with section 158(1), on the ground that it is incompatible with the Convention rights scheduled to the Human Rights Act 1998. It is logical to take these issues in that order, since section 108(6) operates as a restriction on the Assembly's legislative competence in respect of matters which fall within section 108(4) and/or (5). The Counsel General must however succeed on both issues in order to make good his submission that the Bill is within the Assembly's legislative competence.

The issue whether the Bill falls within section 108(4) and/or (5) was not originally raised by the interveners or therefore addressed in the Counsel General's written case. It was nonetheless raised squarely in the interveners' written case, and has been covered by oral submissions and written notes on both sides.

Competence under section 108(4) and (5)

Consequent upon the referendum held in 2011 under section 105(1) of GOWA, the competence of the Welsh Assembly is no longer determined by section 94 read with Schedule 5 to the Act. Section 94 has, along with the rest of Part 3 of the Act, ceased under section 106(1) to have effect. Instead the Welsh Assembly has (since 5 May 2011: see The Government of Wales Act 2006 (Commencement of Assembly Act Provisions, Transitional and Saving Provisions and Modifications) Order 2011 No 1011 (W.150)) had the expanded legislative competence provided by sections 108 and 109 read with Schedule 7. Under section 108(3) a provision is only within the Assembly's legislative competence if it falls within subsection (4) or (5). A provision falls within section 108(4)

"if it relates to one or more of the subjects listed under any of the headings in Part 1 of Schedule 7 and does not fall within any of the exceptions specified in that Part of that Schedule …."

A provision falls within section 108(5)

"if (a) it provides for the enforcement of a provision (of that or any other Act of the Assembly) which falls within subsection (4) or a provision of an Assembly Measure or it is otherwise appropriate for making such a provision effective, or (b) it is otherwise incidental to, or consequential on, such a provision."

The relevant matter specified in Part 1 of Schedule 7 on which reliance is placed to establish competence to enact the Bill is para 9 headed "Health and health services" and reading:

"Promotion of health. Prevention, treatment and alleviation of disease, illness, injury, disability and mental disorder. Control of disease. Family planning. Provision of health services, including medical, dental, ophthalmic, pharmaceutical and ancillary services and facilities. Clinical governance and standards of health care. Organisation and funding of national health service.

Exceptions -

Abortion.

Human genetics, human fertilisation, human embryology, surrogacy arrangements.

Xenotransplantation.

Regulation of health professionals (including persons dispensing hearing aids).

Poisons.

Misuse of and dealing in drugs.

Human medicines and medicinal products, including authorisations for use and regulation of prices.

Standards for, and testing of, biological substances (that is, substances the purity or potency of which cannot be adequately tested by chemical means).

Vaccine damage payments.

Welfare foods.

... Health and Safety Executive and Employment Medical Advisory Service and provision made by health and safety regulations."

The critical phrase is "Organisation and funding of national health service". The questions arise, firstly, whether this covers the imposition of a statutory liability on compensators who were or are alleged to have been wrongdoers, and, secondly, if it does, whether it also covers the amendment of any insurance contracts which would cover such compensators "to any extent" for any liability they had to the sufferers of the relevant asbestos-related disease, so as to make the relevant insurers answer for any compensation payment made irrespective otherwise of the terms of the insurance contract.

These questions raise for consideration the vires of the core elements of the Bill under section 108(4) and para 9. But, if the conclusion is that section 2 does, but section 14 does not, fall within section 108(4) and para 9, then the question still arises whether section 14 can be regarded as providing for the enforcement of that provision or as being "otherwise appropriate for making such a provision effective or … otherwise incidental to, or consequential on, such a provision" within section 108(5).

On behalf of the interveners, Mr Michael Fordham QC submits that para 9 gives general competence to regulate the Welsh NHS, the services which it provides and the standards its meets, but that it lacks, noticeably, any provision enabling charging for such services. The phrase "Organisation and funding of national health service" concerns, in his submission, the allocation by the Welsh Ministers of monies to fund the Welsh NHS and their control of spending by the Welsh NHS of any other monies available to it under (now) section 175 of the NHS (Wales) Act, enacted by the United Kingdom Parliament on 8th November 2006, just over three months after GOWA. He submits that there is nothing in para 9 to suggest any wider meaning.

More specifically, on the interveners' case, para 9 gives general competence in areas such as those dealt with specifically in Chapter 6 (Finance) of Part 11 of the NHS (Wales) Act. Chapter 6 provides that the Welsh Ministers are to decide what funds to allot to Special Health Authorities, what directions to give or conditions to attach regarding such funds (section 171) and what duties and resource limits to impose on such Authorities (sections 172 and 173). It further identifies what funding the Welsh Ministers must in each financial year provide to each Local Health Board (section 174) and the financial duties and resource limits to which such Boards are subject (sections 175 and 176) and makes further provision about the expenditure of such Boards (section 177 and Schedule 8). Exercising the competence provided in these areas, the Welsh Assembly has by the National Health Service Finance (Wales) Act 2014 recently amended section 175, to provide for each Local Health Board to balance its expenditure and income in each three-year accounting period, rather than in respect of each financial year as originally enacted. But what para 9 is not, Mr Michael Fordham QC submits, is a provision which itself enables the Welsh Assembly to impose (or authorise the Welsh Ministers to impose) charges on anyone either for Welsh NHS services or on any other basis.

It is common ground that the Welsh Ministers do not have (and the Welsh Assembly does not have and cannot confer) general fiscal powers, an exception noted expressly in relation to economic development in paragraph 4 of Part 1 of Schedule 7 GOWA. The Welsh Government has large spending powers, but its funding of the services it supports is, at present, fundamentally dependent on the United Kingdom's block grant. The Welsh Assembly has limited powers or control in respect of business rates and council tax, in which connection the reference to "Local government finance" in paragraph 12 of Schedule 7 is relevant. That paragraph gives competence (subject to exceptions which I need not set out here) in respect of the

"Constitution, structure and areas of local authorities. Electoral arrangements for local authorities. Powers and duties of local authorities and their members and officers. Local government finance."

The framework within which business rates and council tax are charged is provided by the Local Government Finance Act 1988, as amended in 1992 and 2012. Such taxes are payable to the relevant local government authorities, not to the Welsh Ministers. The reference to "Local government finance" enables the Welsh Assembly, for example, to determine the level of business rates or limit council tax increases chargeable under these statutes (though, under the block grant system, this does not appear to affect the overall level of funding available to the Welsh Government). But, it cannot on any view be read as a general power enabling the Welsh Assembly to raise funds in any way it may decide, even if such funds are ear-marked for use to support local government activities.

In support of a generous interpretation of the concept of "Organisation and funding of national health service", the Counsel General drew attention to the previous legislative competence under section 94 and Schedule 5 of GOWA, to enact measures relating to the red meat industry in relation to increasing efficiency or productivity, improving marketing, improving or developing services or ways in which the industry contributes to sustainable development. This was treated by the Welsh Assembly as enabling the enactment of the Red Meat Industry (Wales) Measure 2010, permitting the imposition of a levy to meet expenditure incurred on such objectives. He points out that that measure was not challenged. Equally, this means that there is no authority throwing light on its competence. The argument in favour of a generous interpretation can be further advanced, as Lord Thomas notes, by the consideration that the Welsh Assembly is undoubtedly entitled to expend monies out of the block grant on matters covered by other paragraphs of Schedule 7, such as para 5 covering education, training and the careers service, which do not make any specific reference to finance or funding. The specific reference to "funding" in para 9 may therefore suggest an intention to cover matters other than mere allocation of funds. I do not on the other hand find any assistance in the exception to para 9 relating to the regulation of prices of human medicines and medicinal products. Schedule 8 to the NHS (Wales) Act contains provisions relating to the reimbursement of any remuneration referable to the cost of drugs which is paid by any Local Health Board in any year. The exception in para 9 appears simply to make clear that the Welsh Assembly has no competence to regulate the price of such drugs. It does not to my mind carry either side's argument on the present issues.

The language of paragraph 9 of Schedule 7 addresses matters all closely linked to the internal organisation and the delivery of national health services - promoting health, preventing, treating and alleviating (or controlling) disease, illness, injury, disability or mental disorder, providing services, governance and standards of care and finally "organisation and funding of national health service". A natural inference is, I think, that "funding" was also seen as closely linked with the internal organisation and delivery of health services.

As background to an understanding of para 9, it is not, I consider, inadmissible to take note of the position regarding charging for health services as it was under the National Health Service Act 1977 in force when GOWA was passed and as it was re-enacted, in relation to Wales, by the NHS (Wales) Act 2006, passed three months after GOWA was enacted, and still in force. A fundamental tenet of the National Health Service from its outset has been that the services it provides should be free of charge, except where any relevant statutory provision expressly provides for the making and recovery of charges: section 1(2) of the National Health Service Act 1946, section 1(2) of the National Health Service Act 1977, and, now, in relation to Wales, section 1(3) of the NHS (Wales) Act 2006, described as "an Act to consolidate certain enactments relating to the health service", among which were necessarily the National Health Service Act 1977 so far as it concerned Wales. Section 1(3) provides that the services provided "must be free of charge except in so far as the making and recovery of charges is expressly provided for by or under any enactment, whenever passed". The Counsel General addressed submissions to the question whether in context this refers only to enactments by the Westminster Parliament, or whether it extends to the Welsh Assembly. I have no difficulty in accepting that it extends to the latter, but it does not itself confer competence. Competence to provide for such charges must be found elsewhere.

Within the NHS (Wales) Act itself there are provisions which do expressly confer on the Welsh Ministers power to make regulations providing for the making and recovery of charges prescribed in respect of the supply under that Act of drugs, medicines or appliances - except for a patient who is resident in hospital - or in respect of pharmaceutical services: see sections 121 and 122. These, as the Counsel General points out, are the successors to the powers to make or remit prescription charges formerly existing under sections 77, 83, 83A (as inserted by section 14(1) of the Social Security Act 1988) and 126(4) of the National Health Service Act 1977, which powers were then devolved to the Welsh Assembly by Schedule 1 to the National Assembly for Wales (Transfer of Functions) Order 1999 (SI 1999/672), under the Government of Wales Act 1998. Section 126(1) of the 1977 Act was also amended by section 6 of the National Health Service Reform and Health Care Professions Act 2002 to provide expressly that regulations made under the 1977 Act might be made by an instrument made by the Welsh Assembly. It was pursuant to the powers so devolved and conferred that the Welsh Assembly enacted its flagship reform, The National Health Service (Free Prescription and Charges for Drugs and Appliances) (Wales) Regulations 2007 (SI 2007/121 W11), which abolished prescription charges with effect from 1st April 2007. By the same token, if it were so decided, prescription charges could now be restored by regulations made by the Welsh Ministers under sections 121 and 122. But section 122 would in terms prevent their imposition in respect of a patient resident in hospital. Another provision of the same Act enables the Welsh Ministers to recover in respect of "accommodation in single rooms or small wards which is not needed by any patient on medical grounds": section 137.

On the Counsel General's case, the coming into force on 5 May 2011 of paragraph 9 of Schedule 7 of GOWA gives the Welsh Assembly competence to override or vary the scheme - which existed under the 1977 Act when GOWA was passed and was consolidated in relation to Wales three months later by the NHS (Wales) Act - by imposing charges on any basis which can be said to contribute to funding the Welsh NHS (with the sole qualification that the exception from para 9 would preclude it regulating the prices of "Human medicines and medicinal products"). The schemes of the National Health Service Acts and of GOWA are legally separate, and nothing in principle prevents the conclusion which he advocates.

Against such a conclusion, it may however be said that it gives rise to duplication of competences, with the Welsh Assembly having legislative competence in areas where the Welsh Ministers have delegated powers under the NHS (Wales) Act, and that it gives para 9 an extended scope of uncertain width, when its more obvious aim is the allocation to health boards and other health authorities or professionals of resources available to the Welsh Ministers and the Welsh National Health Service, rather than the raising of revenue. I do not consider that the essentially budgetary, accounting, auditing and macro-financial provisions of Part 5 (sections 117-145) of GOWA are by themselves a necessary answer to this point.

In these circumstances, although I see the force of the Counsel General's submission that "organisation and funding" in para 9 goes beyond allocation of resources, I prefer to approach the present appeal on an assumption, rather than deciding, that para 9 is, at least to some extent, capable of covering the raising of monies, for example by levying charges for services. But this cannot, in my opinion, mean that para 9 confers on the Welsh Assembly a general power to raise monies, even if they are to a greater or lesser extent hypothecated to the Welsh Health Service (as to which, see further para 28 below). The key question is whether, on the assumption I am making, GOWA provides legislative competence for the imposition of liabilities on compensators and insurers, and to this I therefore turn.

Section 108(7) provides that

"For the purposes of this section the question whether a provision of an Act of the Assembly relates to one or more of the subjects listed in Part 1 of Schedule 7 (or falls within any of the exceptions specified in that Part of that Schedule) is to be determined by reference to the purpose of the provision, having regard (among other things) to its effect in all the circumstances."

The expression "relates to", used in section 108(4), has been examined in the context of the Scotland Act, where it is by section 29(3) given a definition identical to that in section 108(7) of GOWA. But it is used in the Scotland Act 1998 to define not the competence conferred to the devolved Parliament, but the competence reserved to the Westminster Parliament. Despite this difference, there is no reason to give the words a different meaning in the two pieces of legislation. The expression involves words of neutral meaning, used to define the parameters of competence. In a Scottish context, it was considered by the Supreme Court in Martin v Most [2010] UKSC 10, 2010 SC (UKSC) 40, paras 15 and 49 and in Imperial Tobacco Ltd v Lord Advocate [2012] UKSC 61, 2013 SLT 2, para 16. In Martin v Most Lord Walker said that the expression was "familiar in this sort of context, indicating more than a loose or consequential connection, and the language of section 29(3), referring to a provision's purpose and effect, reinforces that" (para 49). In Imperial Tobacco, Lord Hope, in a judgment with which all other members of the court agreed, endorsed Lord Walker's approach that the expression "indicates something more than a loose or consequential connection" (para 16). In a Welsh context, the test adopted in both these authorities was referred to with approval in the recent decision in In re Agricultural Sector (Wales) Bill [2014] UKSC 43; [2014] 1 WLR 2622, para 50, where the Supreme Court added that "As the section requires the purpose of the provision to be examined it is necessary to look not merely at what can be discerned from an objective consideration of the effect of its terms".

The provision of health services and the organisation and funding of the Welsh Health Service clearly cannot permit the Welsh Assembly to raise monies generally, by relying on the fact that any monies raised from any source increase the funds available for all its spending, including spending on the Health Service. The question is whether the position is different if the monies raised can be said to be specifically intended or hypothecated to provide funds for use in the Health Service. But, if that were sufficient, it would be difficult to see any real limit to the persons on whom or basis on which such charges might be imposed, provided only that the charges were levied on that express basis. The reality is also that, unless the charges are for research, treatment or other services which would not otherwise be undertaken or provided by the National Health Service, even a hypothecated charge is in substance no different from a general charge boosting the Welsh Government's resources.

In these circumstances, any raising of charges permissible under para 9 would have, in my opinion, to be more directly connected with the service provided and its funding. The mere purpose and effect of raising money which can or will be used to cover part of the costs of the Welsh NHS could not constitute a sufficiently close connection. In the case of prescription or other charges to users of the Welsh NHS service, a direct connection with the service and its funding exists, in that users are directly involved with and benefitting by the service. In the case of charges under section 2, the argument would have to be that a sufficient connection can be found in the actual or alleged wrongdoing that led to a compensator making a compensation payment to or in respect of a sufferer from an asbestos-related disease. But that is at best an indirect, loose or consequential connection. The expression "organisation and funding of national health service" could not, in my opinion, have been conceived with a view to covering what would amount in reality to rewriting the law of tort and breach of statutory duty by imposing on third persons (the compensators), having no other direct connection in law with the NHS, liability towards the Welsh Ministers to meet costs of NHS services provided to sufferers from asbestos-related diseases towards whom such third persons decide to make a compensation payment for liability which may or may not exist or have been established or admitted.

I add that, even if (contrary to my view) hypothecation were the test of part of the test of competence, section 15(1) of the Bill does not achieve it in terms. Under section 15, the Welsh Ministers must have regard to the "desirability" of expending amounts equalling the charges levied under section 2 on research into, treatment of, or other services relating to, asbestos-related diseases. If what is desirable is achieved, then, whether or not the expenditure on such research, treatment or other services would anyway have occurred, the effect would be to cover part of the Welsh Minister's budget for NHS services. But what is desirable is not necessarily achievable or achieved. Lord Thomas suggests (para 90) that the effect of para 9 would anyway be to confine the use of any monies raised to the Health Service, even if they were not used in relation to asbestos-related diseases. But for the reasons already given, para 9 cannot in my opinion permit the Welsh Ministers to raise money in any way they choose even if the only purpose for which the monies raised can be used is on the Welsh NHS.

Even if a different view were to be taken about the existence of a sufficient connection in the case of section 2, I have no doubt that section 14 would fall outside the Welsh Assembly's legislative competence. It is argued that, assuming that section 2 falls within section 108(4) GOWA, then section 14 falls within section 108(5). That was also the basis on which the Presiding Officer made her statement of compatibility regarding section 14. But in my opinion it is not sustainable. The provisions of sections 5-13, summarised in para 3 above, could all be capable of being regarded as providing for the enforcement of, or otherwise appropriate for making effective, or incidental or consequential on, the provision contained in section 2, whereby compensators must pay the Welsh Ministers charges for NHS services provided to sufferers. But section 14 is directed to an entirely different relationship, that between compensators and their liability insurers. The only basis on which it could be argued to provide for enforcement of section 2, or be otherwise appropriate for making it effective, or be incidental or consequential on it, is financial. Without section 14, compensators required to pay under section 2 may lack the funds to do so. But section 108(5) is not, in my opinion, directed to or wide enough to cover what amounts to a separate scheme for the provision of financial recourse against third party insurers by the compensators who are primarily affected by the scheme introduced under section 108(4), as opposed to provisions enhancing the legal enforceability or, maybe, even the practical effectiveness of the scheme as against compensators. In law and practice, section 2 is part of a coherent, enforceable and effective scheme, irrespective of the financial means of compensators. And section 14 is just as incapable of being regarded as incidental or consequential to section 2. The limited role of the words "incidental to, or consequential on" is clear from Martin v Most 2010 SC (UKSC) 40. In that case, Lord Rodger at para 128 spoke of "the kinds of modifications which are obviously necessary to give effect to a piece of devolved legislation, but which raise no separate issue of principle", contrasting these with other provisions which were "independent and deal with distinct aspects of the situation". This guidance was adopted as being of assistance in the context of GOWA in Attorney General v National Assembly for Wales Commission [2012] UKSC 53, [2013] 1 AC 792, at paras 50-53 by Lord Neuberger, with the agreement of three other members of the court, while Lord Hope, with whom the same three members also agreed, distinguished at para 83 between, on the one hand, provisions which are "merely subsidiary" to other provisions and have consequences which can be seen to be "minor or unimportant in the context of the Act as a whole" and, on the other, provisions with an end and purpose of their own.

Section 14 clearly raises important issues of principle separate from sections 2-13. Unlike compensators, insurers are neither actual nor alleged wrongdoers. The rationale which exists for imposing liability for NHS charges on compensators does not apply to insurers. The rationales of imposing liability on insurers towards compensators are no doubt (i) that this favours the Welsh Ministers' prospects of making a financial recovery under section 2, and (ii) perhaps also that it lessens the blow for, and is likely to avoid objections by, compensators, or at least those who remain solvent and had arranged liability insurance. But legislation imposing on insurers new contractual liabilities under old insurance policies years after they were made engages obvious and important general principles. None of the provisions of section 108(5) could in my opinion justify section 14, and the Bill would be outside the legislative competence of the Welsh Assembly on this ground also.

Lord Thomas suggests (paras 96-98) that any doubt about competence can be resolved by reference to the consideration that, if the present legislation had imposed charges in respect of National Health Service services on National Health Service patients generally or on victims of asbestos-related diseases specifically, then neither the compensators nor their insurers could have had any complaint. The compensator would then have had to meet them, as any other loss, and they would have been recoverable from any liability insurer of the compensator subject to the terms of cover. This is a submission on which the Counsel General also relies in relation to the case under A1P1 (to which I turn later in this judgment), in which context Mr Michael Fordham QC for the interveners accepts that, if this is what had occurred, the compensators and insurers would have no case under A1P1. Their possessions would not have been disturbed, because what happened would have been within the scope of the legal obligations which they had incurred under the existing law of tort and the insurance contracts into which they had entered.

However, in the context of competence, reference to what might or might not have been done by other routes is in my view both irrelevant and detrimental for the coherent development and application of provisions of the kind contained in the devolution legislation. Either the Welsh Assembly has competence to do what it proposes, or it does not. It cannot confer competence on itself by hypothesising (however accurately) that it might legitimately have chosen a different route. The fact would remain that it had not chosen the right route. Questions of competence depend on whether what is done is permitted, not on whether something which has not been done would have been permitted. I know of no authority for a contrary proposition, which would seem to me not only novel but confusing, deleterious and likely to give rise to extensive difficulties and arguments in application. The scenario in the present case also appears an unreal one. The suggested alternative route has not been used, and it seems highly improbable that it would be attempted. The National Health Service is a prized asset throughout the United Kingdom, founded on the basic principle of free care according to needs. Imposing NHS hospitalisation charges on sufferers of asbestos-related diseases would seem even less thinkable than charging patients generally.

It was also suggested that charges might be imposed on sufferers only insofar as such sufferers were able to recover from others in respect of them. This is not in fact what the Bill proposes - it makes compensators liable in the first instance, although it aims to assist those with relevant insurance to recover under it and to do so also overrides or varies the insurance terms as far as necessary. The suggested scenario does not therefore match the Bill; it would be artificial and would highlight the reality that what were in reality being imposed were liabilities on compensators and insurers, not on victims. But in any event it is irrelevant, for the basic reason that competence must be judged by reference to what the Bill proposes, not by reference to some different scheme the competence to enact which would have to be assessed in the light of its own terms.

For all these reasons, I conclude that the Bill falls outside the legislative competence of the Welsh Assembly, in that it does not relate to any of the subjects listed in paragraph 9 of Part 1 of Schedule 7 to the Government of Wales Act 2006, and I would answer the Counsel General's reference accordingly.

Does the Bill infringe A1P1?

In the light of the conclusion I have reached in paras 27, 30 and 34, this issue does not strictly arise for decision. But it has been fully argued, and involves a disagreement about the applicable principles which has general importance. I will therefore express my views on it. For this purpose, it is necessary to assume, contrary to my conclusion in para 34, that the Bill falls within section 108(4) and/or (5) of GOWA. The question is whether, on that basis, it is compatible with the Convention rights scheduled to the Human Rights Act 1998 as required by section 108(6)(c). The relevant right allegedly infringed is article 1 of Protocol No 1 ("A1P1"). This reads:

"Every natural or legal person is entitled to the peaceful enjoyment of his possessions. No one shall be deprived of his possessions except in the public interest and subject to the conditions provided for by law and by the general principles of international law.

The preceding provisions shall not, however, in any way impair the right of a State to enforce such laws as it deems necessary to control the use of property in accordance with the general interest or to secure the payment of taxes or other contributions or penalties."

The relevant NHS costs and compensation payments will be incurred only in the future, once the Bill is in force. But the liability or alleged liability which under section 2 triggers the Welsh Minister's right to recover in respect of them arises from exposure to asbestos which occurred decades ago. The effect of the Bill is therefore to impose on compensators, in the first instance, and their insurers, in the second instance, burdens which have not previously existed. The interveners submit that the Bill would thus deprive both employers and their insurers of their previous legal freedom from exposure to the relevant charges and of their possessions in the form of the assets they would have to use to discharge the new liabilities imposed by the Bill.

The Counsel General for Wales submits, in response, that it is "not free from doubt" whether A1P1 is engaged in these circumstances. Focusing only on the insurance position, his written case argues that "a contract of insurance … operates at the individual level, not at the level of the balance sheet of the insurer". The essence of insurance is however the pooling of risks and premia. The bottom line of an insurer's balance sheet depends upon the rating and writing of individual contracts, which in their totality make up its underwriting book. All individual contracts are a piece of the whole, a part of the main. Any additional liability imposed on a category of policy will feed through into the balance sheet. The "complex inter-relationship" between payments out and past, current and future premium receipts, and (since 1969) compulsory employers' insurance for broadly defined liabilities, to which the Counsel General also refers, cannot obscure this simple truth.

The Counsel General points out, correctly, that insurers could have had no complaint if the sufferer had decided to use and had the means or insurance to cover hospitalisation in a private hospital. The sufferer could then have held the compensator liable and the compensator could in turn have looked to any insurer he had. That is true, but the liability would have arisen by a conventional route, and the likelihood or unlikelihood of its arising is something which compensators and their liability insurers could assess and factor into their accounts and plans. In reality, the likelihood of liability arising by this route must always have been small.

The Counsel General also points out, correctly, that neither the compensators nor their insurers could have had any complaint if the present legislation had imposed the charges on the sufferer. The compensator would then have had to meet them, as any other loss, and they would have been recoverable from any liability insurer of the compensator subject to the terms of cover. In such circumstances, Mr Michael Fordham QC for the interveners accepts that the compensators and insurers would have no case under A1P1. Their possessions would not have been disturbed, because what happened would have been within the scope of the legal obligations which they had incurred under the existing law of tort and the insurance contracts into which they had entered. However, for reasons already noted in paras 32 and 33 above, this scenario is also an unreal one. It has not, and would never have, occurred. The further suggestion that charges might be imposed on sufferers only insofar as such sufferers were able to recover from others in respect of them seems equally remote.

If any of these remote scenarios was to be treated as conceivable, it would fall within the exposure accepted by those causing victims to suffer asbestos-related diseases and the risks accepted by their liability insurers. But it does not mean that either employers or employers' liability insurers are taken to accept other, yet further risks, deriving from the positive intervention of the legislature, cutting across the ordinary law of tort and the agreed policy terms. The present case must again be judged by what the legislature has actually chosen to do – no doubt because it concluded that this was necessary - rather than by reference to remote contingencies, the non-adoption of which by the legislature tends to confirm their unreality.

In my opinion, and in agreement on this point with Lord Thomas (paras 103-104), A1P1 is engaged as regards both compensators and their liability insurers. Both are affected and potentially deprived of their possessions, in that the Bill alters their otherwise existing legal liabilities and imposes on them potentially increased financial burdens arising from events long-past and policies made long ago. "A person's financial resources … are capable of being possessions within the meaning of A1P1", as Lord Hope of Craighead put it in AXA General Insurance Ltd v HM Advocate [2011] UKSC 46, [2012] 1 AC 868, para 25; the question is whether the alleged victim is "a member of a class of people who risk being directly affected by the legislation", rather than subject to some purely hypothetical risk: paras 25-26, with reference to Burden v United Kingdom (2008) 47 EHRR 38, para 34. Lord Hope's judgment on these points carried the support of all members of the House: paras 73, 85-90, 109-114 and 177, with Lord Reed noting at para 111 that the Convention was intended to guarantee rights that were "practical and effective" and that the Convention concept of a "victim" was "correspondingly broad".

In AXA, the Scottish Parliament had by the Damages (Asbestos-related Conditions) (Scotland) Act 2009 Act reversed the House of Lords' decision in in Rothwell v Chemical & Insulating Co Ltd [2007] UKHL 39, [2008] AC 281 that pleural plaques did not constitute damage for the purposes of a claim for breach of tortious or statutory duty. The effect was to make employers liable for loss not previously recoverable occurring as a result of long-past breaches of duty. Employers' insurers challenged the statute because of the additional burden which could thus fall on them. It was objected that they were not victims for the purposes of the Convention rights. The objection failed: paras 23-28 per Lord Hope, para 73 per Lord Brown of Eaton-under-Heywood, paras 85-90 per Lord Mance, paras 109-114 per Lord Reed and para 177 per all three other members of the Court agreeing with Lord Hope and Lord Reed. Lord Brown regarded the answer to the objection as "clear almost beyond argument" (para 73). Lord Reed and I pointed out that the logical consequence of the argument (had it been accepted) would have been that the true or only persons with victim status were employers: paras 110 and 190.

The position under the present Bill is a fortiori to that which existed in AXA. The Bill is clearly directed at both compensators and insurers, but it is also expressly directed at insurers as well as compensators. Moreover it imposes liabilities on both not only in conjunction with existing liabilities, but in addition to them. It does so in the case of compensators by making it irrelevant whether the compensation reflects any actual or admitted liability. It does so in the case of insurers by making them liable in circumstances where the insurance cover which they granted would not apply. For all these reasons, both compensators and insurers are in my opinion entitled to be regarded as victims for the purposes of A1P1.

General principles under A1P1

The European Court of Human Rights has examined the application of A1P1 in a number of cases. These are all cases at an international level, in which the margin of appreciation had therefore an important potential role. We are concerned with the domestic application of the Convention. The margin of appreciation does not apply. Instead, the issue is with what intensity we should review the Bill and what deference is due or weight attaches to the legislature's view as to the appropriateness of the Bill: see per Lord Reed in AXA [2012] 1 AC 868, 131, R (Huitson) v Revenue and Customs Comrs [2011] EWCA Civ 893, [2012] QB 489, 85.

The general principles according to which a court will review legislation for compliance with the Convention rights scheduled to the Human Rights Act 1998 have been comprehensively reviewed in recent case law, particularly Bank Mellat v Her Majesty's Treasury (No 2) [2013] UKSC 39, [2014] AC 700, paras 68-76 per Lord Reed, with whose observations in these paragraphs Lord Sumption, Lady Hale, Lord Kerr and Lord Clarke agreed at para 20 and Lord Neuberger agreed at para 166, and R (Nicklinson) v Ministry of Justice (CNK Alliance Ltd intervening) [2014] UKSC 38, [2014] 3 WLR 200. There are four stages, which I can summarise as involving consideration of (i) whether there is a legitimate aim which could justify a restriction of the relevant protected right, (ii) whether the measure adopted is rationally connected to that aim, (iii) whether the aim could have been achieved by a less intrusive measure and (iv) whether, on a fair balance, the benefits of achieving the aim by the measure outweigh the disbenefits resulting from the restriction of the relevant protected right. The European Court of Human Rights has however indicated that these stages apply in relation to A1P1 with modifications which have themselves been varied over the years.

Initially, in Handyside v United Kingdom (1976) 1 EHRR 737, para 62, followed in Marckx v Belgium (1979) 2 EHRR 330, para 63, the court said that the State was the sole judge of necessity for the purposes of deciding whether a deprivation of property was "in the public interest". That no longer represents the position on any view. But the Counsel General for Wales and Mr Michael Fordham QC disagree as to the current position. The Counsel General submits that the court will at each of the four stages of the analysis "respect the legislature's judgment as to what is 'in the public interest' unless that judgment be manifestly without reasonable foundation": James v United Kingdom (1986) 8 EHRR 123, para 46. Mr Michael Fordham QC on the other hand submits that this passage was or, at least in subsequent authority, has been restricted in application to the first or at all events the first to third stages. In my opinion, Mr Michael Fordham QC is basically correct on this issue, at least as regards the fourth stage which presently matters, although that does not mean that significant weight may not or should not be given to the particular legislative choice even at the fourth stage.

In James itself, the court went on in paras 47-49 to address the question whether the aim of the legislation was a legitimate one in principle concluding that the United Kingdom Parliament's belief in the existence of a social injustice "was not such as could be characterised as manifestly unreasonable". But, turning in para 50 to the "means chosen to achieve the aim", it then said:

"This, however, does not settle the issue. Not only must a measure depriving a person of his property pursue, on the facts as well as in principle, a legitimate aim 'in the public interest', but there must also be a reasonable relationship of proportionality between the means employed and the aim sought to be realised (see, amongst others, and mutatis mutandis, the above-mentioned Ashingdane judgment (1985) 7 EHRR 528, 57). This latter requirement was expressed in other terms in the Sporrong and Lönnroth judgment by the notion of the 'fair balance' that must be struck between the demands of the general interest of the community and the requirements of the protection of the individual's fundamental rights ((1983) 5 EHRR 35, para 69). The requisite balance will not be found if the person concerned has had to bear 'an individual and excessive burden' (para 73). Although the court was speaking in that judgment in the context of the general rule of peaceful enjoyment of property enunciated in the first sentence of the first paragraph, it pointed out that 'the search for this balance is ... reflected in the structure of article 1 (P1-1)' as a whole (para 69)."

Later authority confirms the principle governing the validity of the "means chosen to achieve the aim" is one of "fair balance". The court has developed the distinction introduced in James. The court will accept the legislature's judgment as to what is "in the public interest" unless that judgment is "manifestly without reasonable foundation". But "an interference with peaceful enjoyment of possession must nevertheless strike a 'fair balance' between the demands of the public or general interest of the community and the requirements of the protection of the individual's fundamental rights" : see eg AGOSI v United Kingdom (1987) 9 EHRR 1, at paras 48 and 52, Gasus Dosier- und Fördertechnik v Netherlands (1995) 20 EHRR 403, at para 62, Pressos Cia Naviera SA v Belgium (1996) 21 EHRR 301, at para 35 (covering "in the public interest", with a footnote reference to James) and paras 36-44 (covering "proportionality of the interference"), Bäck v Finland (2005) 40 EHRR 48, at paras 53 and 55, Grainger v United Kingdom (Application No 34940/10) (unreported) 10 July 2012, at paras 35 and 36 and, most recently, Paulet v United Kingdom The Times, 19 May 2014; [2014] ECHR 477, at para 63 (citing AGOSI).

Pressos and Bäck are of particular interest in the present reference as cases of retrospective interference. In Pressos legislation removed retrospectively the tortious right to compensation which shipowners had, on the basis of longstanding Belgian Supreme Court authority, enjoyed. The Belgian government invoked the "enormous" financial implications of such liability (para 40), but the court said:

"43. The financial considerations cited by the Government and their concern to bring Belgian law into line with the law of neighbouring countries could warrant prospective legislation in this area to derogate from the general law of tort.

Such considerations could not justify legislating with retrospective effect with the aim and consequence of depriving the applicants of their claims for compensation.

Such a fundamental interference with the applicants' rights is inconsistent with preserving a fair balance between the interests at stake."

In Bäck retrospective legislation had granted relief to impecunious debtors allowing them to write down their debts very substantially on the basis of a greatly reduced payment schedule. The retrospective nature of this legislation meant "that a special justification [was] required for such interference" with existing contracts. It was however "remedial social legislation" and "in particular in the field of debt adjustment, it must be open to the legislature to take measures affecting the further execution of previously concluded contracts in order to attain the aim of the policy adopted" (para 68). The "striking" amount of the reduction was justified by the consideration that the debt was already worth "much less than its nominal value" and any claim to recover it "had already been rendered highly precarious before the debt adjustment for reasons not attributable to the State" (paras 69-70).

Domestic law is to like effect. Lord Hope in AXA General Insurance Ltd v HM Advocate [2011] UKSC 46, [2012] 1 AC 868 addressed separately the issues of "Legitimate aim" (paras 29-33) and "Proportionality" (paras 34-41). Only in relation to the former did he identify the relevant test as being whether the legislature's choice as to what was 'in the public interest' was "manifestly unreasonable", citing in this connection James, 8 EHRR 123, at para 46. In relation to proportionality, he applied the fair balance test, citing Sporrong and Lönnroth and Pressos. Lord Reed's judgment contains the same distinction in paras 124-125 and 126-128. Save for Lord Brown, all the other four members of the court including myself were content to agree with Lord Hope's and Lord Reed's judgments on this aspect. However, Lord Brown at paras 80 and 83 took a different, rolled-up approach to the issues of legitimate aim and proportionality. His approach would, if adopted, support the Counsel General's approach that any challenge on either score must, to succeed, show that the measure was "manifestly without reasonable justification". If Lord Brown's judgment is read in this way, he was in a minority on the point and his view on it does not in my opinion represent the law.

I conclude that there is Strasbourg authority testing the aim and the public interest by asking whether it was manifestly unreasonable, but the approach in Strasbourg to at least the fourth stage involves asking simply whether, weighing all relevant factors, the measure adopted achieves a fair or proportionate balance between the public interest being promoted and the other interests involved. The court will in this context weigh the benefits of the measure in terms of the aim being promoted against the disbenefits to other interests. Significant respect may be due to the legislature's decision, as one aspect of the margin of appreciation, but the hurdle to intervention will not be expressed at the high level of "manifest unreasonableness". In this connection, it is important that, at the fourth stage of the Convention analysis, all relevant interests fall to be weighed and balanced. That means not merely public, but also all relevant private interests. The court may be especially well placed itself to evaluate the latter interests, which may not always have been fully or appropriately taken into account by the primary decision-maker.

It is also clear that The European Court of Human Rights scrutinises with particular circumspection legislation which confiscates property without compensation or operates retrospectively. In the case of confiscation, it will normally be disproportionate not to afford reasonable compensation, and a total lack of compensation will only be justifiable in "exceptional circumstances". In the case of retrospective legislation, "special justification" will be required before the court will accept that a fair balance has been struck: paras 48-49 above. The Counsel General in his written case (paras 89 and 126) himself states that "It is of course accepted, as the case law … makes clear, that there is a need for special justification where a statutory provision has retrospective effect", while maintaining that this is present in the circumstances of this case.

At the domestic level, the margin of appreciation is not applicable, and the domestic court is not under the same disadvantages of physical and cultural distance as an international court. The fact that a measure is within a national legislature's margin of appreciation is not conclusive of proportionality when a national court is examining a measure at the national level: In re G (Adoption: Unmarried Couple) [2008] UKHL 38, [2009] AC 173; R (Nicklinson) v Ministry of Justice [2014] 3 WLR 200, at paras 71, 163 and 230, per Lord Neuberger, Lord Mance and Lord Sumption. However, domestic courts cannot act as primary decision makers, and principles of institutional competence and respect indicate that they must attach appropriate weight to informed legislative choices at each stage in the Convention analysis: see AXA, para 131, per Lord Reed, R (Huitson) v Revenue and Customs Comrs [2011] EWCA Civ 893, [2012] QB 489, at para 85. But again, and in particular at the fourth stage, when all relevant interests fall to be evaluated, the domestic court may have an especially significant role.

To put a legislative measure in context, domestic courts may (under a rule quite distinct from that in Pepper v Hart [1993] AC 593) examine background material, including a white paper, explanatory departmental notes, ministerial statements and statements by members of parliament in debate: Wilson v First County Trust Ltd (No 2) [2003] UKHL 40, [2004] 1 AC 816. But care must at the same time be taken not to question the "sufficiency" of debate in the United Kingdom Parliament, in a way which would contravene article 9 of the Bill of Rights. In Wilson, at para 67, Lord Nicholls of Birkenhead put this point as follows (para 67):

"Lack of cogent justification in the course of parliamentary debate is not a matter which 'counts against' the legislation on issue of proportionality. The court is called upon to evaluate the proportionality of the legislation, not the adequacy of the minister's exploration of the policy options or of his explanations to Parliament. The latter would contravene article 9 of the Bill of Rights. The court would then be presuming to evaluate the sufficiency of the legislative process leading up to the enactment of the statute."

There is in this connection a potential tension. If, at the fourth stage when the court is considering whether a measure strikes a fair balance, weight attaches to the legislative choice, then the extent to which the legislature has as the primary decision maker been in or put in a position to evaluate the various interests may affect the weight attaching to its assessment: see Belfast City Council v Miss Behavin' Ltd [2007] UKHL 19, [2007] 1 WLR 1420, at paras 27, 37 and 46-47, per Lord Rodger, Lady Hale and Lord Mance. That was a case involving subordinate legislation, to which article 9 of the Bill of Rights does not apply. Perhaps in the light of article 9 there is a relevant distinction between cases concerning primary legislation by the United Kingdom Parliament and other legislative and executive decisions. It is, I think, unnecessary to go further into this difficult area on this reference. On any view, if the admissible background material shows that the Bill was put before and passed by the Welsh Assembly on the basis of a supposed analogy or precedent, it must be possible to consider whether that analogy or precedent actually applies, and, if it does not, the same assistance cannot be obtained from the legislative choice as might otherwise be the case.

Application of A1P1 to the present reference

I have already concluded that the Bill engages A1P1, and addressed the Counsel General's argument that there are other means by which compensators and insurers might have become or been made liable to bear hospitalisation costs, without altering the laws of tort and contract in the way undertaken by the Bill. More generally, the Counsel General also submits that insurers (as well no doubt as compensators) run a considerable risk of unforeseen exposure, and that this is particularly so in relation to asbestos-related diseases, as recent decades have shown. Accepting that as correct, it is, however, no justification for the retrospective imposition of further exposure, which they could legitimately expect could not and would not fall upon them. They could legitimately expect this not only when issuing their original policies, but also when considering their reserves for incurred but as yet unreported claims, as any long-tail insurer must do regularly for accounting and solvency purposes and must no doubt also do when considering what, if any, reinsurance or further reinsurance it should from time to time purchase.

I note in parenthesis, because no such points were developed before us and I do not therefore rely on them, that it is unclear what insurance policies could or would be caught by the Bill. The Bill is limited to Welsh NHS services, but it purports to apply to all insurance contracts issued to compensators. The proper law of such contracts might be English or Scottish or even foreign, and any indemnity might be due for performance outside, rather than in, Wales. It is not clear to me how Welsh legislation could affect a Scottish or foreign policy, and it might be arguable whether it could affect an English policy due for performance in (say) London. Another point on which the Bill is silent is reinsurance. Having imposed on insurers uncovenanted liabilities, the Bill leaves insurers to make whatever recovery they can under any reinsurances which may be in wide enough terms, without alteration, to cover such new liabilities.

The Counsel General relies on the Supreme Court's reasoning as well as the decision in the AXA case. The Counsel General and Mr Michael Fordham QC differ in their analysis of this reasoning. The Counsel General relies upon Lord Hope's identification in paras 37-38 of a special feature of that case as being "that the business in which insurers are engaged and in pursuance of which they wrote the policies that will give rise to the obligation to indemnify is a commercial venture which is inextricably associated with risk" (para 38). Lord Hope went on to point out that phrases such as "bodily injury or disease" might expand as medical knowledge and circumstances changed, that new diseases might become familiar, as occurred with asbestos-related diseases, and that the number, nature and value of claims were always liable to develop in ways that were unpredictable. Lord Hope was addressing the expansion of insurance liabilities by conventional routes, including the relaxed approach to causation taken in cases such as Fairchild v Glenhaven Funeral Services Ltd [2002] UKHL 23, [2003] 1 AC 32 and the Trigger litigation, and using that as a stepping stone for consideration of the issue before the Supreme Court in AXA which was whether a legislative reversal of the prior House of Lords decision in Rothwell v Chemical & Insulating Co Ltd [2007] UKHL 39, [2008] AC 281 could be similarly categorised.

Lord Hope's words commanded the agreement of a majority of other members of the court, but I do not accept the Counsel General's submission that this means that there was disagreement between him and the judgments of either Lord Brown or myself in this area. Lord Hope was careful to make clear in para 40 that the case was not one where the law was settled – the Scottish Parliament was restoring a position which might well have represented the law. He also stressed at para 37 that the liability imposed by the Act depended on establishing negligence and preserved all other defences, other than the single question whether pleural plaques are as such actionable. Consistently with this, Lord Brown at para 83 made clear that the case turned on the absence of any legitimate expectation as to the irrecoverability of damages for pleural plaques, rather than on the fact that "the appellants as insurers are in a business inevitably associated with risks and unpredictable events". I expanded this point at para 91, when I said:

"Retrospectivity. The key to this issue is not in my view that insurance is a contract against risks. There are always limits to the contingencies upon which insurers speculate, provided by the terms and conditions of the policy. Further, insurers are normally entitled to expect that the liabilities, which their insured employers incur 'arising out of and in the course of [their] employment' and which they insured under the specimen copy policy to which I have referred, will be liabilities capable of existing in law at the time of the occurrence during the relevant employment from which such liabilities arise. Hence, the present challenge to the 2009 Act is based on the fact that it retrospectively converts into harm actionable in law physical changes which (it has been held in Rothwell …) were not otherwise such, in the hope or expectation that the relevant policies will respond to that development."

As the outcome of AXA itself shows, the mere fact that legislation changes the pre-existing law retrospectively does not make it incompatible with A1P1. Lord Brown was in AXA (para 78) exercised in this connection by a possible distinction between the power of the courts to "adapt and develop (ie change) the law (albeit within well-recognised constraining limits) to accord with what the judges consider to be the contemporary demands of justice" and the position of the legislature. But the answer to this concern appears to me to lie at least generally in Lord Brown's own words "adapt and develop" and "well-recognised limits". The common law moves, so far as possible, incrementally and, when some greater shift takes place, it can be expected to be based on some general social consensus that it is appropriate. Common law courts have themselves accepted the possibility of prospective overruling, with express reference to its potential utility in a Convention context: In re Spectrum [2005] UKHL 41, [2005] 2 AC 680 and A v HM Treasury (JUSTICE intervening) [2010] UKSC 5, [2010] 2 AC 534, 693-694, at para 17 per Lord Hope. As this implies, common law jurisprudence must itself take account of the principle that special justification is required for retrospective changes upsetting legitimate expectations.

That failure to do this may contravene Convention rights has recently been underlined by the European Court of Human Rights's decision in del Río Prado v Spain (Application No 42750/09) (unreported) given on 21 October 2013. In that case, the periods to be served under various prison sentences had, in accordance with previous case law stemming from a decision of the Spanish Supreme Court dating in 1994, been ordered in 2000 to be combined and capped at 30 years. However, in 2006 the Spanish Supreme Court in its Parot judgment (STS 197/2006) departed from this previous case law, holding that the sentences should be viewed individually, with the result that the applicants' release date was refixed by the Audiencia Nacional in 2008 to expire at a date some nine years later than it would have done. The Strasbourg Court adopted a test of foreseeability (para 130), holding that at the time when the applicant was convicted, detained and notified of the decision to combine the sentences and set a maximum term of imprisonment, she could not have foreseen to a reasonable degree that the method used to apply remissions of sentence for work done in detention would change as a result of a departure from case-law by the Supreme Court in 2006 and that the new approach would be applied to her. Her detention after the expiry of the combined period of 30 years was accordingly unlawful under article 5(1) of the Convention, and Spain was ordered both to compensate her and to ensure her release. Whether the issue of retrospectivity arises in a statutory or common law context, there are therefore potential constraints which reflect the legitimate expectations of those affected.

The Counsel General submits that AXA was a stronger case for treating the legislation as incompatible than the present, yet the Supreme Court did not do so. I do not accept the Counsel General's analysis. The Scottish statute in issue in AXA affected all outstanding and future claims, and the present Bill on its face also affects all future compensation payments made in respect of outstanding and future claims. But the two differ in other important respects:

a. The Scottish statute was passed to rectify a perceived injustice directly affecting those suffering from asbestos-related diseases, and was in this very real sense social remedial legislation. Despite the Counsel General's contrary submission, the same cannot in my opinion be said of the Bill. It has no effect on sufferers from asbestos-related diseases. Its purpose is to transfer the financial burden of costs of their hospitalisation from the Welsh Ministers to compensators and their insurers.

b. The Scottish statute was passed to restore the legal position as it had been understood at first instance for some decades, and it might well have been accepted as being at the highest instance. The present Bill aims to change a well-understood position which has existed since the NHS was created, by introducing a new right of recourse which has never previously existed, though it is one which Parliament could at or at any time since the creation of the NHS have decided to introduce without any legal problem in relation to future events giving rise to liability claims against compensators (and so to liability insurance claims by compensators against their liability insurers).

c. The Scottish statute built on established legal principles, requiring liability to exist before compensators could be compelled to meet claims for pleural plaques and for insurance cover to exist before such compensators could recover from their liability insurers. This was one of the two points stressed by Lord Hope in AXA, as I have mentioned in the preceding paragraph. The Bill bypasses such principles, making the liability of compensators dependent simply on the payment of compensation, even if made without admission of liability and making the liability of insurers arise independently of the terms of the insurance policies issued, by reference to the fact of payment of such compensation, provided such policies would to some extent cover any liability which such compensators would, if it were established, have had.

The first of these points requires further treatment. The Counsel General submits that, although the Bill has no effect on sufferers from asbestos-related diseases, it is a measure passed as a matter of economic and social policy, in relation to which the Welsh Assembly should be recognised as having a wide area of appreciation and discretionary judgment: see Huitson [2012] QB 489, at para 85 per Mummery LJ. He also cites in support the House's decision in Wilson [2004] 1 AC 816. Both these were cases where the relevant legislation had retrospective aspects. But in both there were directly applicable and compelling social interests militating in favour of retrospectivity. Wilson concerned consumer protection legislation regarding the enforceability of loan agreements which failed correctly to state the amount of credit. Huitson concerned legislation protecting a grave challenge to the public exchequer, posed by wholly artificial tax arrangements taking advantage of double taxation treaties to avoid the payment of United Kingdom tax by United Kingdom residents. The arrangements were anyway doubtfully legal and such residents had no legitimate expectation that they could avoid such tax.

Although the Bill would either save the Welsh Ministers money or add to their resources, it is not shown that it would achieve a directly applicable or compelling social or economic interest comparable with those involved in these previous cases. Section 15 of the Bill contains the specific enjoinder that the Ministers should have regard to the "desirability" of equivalent sums being made available for "research into, treatment of or other services relating to asbestos-related diseases", but it is not shown that any such sums so expended would add to existing sums already being spent in these areas, or resolve any exceptional social or economic problem. It is common knowledge that the funding of the National Health Service is under increasing strain throughout the United Kingdom, and it may be so even more in Wales than elsewhere, but that is a different level of general problem to any shown on the authorities to be relevant in the present context.

The Counsel General maintains that special justification exists for the retrospectivity involved in the Bill because, without it, the Bill cannot achieve its legitimate policy aim. That is a circular submission, which, if accepted, would eliminate the important balancing stage of the proportionality exercise identified by Lord Reed in Bank Mellat (para 43 above) by Lord Hope in AXA (para 49 above) and by the Strasbourg Court in its case law (paras 44-48 above). As a matter of legislative policy it could be thought appropriate by the relevant legislature that the Welsh NHS should be able to recover hospitalisation costs from those whose breach of tortious or statutory duty caused them to be incurred. But that is, as I have noted, a provision which could have been made by the United Kingdom when or at any time since the NHS was introduced. It is a provision which would no doubt have been proportionate if introduced in relation to future exposure to asbestos and future insurance contracts. But rewriting historically incurred obligations to impose it in relation to future Welsh NHS costs is a quite different step. It is a step for which, on the authorities and as the Counsel General accepts, special justification is necessary, and none is shown. I therefore conclude that, even assuming the Bill to satisfy section 108(4) and/or (5), it falls outside the legislative competence of the Welsh Assembly.

Lord Thomas attaches great weight to the judgment of the Welsh Assembly that this is a measure which should in the interests of Wales be enacted. I agree that weight should be given to the Welsh Assembly's judgment. But it is the court's function, under GOWA, to evaluate the relevant considerations and to form its own judgment, on the issue both of legislative competence and of consistency with the Convention rights. I would arrive at the conclusions I have, even if the background to the Bill had consisted of a full presentation and appreciation of its implications by those responsible for promoting and passing the legislation. My conclusion is merely reinforced by the consideration that this does not appear to have been the case. Rather, the Bill was seen as a mere extension in degree of a United Kingdom measure which had already been accepted in principle by the United Kingdom Parliament despite its retrospectivity. The measure in question is the Health and Social Care (Community Health and Standards) Act 2003. This applies to enable the recovery from compensators of costs of hospitalisation incurred by the National Health Service "in consequence of any injury, whether physical or psychological": section 150(1) and (2). "Injury" is specifically defined as not including any disease: section 150(5). The exclusion of disease was in the light of strong representations about the retrospective implications of covering disease, and a lead-time of (in the event) three years was allowed before the Act came into force in relation to injury, following representations that a lead-time of two or more years was required to allow insurers to re-rate policies to cover the relative short-tail exposure arising from injury. In short, the 2003 Act shows the United Kingdom Parliament concerned not to legislate in a manner which was to any significant extent retrospective.

The 2003 Act was explained by the Health and Social Care Committee which reported on the Bill for the Welsh Assembly in March 2013 as not differing "in principle" on the question of retrospectivity, though it was said that "due to the lengthy latency period for asbestos-related diseases, compared with the immediacy of accidental injuries, there may well be a difference in scale between the functions of the two pieces of legislation; that is the degree of retrospectivity will be greater in the Bill than the 2003 Act" (para 98). Nevertheless, the Committee went on to add that it was "content that the Bill will not apply to compensation payments that have already made [sic] and that it is inevitable that insurance claims arise for matters and amounts that could not be fully foreseen when the original policies were taken out. We believe that is the nature of the insurance business." (para 99).

The Committee's assessment of the Bill as no different in principle, but only different in degree, from the 2003 Act does not reflect the very real and substantial difference in both aim and effect of the two measures. The Committee's final comment in para 99 would, if carried to a logical conclusion, justify any retrospective re-writing of any insurance contract, and, for the reasons which I gave in AXA, is not a justification for imposing on compensators and insurers unforeseen and unforeseeable new obligations which they had no opportunity to assess, rate or make reserves to cover.

Conclusion

It follows from the above that I regard the Bill as outside the legislative competence of the Welsh Assembly under both section 108(4) and section 108(5) GOWA, and, had I reached a contrary conclusion on that, as outside its legislative competence under section 108(6)(c). I would answer the Counsel General's reference to that effect.

LORD THOMAS: (with whom Lady Hale agrees)

Introduction

I agree with the result set out in the judgment of Lord Mance on the referred question, namely whether the National Assembly for Wales (the Welsh Assembly) had legislative competence to impose the liabilities set out in the Bill on insurers under section 14 of the Bill. However, as my reasons for reaching that conclusion are much narrower and as I have reached a different conclusion on other issues, I will set out my own views.

The original challenge to the legislative competence of the Welsh Assembly was the contention by the Association of British Insurers that section 14 of the Bill was incompatible with the Convention rights of insurers under article 1 of Protocol 1 (A1P1) and therefore infringed section 108(6)(c) of the Government of Wales Act 2006 (GOWA 2006). The Association of British Insurers subsequently raised in their written case the further issue as to whether the legislative competence conferred on the Welsh Assembly under section 108(4) and (5) to pass primary legislation included competence to impose the liabilities set out in the Bill on insurers and others.

It became apparent as the argument developed that, although the question referred by the Counsel General was limited to the legislative competence of the Welsh Assembly to enact section 14 of the Bill which related only to insurers, the issues also necessarily encompassed the position of those within section 2 whose alleged negligence or breach of statutory duty in the past had caused asbestos-related diseases. As those within section 2 will in the overwhelming number of cases be the employers of those who are suffering from asbestos-related diseases, it is convenient to refer to those within section 2 as "employers". It is important to note that it is by no means clear that any employer or any other person encompassed within section 2 objected to the provisions of the Bill which imposed liability on them. Certainly no argument was advanced before the court by anyone instructed on behalf of any such person. The argument was solely advanced by the Association of British Insurers to protect their own interests.

The legislative background

It has been clear since at least the late 1970s that the majority of persons suffering from asbestos-related diseases are employees of industrial enterprises who contracted the disease whilst in such employment. If the negligence or breach of statutory duty of their employer caused the injury giving rise to the disease, the employer will be liable for damages as a tortfeasor. Those damages will include medical expenses incurred by the employee if, for example, the employee has incurred them by seeking private treatment or required a level of care not provided under the National Health Service. The employer, if insured under the usual form of employers' liability policy, will be entitled to recover an indemnity for such damages under the policy, subject to the terms of the policy and any permitted limits or deductibles.

However, as the National Health Service in the United Kingdom (NHS) has, since its establishment under the National Health Service Act 1946, provided care on the basis of the service being free of charge at the point of delivery, the cost of medical treatment and of long term care has for the overwhelming majority of those suffering from asbestos-related diseases been met from the financial allocation made by the State to the NHS. That cost has therefore been a charge to the general revenue of the State rather than being met by the tortfeasor, namely the employer whose negligence brought about the disease, and by the insurers of that employer. It is in reality a state benefit provided by the State to such employers and their insurers which relieves them of some of the consequences of the employers' wrongdoing as a tortfeasor.

In 2006, separate legislative provision was made by the United Kingdom Parliament for the National Health Service in Wales by the National Health Service (Wales) Act 2006 (the NHS (Wales) Act), a consolidating Act which replaced the National Health Service Act 1977 and set out a framework for the National Health Service in Wales (the Welsh NHS). The Act was enacted by the United Kingdom Parliament at a time when the Welsh Assembly did not have legislative competence to pass primary legislation.

In 2013, about two years after the provisions of Part 4 of the GOWA 2006 came into effect, conferring on the Welsh Assembly competence to enact primary legislation in defined areas, the Bill referred was enacted as primary legislation by the Welsh Assembly. On my analysis of the provisions of the Bill, it should be seen as having two distinct aims.

(i) The first and central aim of the Bill is to withdraw the requirement that the Welsh NHS continue the delivery of the benefit to employers and their insurers of not having to meet the cost of medical treatment and care of an employee where the employers are responsible for causing asbestos diseases as tortfeasors. It is intended that the costs of medical treatment and long term care of such employees incurred by the Welsh NHS after the coming into force of the Bill are to be met by employers responsible at any time in the past for causing asbestos-related diseases and by the employers' insurers, rather than being met out of the monies generally provided by the Welsh Government to the Welsh NHS out of the block grant allocated by Her Majesty's Treasury to the Welsh Assembly.

(ii) The second, but necessarily subsidiary, aim is to establish machinery for collection of the costs which is as simple and as efficient as possible and causes those with asbestos-related diseases the least stress. It is intended that the machinery would enable employers to recover under their employers' liability policy the sums payable by way of charges to the Welsh NHS which would have been payable if the liability for such charges had been imposed on the employees and recovered in the conventional way as damages from the employers.

It is against that short summary of the background that I turn to consider the issues of legislative competence under section 108 (4) and (5) and in respect of A1P1. It is important to underline two points at the outset.

(i) The basis of the view I have formed is that the Bill has the two distinct objectives which I have set out and which it is necessary to analyse separately.

(ii) Secondly, it is necessary in such an analysis first to consider the liability of the employer. That is because the effect of the Bill on the liability of insurers under their employers' liability policies depends on an examination of the two distinct objectives of the Bill as they affect any employer who has the benefit of employers' liability insurance.

My approach was not the central focus of the argument, particularly because the only challenge was from the insurance industry and not from any of the employers. However because the procedure to refer a question to this court operates as a direct reference resulting in a final decision without the benefit of a prior decision of another court and because the effect of the judgment of Lord Mance, as the view of the majority of the court, is far reaching and final, it is necessary to set out my own analysis.

The position of employers under section 2 of the Bill

(1) Legislative competence under section 108(4) and (5)

(a) The legislative competence to fund the Health Service under section 108(4) and (5) and Schedule 7

The legislative competence of the Welsh Assembly to enact primary legislation extends to legislating afresh by a new Act of the Welsh Assembly or by amending by means of a new Act of the Welsh Assembly a statute previously enacted by the United Kingdom Parliament. Its competence to do so, apart from compliance with the Convention on Human Rights, is set out in section 108(4) and (5) of the GOWA 2006 and the 20 headings enumerating specific competence set out in Part 1 of Schedule 7. These provisions which operate on a conferred powers model were recently considered and explained in In re Agricultural Sector (Wales) Bill [2014] UKSC 43, [2014] 1 WLR 2622.

The relevant heading in Part 1 of Schedule 7 is Heading 9: "Health and health services":

"Promotion of health. Prevention, treatment and alleviation of disease, illness, injury, disability and mental disorder. Control of disease. Family planning. Provision of health services, including medical, dental, ophthalmic, pharmaceutical and ancillary services and facilities. Clinical governance and standards of health care. Organisation and funding of national health service."

Although none of the exceptions listed under this heading is relevant, it is clear from the whole of Part 1 of Schedule 7 and the exceptions under other headings that no general competence in relation to taxation is conferred on the Welsh Assembly.

The main issue in relation to the specific competence under section 108(4) and (5) to impose charges on employers for the services in providing medical treatment and long term care of employees is therefore whether the Bill relates to the "Organisation and funding of national health service". There are two relevant meanings which the term "funding" might ordinarily bear – (1) raising funds or (2) allocating funds.

Interpreting the GOWA 2006 by giving the words their ordinary meaning in their context, I consider that this term has the first of those meanings - raising funds for the Welsh NHS by, for example, charging for the services it provides. I do not consider that it has the second of those relevant meanings – the provision and allocation to the Welsh NHS of the monies made available to the Welsh Consolidated Fund under sections 118-120 of the GOWA 2006. The extensive powers to allocate expenditure from that Fund are governed by sections 124 to 129. It would therefore be unnecessary to include in Schedule 7 a specific power referable to the Welsh NHS. This is not done elsewhere in Part 1 of Schedule 7; for example, another important part of the expenditure of the Welsh Assembly is expenditure on education, but there is no reference under Heading 5 "education and training" to funding. This strongly supports the interpretation of the phrase "funding of National Health Service" in the context in which it appears in the GOWA 2006 as having the first of these meanings.

The submission to the contrary advanced on behalf of the Association of British Insurers (clearly summarised in paras 15 and 16 of the judgment of Lord Mance) was that "organisation and funding of national health service" should be construed by reference to the subordinate legislative powers conferred under the NHS (Wales) Act and effectively limited to those powers. It is necessary to examine the background in some detail.

Prior to the first phase of devolution in 1999 (as explained at para 19 of the judgment in In re Agricultural Sector (Wales) Bill [2014] 1 WLR 2622, section 1(2) of the National Health Service Act 1977 provided that:

"The services so provided must be free of charge except in so far as the making and recovery of charges is expressly provided for by or under any enactment whenever passed."

Sections 77 to 83A of that Act (as amended prior to 1999) enabled charges to be made for specified services; the powers to set prescription and other charges were set out in section 77. These were exercisable by the Secretary of State by subordinate legislation. Under the National Assembly for Wales (Transfer of Functions) Order 1999, the powers of the Secretary of State under the National Health Service Act 1977 were simply transferred to the Welsh Assembly, as under the first phase of devolution the Welsh Assembly only had the power to make subordinate legislation.

In 2006, the UK Parliament consolidated the legislation in relation to the NHS. It enacted for the National Health Service in England the National Health Service Act 2006 and for Wales the NHS (Wales) Act 2006. The powers under section 77 of the National Health Service Act 1977 (as amended) were re-enacted in section 121 of the NHS (Wales) Act as powers to make subordinate legislation. Section 1(2) was re-enacted as section 1(3).

The NHS (Wales) Act was enacted by the United Kingdom Parliament in 2006 three months after the enactment of the GOWA 2006; the provisions of Part 4 and Schedule 7 of the GOWA 2006 conferring on the Welsh Assembly competence to pass primary legislation required a referendum before such competence would take effect. The legislative competence of the Welsh Assembly under the GOWA 2006 was at first limited under Part 3 and Schedule 5 to what was described as the second phase of Welsh devolution in paras 24-26 of the judgment in In re Agricultural Sector (Wales) Bill. Primary legislative provision relating to the Welsh NHS could only therefore be made by the United Kingdom Parliament, unless specific powers were granted to the Welsh Assembly to pass an Assembly Measure under Part 3.

Whilst the competence of the Welsh Assembly was limited under the second phase of Welsh devolution, it was entirely appropriate to consider Part 3 and Schedule 5 of the GOWA 2006 and the NHS (Wales) Act together. It followed that during the currency of the second phase of Welsh devolution amendments to prescription charges were made under subordinate legislation under section 121 of the NHS (Wales) Act. It was through these powers that the National Health Service (Free Prescription and Charges for Drugs and Appliances) (Wales) Regulations 2007 were by the Welsh Assembly made as subordinate legislation constrained by the terms of the NHS (Wales) Act 2006.

However, since Part 4 and Schedule 7 has come into effect after the referendum and has brought about the third phase of Welsh devolution, the Welsh Assembly may within the competence conferred by Part 4 and Schedule 7 amend legislation passed by the United Kingdom Parliament prior to March 2011 or supplement it by new primary legislation.

The construction advanced on behalf of the Association of British Insurers sought to limit the primary legislative competence of the Welsh Assembly in the third phase of devolution under Part 4 and Schedule 7 by reference to the powers originally conferred by legislation of the United Kingdom Parliament on the Secretary of State to make subordinate legislation and continued under the first and second phases of Welsh devolution. Viewed against the background I have set out, I cannot accept the submission.

First the GOWA 2006 and in particular Part 4 and Schedule 7 should, in my view, be construed by reference to the other terms of the GOWA 2006 and not by reference to other statutes of the United Kingdom Parliament such as the NHS (Wales) Act. The position is, in my view, no different to that set out in para 42 of the judgment in In re Agricultural Sector (Wales) Bill with respect to interpreting the legislative competence of the Welsh Assembly. That has to be determined by an interpretation of the terms of Part 4 and Schedule 7 and not by reference to the way in which functions may have been distributed between the United Kingdom Parliament and United Kingdom Ministers on the one hand and the Welsh Assembly on the other hand in the first and second phases of Welsh devolution.

Second, although the provision in section 121 of the NHS (Wales) Act was necessary to enable the Welsh Assembly to exercise subordinate legislative powers before it received primary legislative competence, once it received primary legislative competence, I see no reason to hold that the powers under the GOWA 2006 should remain so limited. Although the provisions of the National Health Service Act 1977 and the NHS (Wales) Act set out detailed provisions setting out what could be done by secondary legislation and what required primary legislation, there is nothing to suggest that Parliament intended these to be of relevance once the Welsh Assembly acquired primary legislative powers.

Third, there is a clear distinction between exercising general tax raising powers and charging for services provided by the NHS. A specific cost can be attributed to the services. The funds so raised can then be used to defer the costs of those services rather than utilising the grant provided to the Welsh Consolidated Fund. Thus it is entirely consistent with the grant to the Welsh Assembly of primary legislative powers in respect of health under Heading 9, that the Welsh Assembly was given competence to vary the NHS (Wales) Act and to charge for services provided without being constrained by the terms of that Act.

If on the interpretation of Heading 9 in its context in the GOWA 2006, funding means raising funds, then it was open to the Welsh Assembly either to amend the provisions of the NHS (Wales) Act which restrict the services for which a charge can be made or to enact primary legislation which imposes charges for services as an enactment within the scope of section 1(3) of the NHS (Wales) Act:

"The services so provided must be free of charge except in so far as the making and recovery of charges is expressly provided for by or under any enactment, whenever passed."

The terms of the NHS (Wales) Act are not, in my view, therefore relevant to limiting the meaning of Heading 9.

I consider for these reasons that funding has the meaning I have set out in para 83 which I have derived from an interpretation of the terms of the GOWA 2006 without reference to the NHS (Wales) Act. In principle, therefore, the Welsh Assembly has competence to enact legislation that makes provision for charging for services by way of the treatment and long term care of those with asbestos-related diseases provided that the moneys so raised are used exclusively for the Welsh NHS.

(b) Charging employees who can recover from their employers

As I have set out, the first and central aim of the Bill is to transfer the cost of medical treatment and care of an employee from the State to the employer in circumstances where the cost would be recoverable as a recognised head of damages from the employer as a tortfeasor. I do not see what objection there could be in law, given my views of legislative competence, to a scheme where the Welsh NHS would have imposed charges directly on such an employee and the employee would have recovered such charges from the employer. Such a scheme would have been analogous to the scheme for Road Traffic accidents set out in the Republic of Ireland's Health Amendment Act 1986. In my view, the Welsh Assembly could also have made provision in such a scheme which ensured that the employee with an asbestos-related disease, though liable for the charges, did not have to pay until reimbursed or indemnified by the employer or the employer's insurers. Moreover, the employers' liability insurers would have had to indemnify the employer under a standard form liability policy when the employer was called on to pay the charges by way of damages.

It is argued on behalf of the Association of British Insurers that such a scheme would be politically objectionable, but I cannot accept that submission. The Welsh Assembly would, in my view, be seen simply as taking steps to change the position of employers so that for the future they would actually meet the costs of treatment and care of a very serious disease which they had caused through their negligence or breach of statutory duty at some time in the past, rather than that cost continuing to be carried by the State. It is difficult to see what political objections there could be to such a scheme in withdrawing the State benefit to employers and their insurers and providing more funds to the Welsh NHS. The benefits of such a scheme for the Welsh NHS would be no different to the machinery proposed by the Bill, though it would be more expensive to administer and undoubtedly risk causing stress to the persons suffering from asbestos-related diseases.

I therefore consider that the Welsh Assembly could, either by amendment to the NHS (Wales) Act or by separate legislation, have permitted the Welsh NHS to charge employees for treatment if they suffered from an asbestos-related disease as a result of the negligence or breach of statutory duty of the employer. Such a scheme would have achieved the first and central aim of the Bill. No part of the liability of the employers or their insurers would have been re-written; they might simply become liable on ordinary principles if their liability to the employee for the asbestos-related disease was established.

(c) The machinery provided for in the Bill

Instead of achieving the first and central aim of the Bill by such a scheme, the Bill seeks to achieve its aims by choosing machinery which can be seen as a better way of collecting such charges directly from employers by imposing liability for such charges on the employers:

(i) Sections 2 and 3 impose liability to pay the charges for treatment by the Welsh NHS directly on any employer who is or is alleged to be liable to any extent in respect of the asbestos-related disease.

(ii) The charges can only be recovered if incurred after the coming into force of the Bill.

(iii) The liability only arises if a payment of compensation in respect of the asbestos-related disease is made to the employee after the coming into force of the Bill.

(iv) Section 5 provides the means by which Welsh Ministers certify the amount of the charges.

(v) Section 15 provides that Welsh Ministers must have regard to the desirability of securing that an amount equal to the funds it received through these payments is spent on research or treatment of asbestos-related diseases.

As the Welsh Assembly has, in my view, competence to impose such charges directly upon the employees, I can see no objection to the competence of the Welsh Assembly under the provisions of section 108(4) and (5) and Heading 9 of Part 1 of Schedule 7 in imposing such charges directly on the employers to achieve the aims of the Bill. Lord Mance suggests (para 33) that it is not relevant to consider in the context of legislative competence what might have been done. Although I agree that what might have been done may not generally be relevant, that is not the analysis I have set out. I have simply sought by this means to demonstrate that in reality, the imposition of direct liability on employers is no more than machinery for the collection of charges for services which, on my interpretation of Heading 9 of Schedule 7 the Welsh Assembly has legislative competence to impose.

If charges are to be imposed for NHS services in the Welsh NHS, then, in my view, the monies collected have to be used to fund the Welsh NHS, as that is the sole purpose for which there is legislative competence to raise funds by way of the imposition of charges. Section 15 requires Ministers, in the exercise of their functions under the NHS (Wales) Act to have regard to the "desirability" that an amount equal to the monies raised are applied "for the purposes of research into, treatment of, or other services relating to, asbestos-related diseases". Is that sufficient? In my view it is. In the context of the duties under the NHS (Wales) Act, the provision does no more than to require Ministers to have regard to the desirability of applying the monies so collected specifically in relation to asbestos diseases within the work of the Welsh NHS. It does not permit them to use it for any purpose other than for the Welsh NHS.

I thus consider that there is legislative competence under section 108(4) and (5) to impose charges under section 2 directly on employers.

(d) Is there retrospectivity in respect of the liability imposed on employers?

Although the charges which can be recovered are only those that are incurred after the coming into force of the Bill and the liability to pay Ministers arises only where a compensation payment is made after the coming into force of the Bill, there is an element of retrospectivity in the imposition of the machinery of direct liability on employers. The liability imposed, though only in respect of future charges, is retrospective, as it is a new liability owed directly to Welsh Ministers which arises only by reason of negligence or breach of statutory duty which had occurred prior to the coming into force of the Bill. It is not simply an obligation to make future payments to an employee in respect of a recognised head of damages for an established liability, as would be the case if the machinery adopted had been to impose charges directly on the employees and recovery been obtained from employers. In the case of the employers, prior to the Bill, they would have had no such direct liability to Welsh Ministers. Thus the second aim and effect of the Bill has an element of retrospectivity.

I therefore agree with Lord Mance that imposing such direct liabilities retrospectively can be viewed as amounting to the "deprivation" of the "possessions" of the employers (and others within section 2) so as to engage A1P1.

(2) The effect of A1P1

(a) The applicable principles under A1P1

I gratefully adopt the summary of the general principles applicable to A1P1 set out at paras 44 to 53 of Lord Mance's judgment. The paragraphs trace the development in the increase in the jurisdiction of the judicial branch of the State and of the Strasbourg Court under A1P1 to review the judgement of a legislative branch of the State in relation to the legislation it has enacted. I agree that in the light of the judgments in AXA General Insurance Ltd v Lord Advocate [2011] UKSC 46, [2012] 1 AC 868 there are two separate questions which arise. These are:

(i) Can it be said that the judgement of the Welsh Assembly was manifestly unreasonable in its decision to legislate first to make employers bear the future cost of medical treatment of a disease they had caused rather than such costs being borne by the State and secondly to impose machinery that creates a new direct liability? This can be properly described as the issue of legitimate aims.

(ii) Was a fair balance struck, in the judgement of the court, between the demands of the general interest of the community and the requirements of the protection of the employer's fundamental rights? This can properly be described as the issue of proportionality.

(b) Was the Welsh Assembly entitled to view the Bill as having legitimate aims?

I turn therefore to consider the first question. I have set out the main and subsidiary aims of the Bill at para 77. Those aims must be viewed in the social and economic context of Wales and the legislative competence of the Welsh Assembly to which I have referred:

(i) Since the establishment of the NHS in the United Kingdom in 1946, the general expectation has been that it would provide medical treatment and care free at the point of delivery, subject to limited exceptions, such as prescription charges. However, it does not follow from that general expectation that a legislature with responsibility for the NHS cannot change the extent to which its services are funded by the State so that they are not free at the point of delivery. Indeed, charges for NHS services (such as prescription charges) have been imposed or increased on many occasions.

(ii) In Wales there was a concentration of heavy industry. Wales, along with some other parts of the United Kingdom, has a long and direct experience of serious industrial diseases, such as pneumoconiosis, and their devastating effect on employees. It has long been seen as a matter of social justice that proper compensation and care be provided at the expense of employers in those industries to those suffering from such diseases through negligent acts and breach of statutory duty. Given the period of time that elapses after exposure to asbestos before the disease manifests itself, it cannot be an objection that the wrongdoing occurred many years ago.

(iii) The cost of the provision of health services through the Welsh NHS is an escalating cost. One of the reasons is the effect of serious industrial diseases caused by the concentration of heavy industry and the past negligence and breach of statutory duty by employers. The cost of the Welsh NHS is now a very significant part of the expenditure of the Welsh Assembly which has to be met out of the overall grant to the Welsh Assembly by HM Treasury, as described at para 83 above. There can be little doubt that provision of finance for the Welsh NHS and the Welsh NHS' continued ability to provide the requisite health services out of monies made available to it out of the grant to the Welsh Assembly by Her Majesty's Treasury is a matter of pressing legitimate concern to the Welsh Assembly.

Taking into account this context, I consider that the first and central aim of making the employer (when a tortfeasor) pay for the cost of treating the disease caused by it, is an aim which the Welsh Assembly, as a democratically elected legislature within its area of primary legislative competence, is entitled to reach and has an entirely reasonable foundation.

(i) Given the choices which are open to a democratically elected legislature in how the escalating overall cost of health care is to be met and taking account of the very long period of time before an asbestos-related disease caused by the employer manifests itself, the Welsh Assembly has to make a judgement. It must be entitled to consider in such circumstances which benefits and services it is to continue to provide free of charge. I cannot therefore see a reason why it is not open to the Welsh Assembly to make a judgement that there is a real social and economic need to withdraw the benefit of free medical treatment and care and impose charges on the employers in industries where negligence or breach of statutory duty has occurred in the past.

(ii) The fact that the consequences of such wrongdoing take years to manifest themselves and the escalating cost of treating and caring for those suffering from the diseases can indeed be seen as providing a justification for the Welsh Assembly, in the context I have set out, in withdrawing the benefit hitherto provided and allowing the cost to be borne by those tortfeasors in the same way that those tortfeasors bear the other costs of their wrongdoing which has brought about the diseases.

(iii) I cannot therefore see a basis for contending that the Welsh Assembly is not reasonably entitled to reach a judgement that there is a strong public interest in doing so. Nor can I see the basis for questioning as reasonable the judgement of the Welsh Assembly that it would be desirable that the funds so raised would directly benefit those suffering from asbestos-related diseases.

(iv) Choices have to be made in setting overall policy in relation to the level of service, treatment and care to be provided by a national health service, the funding of such services and the services in respect of which charges are to be made. These are choices of social and economic policy which in my view can and should only be made by the Welsh Assembly as a democratically elected legislature.

(v) The Welsh Assembly is also entitled to make the judgement that instead of a scheme which would have involved levying a charge on employees and collecting it from the employers through a scheme of the type I have described at paras 96-98, machinery for direct collection would confer a further benefit on those suffering from asbestos-related diseases by relieving them of further worry and stress.

(vi) That public interest can therefore be seen as reflecting choices of social and economic policy and of social justice in Wales which may be different to the views of social and economic policy and social justice reasonably held in other parts of the United Kingdom or by other people. As these choices are being exercised in matters within the primary legislative competence of the democratically elected Welsh Assembly, the Welsh Assembly is, in my view, reasonably entitled to adopt such choices and views for Wales.

For these reasons therefore the Welsh Assembly's objective in making the tortfeasor pay rather than the public as a whole is a choice which can properly be regarded as having an economic and social purpose. This is clearly an objective on which different views can reasonably be held. However, it is in every respect pre-eminently a political judgement in relation to social and economic policy on which it is for the legislative branch of the State to reach a judgement. The judicial branch of the State should not therefore question this first and central aim of the Bill, as there are manifestly reasonable grounds for reaching the view which the Welsh Assembly has reached: AXA General Insurance Ltd v Lord Advocate at para 49 and following.

(c) No objection could be taken to charges being claimed by employees from employers

If the Welsh Assembly had imposed charges directly on employees as I have set out at paras 96-98 and thus limited the Bill to the first and central aim, there could, in my view, be no question of any rights of employers being affected in any impermissible way. The employers may have had an expectation that the cost of medical treatment and care of a disease caused by their wrongdoing in the past would always be met by the State through the NHS budget rather than by them; and that they would therefore continue to enjoy a benefit from the State in respect of their past wrongdoing.

However, such an expectation gave them no legitimate expectation giving rise to legal rights. A legislature would not be constrained by A1P1 from enacting primary legislation to make them liable for future payments in respect of their past wrongdoing as were made after the coming into force of the Bill because it was doing no more than withdrawing a benefit to which employers had no entitlement to enjoy for an unlimited period of time. Thus, even though the obligation to make such payments arose out of a liability to the employee that had arisen in the past, there would be no issue of retrospectivity.

(i) The payments would be in respect of a recognised head of damages caused by an asbestos-related disease or condition for which liability under existing law had been incurred. The position is, in my view, different to that in AXA General Insurance Ltd v HM Advocate. The Damages (Asbestos-related Conditions) (Scotland) Act 2009 imposed liability for a condition, asymptomatic pleural plaques, where it had been declared by the courts that there was no liability under existing law. The Bill in the Welsh Assembly has imposed no new liability in respect of responsibility for the asbestos-related disease or condition. The Bill is premised on existing liability for the disease, the existing consequent liability to pay damages and an existing well-recognised head of damages, namely medical treatment and care.

(ii) The payments would only be payments made after the coming into force of the Bill.

(iii) The payments could not be recovered in cases where a settlement had been made of the liability incurred by the employer, as the liability would have been discharged.