Tameside & Glossop Acute Services NHS Trust v Thompstone & Ors

This is the judgment of the court to which all members have contributed. It deals with a number of appeals from decisions concerned with the making of Periodical Payment Orders (PPOs) under section 2 of the Damages Act 1996.

The decisions appealed are all concerned with very serious injury to young claimants, suffered at birth as a result of negligence, for which liability had been admitted by different Health Authorities or different Healthcare Trusts i.e. in broad terms the NHS. The damages for future care were in large measure agreed but the question whether a PPO should be made and if so what the form of the order should be was in issue. In such cases, by section 2(1) of the Damages Act 1996 (brought in by way of amendment by the Courts Act 2003) the court has, since 1^st April 2005, been required to consider whether to make an order for periodical payments. Section 2(8) and (9) provide as follows:-

“2(8) An order for periodical payments shall be treated as providing for the amount of payments to vary by reference to the retail prices index (within the meaning of section 833(2) of the Income and Corporation Taxes Act 1988) at such times, and in such a manner, as may be determined by or in accordance with Civil Procedure Rules.

(9)

But an order for periodical payments may include provision-

(a)

disapplying subsection (8), or

(b)

modifying the effect of subsection (8).”

In Flora v Wakom (Heathrow) Ltd [2006] EWCA Civ 1103 the construction of these subsections was considered in the following circumstances. The claimant in that case had been severely injured and claimed damages for inter alia loss of future earnings and future care. The defendant admitted liability. In the statement of case relating to the award of damages, the claimant contended that, if the court made an order for periodic payments, it should disapply or modify section 2(8) and provide for the amount of such payments to vary by reference to a wage-related index rather than the retail prices index (RPI). To support that case, the claimant sought to adduce the expert evidence of Dr Victoria Wass, an academic labour economist based at the Cardiff Business School. The defendants applied to strike out the relevant parts of the statement of case and to exclude the evidence of Dr Wass on the grounds that use could only be made of section 2(9) in exceptional circumstances. Sir Michael Turner refused the defendant’s applications and the Court of Appeal dismissed an appeal, holding that section 2(8) identified a default position and that section 2(9) allowed the court to make the orders identified therein not simply in exceptional circumstances but whenever it appeared appropriate and fair to do so.

In dealing with a point made by Mr Pooles, for the defendants, as to the risk that, if exceptionality was not the test, the courts faced the prospect of trials at which a host of expensive witnesses would be called on each side, Brooke LJ (with whose judgment Sir Mark Potter and Moore-Bick LJ agreed) said this:-

“33.

We are now dealing with a different statutory provision and, if the experience of the past is any useful guide, it is likely that there will be a number of trials at which the expert evidence on each side can be thoroughly tested. A group of appeals will then be brought to this court to enable it to give definitive guidance in the light of the findings of fact made by a number of trial judges. The armies of experts will then be able to strike their tents and return to the offices or academic groves from which they came.”

We now have before us this group of appeals. They are four in number which we will identify simply by the name of the claimants: Thompstone, a decision of Swift J whose judgment was handed down on 23^rd November 2006; De Haas, a decision of Nelson J given on 24^th November 2006; Corbett, a decision of His Honour Judge Bullimore sitting as a High Court Judge handed down 28 March 2007; and RH, a decision of Mackay J handed down on 20 June 2007. In all, liability was admitted and all are concerned with severely injured claimants claiming future losses, particularly costs of future care.

The decision of Swift J was the first in time. The parties had agreed that all future losses should be paid as a lump sum save for the costs of care and case management. The claimant sought a PPO linked to a wage-related index; the defendant wanted RPI to apply. Swift J rejected arguments seeking to distinguish Flora. After considering extensive expert evidence, she took the view that, because wages would increase at a faster rate than prices measured by the RPI and the RPI was thus not likely to be a reliable or accurate indicator of growth in earnings, it was right to investigate alternatives. She did so and considered the advantages and disadvantages of several measures and indices, including the Annual Survey of Hours and Earnings (ASHE) for the occupational group of care assistants and home carers (Occupational Group 6115), produced by the Office of National Statistics (ONS). She concluded that

“it would be fair and reasonable under the provisions of section 2(9) to modify the effect of section 2(8) by providing for the amount of payments to vary by reference to the 75^th percentile of ASHE occupational group 6115.”

In De Haas, Nelson J was unpersuaded that there were arguments unaddressed in Flora or by Swift J, whose decision, with permission to appeal, had been promulgated the day before the hearing in De Haas. He had approved the settlement of the claimant’s claim and was concerned with the form of the order. He decided that there should be a lump sum payment for all heads of damage other than the costs of future care and case management. He adjourned the question of indexation because the claimant by her litigation friend accepted that she would be bound by the result of the proposed appeal in Thompstone.

HH Judge Bullimore and Mackay J applied their own minds to the issues argued out before Swift J and independently rejected arguments seeking to distinguish Flora. They also independently rejected the RPI as a suitable index by reference to which to increase the wages of carers and independently held, after considering extensive expert evidence, that the appropriate measure for indexation was ASHE 6115. Mackay J was also assisted by the fact that on the last day of the hearing before him, 25^th May 2007, Lloyd-Jones J handed down a judgment in Sarwar v Ali and MIB [2007] EWHC 1255. That decision has not been appealed but he also reached the conclusion that RPI was not an appropriate measure by which to index PPOs covering losses relating to wages, and in relation to future care he too applied ASHE 6115; he also used a wage-related measure to index a PPO covering the loss of future wages.

It could be said that the decision in Flora should be treated as saying no more than that the claimant’s contentions were arguable and that he should be allowed to call the evidence of Dr Wass at trial. However, in all these cases Flora has not been treated as simply a decision on arguability but was - we think rightly - treated as authority on the construction of section 2(8) and (9) albeit that attempts were made to limit what Flora actually decided. In this court on the appeals, Flora has rightly been treated as binding on us but again subject to a submission that some points are not covered by the decision. Permission to appeal to the House of Lords was refused in Flora both by this court and the House of Lords, but in the latter case on the basis as their Lordships indicated that it was too early for such an important point to be considered by them.

Those representing the appellants wish to reserve for future argument in the House of Lords that Flora was wrongly decided, but, as we have said, have accepted that a full frontal attack is not open to them in this court. They have however sought to suggest that certain arguments were not considered in Flora and that, accordingly, it is still open to the court to reach, by another route, the conclusion that Flora rejected. In any event, they seek to argue that the power to modify the effect of section 2(8) was more constrained than the wide language of Flora would indicate.

Those representing the appellants recognise that developing the arguments they have sought to develop on the ambit of section 2(8) and (9) post-Flora without being able to make the full frontal attack on Flora has been and remains a difficult task. In this court, it is made the more difficult by the fact that none of the arguments has found favour before any of the judges at first instance. Mr Philip Havers QC and Mr Paul Rees QC recognised the uphill struggle they faced. Furthermore, if they lose the construction arguments, the appellants then ask this court to review the evaluation exercise as to the appropriate index or measure to be adopted in the individual cases on which the trial judges were unanimous in their choice. That again provides the appellants with a difficult hill to climb. They have sought to dress up the points on evaluation as points of law in order to get over the obvious difficulty that this court would be unlikely to interfere with a judge’s factual evaluation exercise, but their arguments must be seen for what they in fact are.

All that said, however, this is the first opportunity for this court to consider the operation of these subsections in practice, and in a context where it is likely that an attempt will be made to take the whole question of their proper construction to the House of Lords. The task of this court, as we see it, is thus to consider the approach of the judges in these cases and by reference thereto to give guidance that will assist those who hereafter have to deal with this type of case. We hope that it will enable the experts to strike their tents and retreat, as Brooke LJ so figuratively put it, but, if not, at least it might help to focus their attention on the needs of particular claimants and the particular circumstances of individual cases, and might help to make agreement between those experts more likely. We must also keep in mind the possibility that Flora, together with these cases, may at some stage be considered at a higher appellate level.

Section 2(1) of the 1996 Act (as substituted by the Courts Act 2003) provides:

"A court awarding damages for future pecuniary loss in respect of personal injury:-

(a)

may order that the damages are wholly or partly to take the form of periodical payments, and

(b)

shall consider whether to make the order."

The sections 2(8) and (9) we have already set out.

CPR 41.7 further provides:

"When considering -

(a)

[inapplicable in this case]

(b)

whether to make an order under section 2(1)(a) of the 1996 Act,

the court shall have regard to all the circumstances of the case and in particular the form of award which best meets the claimant's needs, having regard to the factors set out in the practice direction."

The relevant practice direction provides, at 41BPD.1:

"The factors which the court shall have regard to under rule 41.7 include:

(1)

the scale of the annual payments taking into account any deductions for contributory negligence;

(2)

the form of award preferred by the claimant including-

(a)

the reasons for the claimant's preference; and

(b)

the nature of any financial advice received by the claimant when considering the form of award; and

(3)

the form of award preferred by the defendant including the reasons for the defendant's preference.”

We cannot do better than to take the background from the first of the judgments we have to consider, that of Swift J in Thompstone:-

“Lump Sum Awards

14.

In the past, damages awards have in general been paid by way of a single lump sum, comprising (as applicable) awards for pain, suffering and loss of amenity, past losses, interest and future losses. In the case of a claim for future loss, a lump sum is calculated by reference to a series of multiplicands, representing the claimant's future annual loss or cost in respect of each head of damage. To those multiplicands are applied multipliers derived from a combination of the claimant's estimated life expectancy and a discount rate based on an assumption that the capital, notionally invested, will produce an annual return equivalent to 2.5% net per annum (pa) over and above the increase in the RPI.

15.

A lump sum award has a number of advantages. It is final, simple and offers flexibility to a claimant, who can decide for himself how to prioritise his various needs and wants. However, the 'once and for all' approach frequently results in over- or under-compensation. The multiplier is calculated by reference to average life expectancy which may have little bearing on the actual life expectancy of the individual claimant concerned. A claimant may die well before his expected time; in that event, the defendant cannot recover the excess of damages paid and that excess constitutes a windfall for the claimant's family. Conversely, a claimant may survive longer than expected, in which case his damages may be insufficient to meet his needs during the last years of his life. Investment returns will vary according to an individual's investment strategy and the economic conditions prevailing at the time. The returns may be above or below those assumed by the discount rate. In addition, a claimant's needs may alter from those anticipated at the time of trial or settlement, and costs which were estimated at rates current at the trial date may increase significantly thereafter.

Structured Settlements

16.

In an attempt to meet some of these difficulties and, in particular, to ensure that claimants were protected against the risk that, at some point, their damages might be exhausted, a new means of paying damages by way of 'structured settlement' was developed. From 1988, the purchase of an annuity by a defendant or its insurer out of, and at the same time, as an award of compensation resulted in the annual payments from the annuity being tax-free in a claimant's hands. The first structured settlement was agreed in 1989 and provided for the payment to the claimant of a combination of a lump sum and a stream of tax-free payments payable for her lifetime. Since 1989, there have been various developments in the arrangements for structured settlements which I need not describe here. Various factors have served to limit the number of claims resolved by means of a structured settlement. A major restriction was that structured settlements could be implemented only with the agreement of both parties.

Changes to the System

17.

Section 2(1) of the 1996 Act permitted the court to make an order that damages were wholly or partly to take the form of periodical payments. Such an order could, however, only be made if both parties agreed.

18.

The shortcomings of the existing system were highlighted by Lord Steyn in the leading case of Wells v Wells [1999] 1 AC 345 at 384B:

" … there is a major structural flaw in the present system. It is the inflexibility of the lump sum system which requires an assessment of damages once and for all of future pecuniary losses. In the case of the great majority of relatively minor injuries the plaintiff will have recovered before his damages are assessed and the lump sum system works satisfactorily. But the lump sum system causes acute problems in cases of serious injuries with consequences enduring after the assessment of damages. In such cases the judge must often resort to guesswork about the future. Inevitably, judges will strain to ensure that a seriously injured plaintiff is properly cared for whatever the future may have in store for him. It is a wasteful system since the courts are sometimes compelled to award large sums that turn out not to be needed. It is true, of course, that there is statutory provision for periodic payments: see section 2 of the Damages Act 1996. But the court only has this power if both parties agree. Such agreement is never, or virtually never, forthcoming. The present power to order periodic payments is a dead letter. The solution is relatively straightforward. The court ought to be given the power of its own motion to make an award for periodic payments rather than a lump sum in appropriate cases. Such a power is perfectly consistent with the principle of full compensation for pecuniary loss. Except perhaps for the distaste of personal injury lawyers for change to a familiar system, I can think of no substantial argument to the contrary. But the judges cannot make the change. Only Parliament can solve the problem."

19.

In 2002, the Master of the Rolls' Working Party published its report entitled "Structured Settlements". The disadvantages of a conventional lump sum award were summarised at paragraph 12 of the report:

"The one thing which is certain about a once and for all lump sum award in respect of future loss is that it will inevitably either over-compensate or under-compensate. This will happen particularly where the claimant survives beyond the life expectancy estimated at the time of trial, or alternatively dies earlier. It will frequently be the case in practice that there is over-compensation in six figure sums, or, correspondingly, that a combination of increased life expectancy, the cost of care, and (it may be) the cost of new but necessary medical treatments is such that the sum needed exceeds anything that might have been awarded at the date of trial."

20.

The Working Party therefore concluded at paragraph 21:

" … of the features we have identified that of accuracy is the most important. We are concerned that a consequence of a system of once and for all lump sum awards is that there will be under or over-compensation (in some cases considerable) and particularly concerned that a proportion of claimants whose life expectancy is uncertain, and who need significant continuing care, might be left with significant uncompensated need. It adds to our concern that this is likely to occur later in life when the consequences will be particularly hard to manage. It is also of concern that appreciation of this may give rise to excessive prudence and under expenditure in earlier years. Accordingly, we prefer a system that is better able to meet future needs as and when they arise. Such a system may also have its defects – as we shall go on to point out – but we believe the advantages outweigh them."

21.

Following the report of the Working Party, amendments were made to the 1996 Act. As a result of these amendments, contained in the Courts Act 2003, it has since 1 April 2005 been open to a court to make an order for periodical payments whether or not the parties agree. Indeed, it is now mandatory, when a court is making an award of damages for future pecuniary loss in respect of personal injury, for it to consider whether the damages - or part of them – should be paid by way of periodical payments.

Periodical Payments Orders

22.

Periodical payments avoid many of the problems caused by lump sum awards. They provide a guarantee for a claimant that he will continue to receive regular annual payments for the duration of his life so that his damages will never be exhausted. The annual payments will be free of tax, which removes any uncertainties associated with possible future changes to the arrangements for taxation of investment income. The fact that the annual payments cease on death means that there is far less risk of large sums paid by defendants going to persons other than the injured person for whose benefit they were intended. All these factors represent considerable advantages over the lump sum award.

23.

When a lump sum award is made, the responsibility for deciding how to invest his damages lies with the claimant or those acting on his behalf. It is for him (or them) to husband his resources carefully, with a view to ensuring that they have the best possible opportunity to meet his lifelong needs. Where the claimant has a large sum of damages and a long life expectancy, his damages will require skilled financial management in order, not only to guard against the risk that he might survive longer than the estimate of life expectancy on which the multiplier was calculated, but also to meet the cost of his annual needs, which cost will increase year by year from the date when the award is made.

24.

The management of a claimant's damages involves taking decisions about risk. All investments carry with them some element of risk. The more risk an investor is prepared to take, the greater the potential returns. An investment that offers the prospect of high returns in terms of capital growth also in general carries with it the risk that the capital value of the investment might decrease. An investment carrying a lower rate of risk is likely to produce lower returns but greater security of capital. Ideally, a claimant who is entirely reliant on his damages for his future quality of life should not be compelled to expose himself to a significant degree of investment risk. However, for a claimant with a long life expectancy and a 'once and for all' damages award, this is often unavoidable. He will probably be advised to invest in a combination of low risk and higher risk investments, this being the only way in which he can hope to be in a position to meet increasing annual costs and the potential additional expense associated with a longer than expected life expectancy.

25.

One important effect of a periodical payments order is to transfer the risk associated with the investment of damages away from the claimant. The award will guarantee the claimant annual payments for his lifetime. Furthermore, the amendments to the 1996 Act provide for uplifts to the annual payments in accordance with the RPI. Provided that indexation to the RPI accurately reflects actual increases in the relevant annual costs, the claimant will be protected against the effects of future increases in those costs.

26.

If, however, the annual uplifts do not properly reflect the actual increases in the relevant costs, the claimant will have no means of protecting himself against the consequences of the shortfall. He will not have the option of pursuing an investment policy aimed at achieving capital growth in order to meet shortfalls in the future. The transfer of risk away from him results also in a loss of opportunity for gain. He could be caught in a situation whereby his annual payments fall further and further below the level which, at the time the periodical payments order was made, was agreed or assessed as being required to meet his needs.

Awards for Future Care Costs

27.

In most serious personal injury cases, the largest component of the award of damages is the cost of future care. The provision of a proper level of good quality care throughout life is vital to a severely injured claimant's future well being and quality of life. The importance of this head of damage has been recognised by the courts. In Wells, Lord Hutton said at 403C-D:

"Unlike the great majority of persons who invest their capital, it is vital for the plaintiffs that they receive constant and costly nursing care for the remainder of their lives and that they should be able to pay for it, and any fall in income or depreciation in the capital value of their investments will affect them much more severely than persons in better health who depend on their investments for support."

In the same case, Lord Hope observed at 400E:

"Whatever policy reasons there might have been for regarding it as acceptable that there may be less than a full recovery in regard to wage loss – and I should make it clear that I do not subscribe to that policy – there can be no good reason for a shortfall in the amount required for future care or to meet all the other outlays which have been rendered necessary by the disability. The calculation should make the best use of such tools to assist that process as are available."

Lord Clyde made the same point at 394G:

"The problem of sufficiently providing for the future care of the very severely disabled plaintiff gives rise to particular concern since any inadequacy of the award in that respect could be particularly serious."

28.

The effect of a shortfall in the damages available for future care is that the claimant will be dependent on the State to make up the balance needed to meet his care needs. Such assistance may or may not be forthcoming.”

The principle that there should be full compensation (the 100% principle) was fundamental to the decision in Flora and to the decisions of all the trial judges who have had to consider the matter. It is dealt with most fully in the judgment of Swift J but was of no less importance to the others (see in particular to HH Judge Bullimore in Corbett paragraph 92, Mackay J in RH paragraph 69 and Lloyd Jones J in Sarwar paragraph 113). Swift J cited the most relevant and important passages and again it is convenient to take the following passage from her judgment:-

“The Principle of Full Compensation

29.

Awards of damages should be calculated so as to achieve, as nearly as possible, full compensation for the claimant. In Wells, Lord Hope set out what is sometimes known as the '100% principle' at 390A:

“ … the object of the award of damages for future expenditure is to place the injured party as nearly as possible in the same financial position as he or she would have been in but for the accident. The aim is to award such a sum of money as will amount to no more, and at the same time no less, than the net loss."

In the same case, Lord Lloyd said at 363H:

“It is of the nature of a lump sum payment that it may, in respect of future pecuniary loss, prove to be either too little or too much. So far as the multiplier is concerned, the plaintiff may die the next day, or he may live beyond his normal expectation of life. So far as the multiplicand is concerned, the cost of future care may exceed everyone's best estimate. Or a new cure or less expensive form of treatment may be discovered. But these uncertainties do not affect the basic principle. The purpose of the award is to put the plaintiff in the same position, financially, as if he had not been injured. The sum should be calculated as accurately as possible, making just allowance, where this is appropriate, for contingencies. But once the calculation is done, there is no justification for imposing an artificial cap on the multiplier. There is no room for a judicial scaling down.”

The next two paragraphs in Swift J’s judgment we quote because they seem to us to reflect accurately the proper approach to subsections (8) and (9):-

“30.

In the case of Flora, the Court considered the '100% principle' in the context of the indexation of periodical payments. At paragraph 19 of his judgment, Brooke LJ said:

"There is no indication in s 2 of the 1996 Act, as substituted, that Parliament intended to depart from this well-known principle (i.e. the 100% principle) … "

He repeated his view at paragraph 28 and then went on to say at paragraph 29:

"For this reason I reject the argument that in enacting s 2(8) and 2(9) of the 1996 Act Parliament must be taken to have intended to provide compensation lower than that which would be awarded through adherence to the 100% principle if a periodical payments order was to be made."

31.

It follows, therefore, that indexation of a periodical payments order must be directed at ensuring, as far as is possible, that the real value of the annual payments is retained over the whole period for which the payments will be payable. Any other result would mean that the '100% principle' was breached and would result in injustice to a claimant in respect of whom a periodical payments order was made. It would also have the effect of rendering the new statutory provisions nugatory. If it were clear to claimants' advisers and to courts hearing personal injury cases that the real value of periodical payments would not be retained in the future, orders for periodical payments would not be agreed or made. Instead, awards of damages by way of lump sums – with all the attendant disadvantages which have been identified and addressed – would continue to be the norm. It was this possibility to which Brooke LJ referred in Flora when he said at paragraph 35:

"In enacting s 2 of the 1998 (sic) Act, as substituted, it cannot have been Parliament's purpose to create a scheme which no properly advised claimant would ever wish to use."”

In making their submissions on behalf of the appellant defendants, before the trial judges and before us, Mr Havers and Mr Rees accepted that the 100% principle must be applied. They maintained before the trial judges and before us that their arguments did not cause a departure from it. The trial judges were unconvinced and, as we made clear at the hearing, we also remained unconvinced.

It is convenient, against the above background, to turn to the arguments raised on these appeals. Although the grounds of appeal in each of the four cases were put in slightly different terms, the appellants helpfully amalgamated all the points to be decided on by this court into a list of issues. We will deal with these either singly or in groups, excluding those which were not pursued.

We will consider issues 1 and 2 together. Issue 1 requires the court to consider:

whether, as a matter of law and statutory construction, section 2(8) of the Damages act 1996 can only be modified in ‘exceptional circumstances’.

Issue 2 asks:

whether as a matter of law and precedent section 2(8) of the Damages Act 1996 can only be modified in ‘exceptional circumstances’.

It was accepted before the trial judges and before us that Flora had rejected the argument that it was only in exceptional circumstances that section 2(9) could be used to modify section 2(8). A convenient summary of what Flora decided is contained in the judgment of Lloyd Jones J in Sarwar:-

“111.

This new system governing periodical payments was considered by the Court of Appeal in Flora v Wakom (Heathrow) Limited [2006] EWCA Civ. 1103.  . . . .   The judgment of Brooke L.J., with which the other members of the court agreed, establishes the following principles:

(1)

If a periodical payments order does not identify on its face the manner in which the amount of the payments is to vary in order to maintain their real value, the effect of section 2(8) is that it is to be treated as providing for what is set out in that subsection unless the order contains a provision of a type identified in section 2(9).  There is nothing in the language of these sub-sections to suggest that the power to make provision such as is identified in section 2(9) may only be exercised in an exceptional case. (Paragraph 10)

(2)

There is no indication in section 2 of the 1996 Act, as substituted, that Parliament intended the courts to depart from the "100% principle" formulated by Lord Blackburn in Livingstone v Rawyards Coal Company [1880] 5 App. Cas. 25, 39 and by Lord Hope in Wells v Wells [1999] 1 AC 345, 390 A-B, namely that a victim of a tort is entitled to be compensated as nearly as possible in full for all pecuniary losses.  Accordingly, the Court of Appeal rejected the submission that in enacting subsections 2(8) and (9) of the 1996 Act Parliament must be taken to have intended to provide compensation lower than that which would be awarded through adherence to the 100% principle, if a periodical payments order was to be made. (Paragraphs 18, 19, 27-29)

(3)

In a case where periodical payments are claimed, the trial judge should decide whether it is appropriate to use the powers given by Parliament in sub-section 2(9) and make such an order for index-linking the periodical payments (if a periodical payments order is in fact made) as he considers appropriate and fair in all the circumstances, without being obliged to detect exceptional circumstances before he is at liberty to depart from indexation linked to the RPI. (Paragraph 37)”

All the trial judges accepted the binding nature of Flora but, before Mackay J in RH and it would seem before Nelson J in de Haas and then before us, an attempt was made to argue that Flora had been decided per incuriam and thus that it was not binding. It was said that an argument based on the decision in the Court of Appeal in Cooke v United Bristol Healthcare NHS Trust [2004] 1 WLR 251, as to why section 2(9) should only apply in exceptional circumstances, had not been made or considered in Flora. Cooke was concerned with an award of a lump sum for future care in relation to a seriously injured claimant, where the Court of Appeal disallowed what it held to be an attempt to get round the discount rate laid down by the Lord Chancellor under section 1 of the Damages Act, a rate effectively calculated by reference to the RPI. The argument in broad terms is that the same principle should apply when assessing PPOs as applies when assessing lump sum damages, and if Flora is right that will not be so.

The argument in more detail goes as follows. It traces the background to Cooke chronologically as follows. First, by s.1 of the Damages Act 1996, it was provided as follows:-

“1(1) In determining the return to be expected from the investment of a sum awarded as damages for future pecuniary loss in an action for personal injury, the court shall, subject to and in accordance with rules of court made for the purposes of this section, take into account such rate of return (if any) as may from time to time be prescribed by an order made by the Lord Chancellor.”

(2)

Subsection (1) above shall not however prevent the court taking a different rate of return into account if any party to the proceedings shows that it is more appropriate in the case in question.”

Before the Lord Chancellor prescribed a rate in accordance with s.1(1), Wells v Wells was decided by the House of Lords and by their decision they ruled that, pending the Lord Chancellor exercising his power under s.1(1), 3% should be the discount rate assessed by reference to Index-linked Government Stock (ILGS). Again, prior to the Lord Chancellor prescribing a rate, Wells had been interpreted as saying that 3% should be the discount rate “save in very exceptional circumstances” (see Stuart-Smith LJ in Warren v Northern General Hospital NHS Trust (2) [2000] 1 WLR 1404 at 1408).

When, on 25^th June 2001, the Lord Chancellor prescribed a discount rate of 2.5% he published reasons. Those reasons were first published on 27^th June 2001 but, following his attention being brought to an error, further reasons were published on 27^th July 2001. The key reasons are summarised by Dyson LJ in Warriner v Warriner [2002] 1WLR 1703 at paragraph 14 in the following terms:-

“[a] the market in ILGS was distorted so that the prevailing yields were artificially low; [b] the Court of Protection, even in the wake of the decision of the House of Lords in Wells v Wells had continued to invest on behalf of claimants in multi-asset portfolios, such that real rates of return well in excess of 2.5% could be expected; and [c] it was likely that that “real” claimants with larger awards of compensation would not be advised to invest solely, or even primarily, in ILGS, but rather in a mixed portfolio.”

Dyson LJ then quoted the Lord Chancellor as saying:-

“Finally, in deciding that a single rate of 2.5% should have been set by me on 25 June 2001, I have borne in mind that it will, of course, remain open for the courts, under section 1(2) of the Damages Act 1996, to adopt a different rate in any particular case if there are exceptional circumstances which justify it in doing so.”

In Warriner v Warriner the Court of Appeal was concerned with a case in which an expert (Mr Hogg) was suggesting that the court should exercise its power under s.1(2) to apply a lower discount than 2.5% on the ground that it was, as the sub-section required, “more appropriate” in the case in question. The argument on behalf of the defendant on the appeal was that, albeit “more appropriate” were the words of the sub-section, the Lord Chancellor’s statement that sub-section 1(2) would be applied in “exceptional circumstances” was correct, in that it could never be “more appropriate” to apply a different rate unless the circumstances were outside those taken into account by the Lord Chancellor in fixing the rate at 2.5%.

In Cooke, the expert for the claimant (indeed it would appear the same Mr Hogg) sought to deal with the fact that a lump sum for care, if discounted at 2.5%, would not provide full compensation but, recognising that the case was not exceptional, advised that there should be a variation in the multiplicand.

When the case came to the Court of Appeal, that court held that it was impermissible to alter the multiplicand. Laws LJ said this:-

“30.

In the end, the central issue in these appeals falls to be resolved upon what, I have to say, seems to me to be a very straightforward basis. Once it is accepted that the discount rate is intended in any given personal injury case to be the only factor (in the equation ultimately yielding the claimant's lump sum payment) to allow for any future inflation relevant to the case, then the multiplicand cannot be taken as allowing for the same thing, or any part of it, without usurping the basis on which the multiplier has been fixed. And it must be accepted that the discount rate was so intended: by the House in Wells, by Parliament in the Act of 1996, and by the Lord Chancellor in making his order under the Act. Mr Hogg's attempt to treat his calculation of the multiplicand as a "separate issue" from the discount rate, and counsel's submissions supporting that position, are in the end nothing but smoke and mirrors. It follows that the substance of these appeals constitutes an illegitimate assault on the Lord Chancellor's discount rate, and on the efficacy of the 1996 Act itself, and (subject to Mr Hogarth's point on s.1(2)) they must in my judgment be dismissed.”

Dyson LJ said this:-

“42.

The compensation principle is not in issue. It was acknowledged by their Lordships in Wells. They well understood that the cost of future care would or might exceed RPI inflation: see pages 354E and 367H. And yet they fixed a rate of 3 per cent, which was substantially based on a return for ILGS (where inflation is measured by the RPI). Their Lordships recognised that this would not produce a return that would match future inflation precisely. Thus, for example, Lord Steyn spoke of an element of "arbitrariness in any figure" (388D); and Lord Hope referred to there always remaining "an element of uncertainty in prediction which may only in a rough and ready way satisfy the desire that justice be done between both parties" (394G). Their Lordships recognised that a single rate was a somewhat crude instrument, but they adopted it for the public policy reasons that certainty was necessary in order to facilitate settlements and save costs.

43.

These same considerations informed the decision of the Lord Chancellor to select the single rate of 2.5 per cent. He too was aware of the compensation principle, and stated explicitly that this was the principle that he "must strive to apply". He could have chosen different rates for different heads of loss. But he decided not to do so in order to "eliminate scope for uncertainty and argument about the applicable rate".

44.

It is clear, therefore, that both the House of Lords in Wells and the Lord Chancellor in fixing the rate at 2.5 per cent purported to be giving effect to the compensation principle. The submissions made on behalf of the appellants amount to saying that neither the House of Lords nor the Lord Chancellor achieved what they expressly said they were setting out to achieve. If the present appeals were being decided without reference to the fact that the Lord Chancellor has fixed the discount rate, this challenge could not be made since the decision in Wells would be binding on this court. But, of course, the fact that the Lord Chancellor has fixed the rate cannot be ignored. Once it is shown that, in fixing the rate, the Lord Chancellor intended his single rate to allow for future inflation in respect of all heads of loss, an attempt to persuade a court to calculate damages by allowing for future inflation of certain heads of loss by a different method can be seen for what it is. Despite the disavowals vigorously made on behalf of all three appellants, these challenges are, in my view, nothing less than a plain attempt to subvert the Lord Chancellor's rate itself.”

Mr Havers, who argued this point before us, submitted that it should be impermissible to have two ostensibly parallel but in fact divergent systems of compensation producing different outcomes. If applying a discount calculated by reference to RPI was consistent with the 100% principle when calculating a lump sum award, it must also be consistent with the 100% recovery principle to make PPOs indexed by reference to RPI.

The answer to this point is provided in a passage in the judgment of Brooke LJ in Flora supported indeed by the reasons of the Lord Chancellor, as summarised by Dyson LJ in Warriner, quoted above. At paragraph 27, Brooke LJ, having traced the recent history of the discount rate, culminating, incidentally, with a quotation of a passage from the judgment of Laws LJ in Cooke, said this:-

“27.

This brief summary of the recent history of the discount rate used for the purpose of calculating lump sum awards for future pecuniary loss is sufficient to show that an award of a lump sum is entirely different in character from an award of periodical payments as a mechanism for compensating for such loss. When setting the appropriate discount rate in the context of a lump sum award the House of Lords or the Lord Chancellor had to guess the future and to hope that prudent investment policy would enable a seriously injured claimant to benefit fully from the award for the whole of the period for which it was designed to provide him/her with appropriate compensation.

28.

A periodical payments order is quite different. This risk is taken away from the claimant. The award will provide him or her year by year with appropriate compensation, and the use of an appropriate index will protect him/her from the effects of future inflation. If he or she dies early the defendants will benefit because payments will then cease. It is unnecessary in the context of this statutory scheme to make the kind of guesses that were needed in the context of setting a discount rate. The fact that these two quite different mechanisms now sit side by side in the same Act of Parliament does not in my judgment mean that the problems that infected the operation of the one should be allowed to infect the operation of the other. There is nothing in the statute to indicate that in implementing s.2 of the 1996 Act (as substituted) Parliament intended the courts to depart from what Lord Steyn described in Wells v Wells at pp 382H-383 B as the "100% principle", namely that a victim of a tort was entitled to be compensated as nearly as possible in full for all pecuniary losses: . . . .”

As Brooke LJ emphasised, the two regimes are quite different. We cannot put more eloquently the reasons for their difference than are given in the passage quoted.

In any event the argument that in some way Flora was decided per incuriam was hopeless. Cooke was cited and considered at length. Even if one cannot find the precise argument as Mr Havers put it being considered in the judgment of Brooke LJ, we would take some persuading that something very close to it was not being considered. In any event the point was certainly there to be taken and the decision was not per incuriam.

In Young v Bristol Aeroplane Co Ltd [1944] 1 KB 718, Lord Greene MR set out the circumstances where one Court of Appeal might not follow the decision of another including where the first decision was per incuriam. He did not purport to define with absolute precision the breadth of the expression per incuriam saying this at 729:-

“Where the court has construed a statute or a rule having the force of a statute its decision stands on the same footing as any other decision on a question of law, but where the court is satisfied that an earlier decision was given in ignorance of the terms of a statute, or a rule having the force of a statute, the position is very different. It cannot, in our opinion, be right to say that in such a case the court is entitled to disregard the statutory provision and is bound to follow a decision of its own, given when that position was not present to its mind. Cases of this description are examples of decisions given per incuriam. We do not think it would be right to say there may not be other cases of decisions given per incuriam in which this court might properly consider itself entitled not to follow an earlier decision of its own. Such cases would obviously be of the rarest occurrence and must be dealt with in accordance with their special facts. Two classes of decisions per incuriam fall outside the scope of our enquiry, namely, those where the court has acted in ignorance of a previous decision of its own or of a court of co-ordinate jurisdiction which covers the case before it – in such a case a subsequent court must decide which of the two decisions it ought to follow; and those where it has acted in ignorance of a decision of the House of Lords which covers the point – in such a case a subsequent case is bound by the decision of the House of Lords.”

That passage from Lord Greene’s judgment hardly encourages an argument that a previous decision is per incuriam if (as Mr Havers would suggest) an argument readily open to one party was not developed before the court suggested to have acted per incuriam. That that is the correct interpretation of Lord Greene’s judgment, and that the argument is hopeless, is confirmed by the decision of the Court of Appeal in Bryers v Canadian Pacific Steamships Ltd [1957] 1 QB 134. All members of the court were clear that they were bound by a previous decision of the Court of Appeal and Singleton LJ put the matter this way at page 147:-

“It has been argued in this court that the decision is not binding upon us because the point now taken . . . was not argued before the court in [the previous decision]. As I said, the point was there if it was to be taken. The circumstances of the case were considered: Regulation 42 was applied by the Court of Appeal in a case in which there were different repairers. The case ought to be regarded as binding upon the court in this case. It is no part of the duty of the court to look for a reason for not following a decision if the decision, on the face of it, covers the particular case. If one applies the test laid down by Lord Greene MR in Young v Bristol Airplane Co Ltd one finds (as I think) every warrant for the proposition made by Miss Heilbron that the decision in [the previous Court of Appeal decision] governs this case.”

In the end, Mr Havers frankly accepted that he was unlikely to succeed in persuading us that Flora was decided per incuriam, but asked us to deal with the argument anyway, in case the House of Lords had to consider the whole matter hereafter. That we have done.

This issue requires us to consider:

whether as a matter of law section 2(9)(b) of the Damages Act 1996 and the words ‘modifying the effect of’ can encompass and permit the deletion of the Retail Prices Index and/or its substitution by a measure based on Annual Earnings and converted to an index.

The appellants seek to persuade us that the only form of modification permitted by the subsection is modification of RPI and to reject any possibility of an alternative to RPI. It is said that ‘modification’ under the section relates to the ‘index’ and thus must be limited to increasing or decreasing the RPI, as for example by ordering indexation by reference to say, RPI plus 10%. It is said that Flora did not deal with how RPI could be modified; it looked only at when. In other words (as we would understand the argument), in Flora, the Court of Appeal must be taken to have decided that, when the matter returned to the trial judge, the trial judge was bound to rule that the index must remain the RPI, with the possibility of an uplift to the RPI, and that no consideration could be given to some alternative index or measure.

It may be that a judge would be free to modify the RPI index if that was the most accurate way to ensure 100% compensation, but it seems to us quite clear from the judgment in Flora that the Court of Appeal was not saying that that was the limit of the judge’s powers. Dr Wass was advocating alternatives to RPI and not a variation to RPI itself. The Court of Appeal was dealing with the possibility that the alternatives being advocated in the evidence of Dr Wass could be adopted and was encouraging the trial judge to consider those alternatives.

In our view we are bound by the Court of Appeal decision in Flora to reject this argument. But we add that it seems to us that the appellants’ argument is entirely misplaced. In their Consolidated Skeleton, those representing the appellants have demonstrated great industry in quoting many dictionary definitions of modification, many authorities on modification and statutes dealing with modification. But they also demonstrate the trap into which they have fallen by stating, in paragraph 47, that it is their case “that any modification must be by an adjustment to the index”. What section 2(9) actually says is that an order for PPOs may include a provision “modifying the effect of subsection (8)”. It does not confine itself to “modifying the index”. The language clearly includes applying another measure for indexation.

This issues asks:

whether as a matter of law or alternatively discretion, the remedy of modification should be denied by application of the principle of Distributive Justice.

Here, the appellants seek to persuade us once again that, save in exceptional circumstances, section 2(9) should not be used to disapply RPI. Reliance is placed on the principle of distributive justice, and it suggested that this argument was not considered in Flora.

Mr Havers reminded us of some well known dicta where, in recent years, reference has been made to the concept of distributive justice. He cited from Frost v Chief Constable of Yorkshire [1999] 2 AC 455, McFarlane v Tayside Health Board [2000] 2 AC 59, Rees v Darlington Memorial NHS Trust [2004] 1 AC 309 and Heil v Rankin [2001] QB 272.

Mr Havers’ difficulty is that ‘distributive justice’ is not a principle of English law recently adopted so as to allow free rein to ignore basic principles long established. It may come into play when considering whether it is fair, just and reasonable to hold that a duty of care is owed (as in Frost and Rees) or in considering a public policy question such as damages for the birth of a healthy child (as in McFarlane). It is perhaps also understandable how it plays some part in considering the essentially judgemental question of whether the level of general damages should be increased (as in Heil), but this is all a far cry from seeking to influence the calculation of actual financial loss where the 100% recovery principle is fundamental. Once liability is established and once financial loss is being assessed, it is ‘corrective justice’ and not distributive justice with which the court should be concerned.

In our view Brooke LJ was reflecting that thought when in Flora he said at paragraph 29:-

“For the same reason I reject the argument that the court should consider questions of affordability when determining what order to make because, as Lord Steyn said in Wells v Wells, . . . policy arguments based on affordability are a matter for Parliament and not for the court. It is true that in Heil v Rankin . . . this court took into account questions of affordability when determining what amount for general damages, for pain, suffering and loss of amenity, the public would perceive as fair, reasonable and just. There is no material, however, on which a court could safely rely in deciding whether the public perceive it to be fair, reasonable and just for compensation for future pecuniary losses to be reduced simply on affordability grounds. It would have been easy for Parliament to decree that this should be so (and to be willing to incur the accompanying political odium for doing so) but there is no evidence in the language of s.2 of the 1996 Act that this was Parliament’s intention.”

Indeed that passage seems to indicate that something akin to the concept of distributive justice was being canvassed in Flora and, in our view, it is inconsistent with Flora even to contemplate an argument which seeks to invoke distributive justice.

This issue requires the court to consider :

whether as a matter of law a party when seeking to trigger a statutory proviso under section 2(9)(b) of the Damages Act 1996 must discharge a legal burden by identifying and proving that a specified alternative on its own merits can displace the presumed index of the Retail Prices or whether the exercise under Section 2(9)(b) is a radical one of a quasi inquisitorial review undertaken by the Court as part of an extended approval function.

The appellants’ argument here was that there is a legal burden of proof on a claimant who seeks to persuade the court to modify the effect of section 2(8). There were various aspects to the argument. First it was said that it was not permissible for a claimant to call expert evidence which put a choice of alternatives to the RPI index (as happened before Swift J, Lloyd-Jones J and Mackay J). It was said that a claimant must make a choice and seek to establish that choice. Since, before HH Judge Bullimore, the claimant was putting the choices as alternatives, and since before us that was likely to be the respondents’ position, Mr Rees did not really pursue this point.

In our view, however, the point was always bad. The judge is bound by section 2 to consider whether to make a PPO. There is thus an inquisitorial role for the court (albeit not as part of an approval process) and it is helpful to the court to have choices with the good and bad points exposed in a balanced way. If anything, it adds weight to the expert evidence of Dr Wass that she was prepared to give evidence about the strengths and weaknesses of the choices put before the court and then leave the decision to the court.

Second, it was submitted that there was a burden on the claimant to show on the balance of probabilities that there was an appropriate alternative to the RPI and, if that could not be established on the balance of probabilities, then the RPI must apply. The extension of that argument was that, because the court has to consider whether a PPO should be ordered, the court has to have in mind the legal burden of being satisfied on the balance of probabilities that there is an appropriate alternative to RPI.

This point interlinks with the third aspect of the argument which is whether the judge has to take a ‘stand alone’ decision in relation to the alternative index proposed rather than a carrying out a comparative exercise.

In our view, the exercise to be carried out under section 2 should not be complicated by considerations of legal burdens. A party who seeks to prove a fact will have the evidential burden of establishing that fact, and the balance of probabilities is the standard of proof in all civil cases. But if it is suggested that a claimant always has a burden to discharge when the court is considering whether to make a PPO, that suggestion must be rejected. The court itself has to consider whether it is appropriate. Once embarked on that exercise and on whether a PPO will best meet the claimant’s needs, the question of indexation will have to be considered. The question whether RPI should be replaced will depend on the alternatives available and is bound to be a comparative exercise. Lloyd Jones J in Sarwar encapsulates both what we understand to be the argument and what, in our view, is the answer in the following paragraphs:-

“114.

On behalf of the Claimant it was submitted that the exercise with which we are concerned is one of finding the best or "least worst" match so that as far as possible any periodical payments are adjusted to take account of the future effects of inflation.  This was vigorously contested by Mr. Methuen on behalf of the Second Defendant.  He submitted that before the court could modify the application of subsection 2(8) by adopting an index or measure other that RPI, it would need to be satisfied that that index or measure truly fits the bill.  He submitted that the court should not approach this issue on the basis of a comparison of the various indices or measures available.  In his submission the fact that an index might be more appropriate than RPI in the circumstances of this particular case would not of itself justify the displacement of RPI.  Accordingly, he submitted that while an alternative index or measure need not be a perfect match it must be a very close match before it could be adopted.

115.

A consideration of the suitability of an index or measure will inevitably involve a comparison between that index or measure and the RPI which it is intended it should replace and a consideration of the extent to which the use of each is likely to assist in achieving the objective I have identified.  It will also necessarily involve a comparison of potential replacement indices or measures. To my mind this is an essential part of the consideration of the appropriateness and fairness of the substitution of an alternative index or measure described by Brooke L.J. in his judgment in Flora v Wakom. I note that this process of comparison is the basis on which all the experts have approached the issue. I would however accept that in the absence of a relatively high degree of assurance that a given index or measure is likely to achieve that objective, the court is unlikely to adopt it.”

His Honour Judge Bullimore took the view that the exercise is bound to be a comparative one but that any alternative that was unsuitable would have to be rejected. That must be right, and before Mackay J the criteria for suitability were hardly an issue. He identified them in the following words:-

“70.

Before considering individual measures proposed I should consider the criteria that should be applied when making what I consider to be a comparative assessment as to whether each meets the test of fairness of appropriateness defined above.

71.

The experts helpfully agreed the criteria for the suitability of an index as being:-

i)

accuracy of match of the particular data series to the loss or expenditure being compensated;

ii)

authority of the collector of the data;

iii)

statistical reliability;

iv)

accessibility;

v)

consistency over time;

vi)

reproducibility in the future;

vii)

simplicity and consistency in application

This appears to me an entirely appropriate and sensible list of the qualities which are to be looked for. Mr Hall sought to add that the candidate measure should be “free of distorting factors”. Dr Wass, more realistically in my view, said that that is in effect asking for the impossible though it should be as free as possible.”

Mackay J then considered each of the alternatives by reference to the identified criteria and ultimately, by reference to ASHE 6115, expressed his answer in these words:-

“In my view 6115 is markedly superior to RPI and I should modify the Act so as to order its application to the Claimant’s future care damages, unless any of the other arguments against its application, which I deal with below, make it inappropriate or unfair to do so.”

We do not see any great distinction between the approaches of the judges as reflected in the above quotations. The court is seeking to provide an answer which, on the information it has at the trial, will, through the use of a PPO, best provide the claimant with 100% compensation. If, in the context of future care, of which the main element is the wages of the carers, the RPI is not suitable for the purpose of tracking wage inflation, the question is whether a more suitable index or measure is available. Suitability should be tested against the criteria set out by Mackay J quoted above. If an alternative is more suitable, it must be open to the court to accept that alternative even if some criticisms can be made of it. If the alternative is less suitable than RPI it obviously could not be chosen. But this is not a ‘stand alone’ exercise under which the court would have to disqualify an alternative because of criticisms of its suitability even though the alternative was more suitable than RPI.

This issue requires the court to consider :

whether as a matter of law any measure which operates so as to rewrite the multiplicand when applying the element of pay drift in a reported earnings measure converted to an index contravenes the requirement of Cookson v Knowles and in fact rather than a multiplicand fixed at the date of trial which is indexed the annual figure forms the base of what then operates as an infinitely variable periodical payment.

The formulation of this issue purports to raise the question of ‘pay drift’. However, in our view it does not. We will deal with pay drift later in this judgment when we are considering the arguments relating to the suitability of ASHE 6115 as an index for the tracking of care and case management costs. The appellants’ argument under this issue is that to use an index or measure such as ASHE 6115 (or indeed any index which is based on actual earnings in the market) contravenes the principle on which future losses should be assessed as set out in Cookson v Knowles [1979] AC 556. That principle is now so well known that it hardly requires explanation. It is that in calculating future loss (on a lump sum basis of course which was the only method possible at the time) the court should take as a multiplicand the current annual loss as at the date of trial and should apply the appropriate multiplier. No attempt should be made to increase the multiplicand to take account of future inflation. Inflation is to be catered for through the selection of an appropriate multiplier. Although Cookson was a fatal accident case, the same principle is to be applied to the assessment of future loss in personal injury cases.

The appellants’ argument is that by using an index such as ASHE 6115, which will reflect the changes in earnings levels of a group of care workers, the court will in effect be altering the multiplicand annually and this will contravene the principle in Cookson. In our view, this submission is misconceived. As we have said, Cookson was concerned with lump sum awards. In the present case, the court is concerned with a wholly different creature, a PPO. When making a PPO under section 2 of the Damages Act, there is neither a multiplicand nor a multiplier. Parliament has decreed that there should be an annual sum, which is to be indexed every year, either by reference to the RPI under section 2(8) or as modified under section 2(9), for so long as the claimant lives or until he reaches a particular age. No question can arise of any contravention of the principle in Cookson.

In Flora the defendant sought to strike out the whole of the claimant’s case as to wages-related periodical payments. We have already explained how the objections to any form of wages-related indexation as a matter of law or principle failed in Flora; and fail in this case not only as a matter of precedent, but also because this court, having heard the objections rehearsed and supplemented before it, is as little persuaded by them as was its predecessors in Flora. The defendants however had and have objections to the cases advanced in Flora and in the present claims on a more particular level. The claimant in Flora appears to have sought indexation based on the AEI and in our cases they contended for ASHE 6115. The defendants’ case was that ASHE 6115 was open to so many objections that it could not properly be used in any event, even if in principle some wages-related index or assessment should be available to the claimants; and so, since other wages-related indices were even more objectionable than ASHE 6115, and were not supported by the claimants, the success that the claimant in Flora had achieved in avoiding in principle the imposition of indexation by the RPI beat the air, since there was in practice no index that could be used to implement that principle.

This issue, of the appropriateness and usability of ASHE 6115, was left by this court in Flora to the judgement of trial judges in forthcoming cases. That was the enquiry conducted in the judgments under appeal before us, and in most detail in the judgments in Thompstone and RH. The parties were fortunate that that enquiry fell to be conducted by judges of the distinguished experience in this area of work, not only on the bench but also at the bar, that has been enjoyed by Swift J and Mackay J.

This part of the appeal was expressed before us in the terms of issues 7, 8 and 9. However, before addressing the specific objections raised by the appellants it is necessary to revert to the issues of the nature of the enquiry and the burden of proof within it. Those questions were debated under the head of issue 5, but the correct answer to them is of importance when approaching issues 7- 9.

The appellants repeated, in relation to the assessment of particular indices, the argument that they had adduced in relation to section 2(9) as applied to the prior question of whether RPI should be displaced at all. Mr Rees argued that the effect of section 2(9) was to create a proviso to the primary rule applying the RPI. A party who relies on a proviso bears the burden of establishing that the proviso applies to the instant case; so it was for the claimants, who wanted ASHE 6115 to be applied, to show that ASHE 6115 met the requirements, whatever they were, of a properly useable index. If on any issue they had failed to discharge that legal burden of proof, to the standard of balance of probabilities, then ASHE 6115 could not be applied by the court.

This is misconceived because it misunderstands the nature of the exercise on which the court is engaged. Mr Rees sought to cite in support of it cases on the proof of fact, such as, conspicuously, The Popi M, in which courts have said that a tribunal must always apply the rules as to burden and standard of proof, and if those rules do not yield a solution then the matter must remain unproved. Mr Rees, as we understood him, submitted that in respect of various elements relevant to ASHE 6115 the case remained unproved in that sense, and so it had not been properly open to the judges below to accept the validity of ASHE 6115. But the issues that arise in relation to ASHE 6115 are not pure issues of fact, and do not lend themselves to this treatment. To take one example, discussed further below, the question of whether ASHE 6115 is too volatile to be properly applied to a long period in the future. Certainly the basic facts have to be established; but then the answer to the question of whether the index is inappropriately volatile is a matter of judgement and appreciation for the court. In that exercise, the judge, unlike the trial judge addressing the competing theories as to factual events in the Popi M, does have to make up his mind in one direction or the other. He cannot resolve the matter simply by placing a burden on one party and then finding that that burden has not been discharged.

We cannot express these fundamental propositions better than they were put by Swift J in paragraph 52 of her judgment in Thompstone:

“My task is to decide what form of order will best meet the claimant’s needs and, so far as section 2(8) and (9) is concerned, to determine what is appropriate, fair and reasonable. These matters do not lend themselves to determination by the burden of proof. Insofar as the claimant does bear any burden, it seems to me that this is an evidential burden, i.e. an obligation to adduce evidence sufficient to establish a case that the RPI is an inappropriate measure of indexation and there is at least one alternative, more appropriate, measure that the court might adopt in its stead.”

That statement is subject to one qualification in its application to this part of the case. Mr Rees said that the issue was not simply whether ASHE 6115 was preferable to the RPI as an instrument of indexation. That was because, in the light of the analysis in Flora, it was difficult to gainsay a proposition that some wages-related index would serve the fundamental objective of 100% compensation better than does indexation by the RPI. Rather, he submitted, the index supported by the claimants must be shown to be reliable and supportable in its own right. We accept that contention to the limited extent that, if an index is seen to be statistically unreliable or unlikely to be reproducible in future, it will not be a suitable alternative to RPI. But, given that a particular index might be ruled out for that kind of reason, those which pass that initial test must then be subject to a comparative exercise to determine which is the most suitable. But the enquiry still remains, as Swift J said, a task of evaluation for the court.

We therefore revert, issue by issue, to the enquiry into ASHE 6115. In doing so we say now, for the avoidance of doubt, that in respect of every issue we are satisfied that, even if there had been applied to it the approach urged by Mr Rees, the outcome would have been the same.

ASHE is an annual earnings survey published by ONS by means of a sample survey of the actual earnings of employees. It is expressed as a statement of earnings levels, rather than as an index. In addition, ONS publishes disaggregated data for the various Standard Occupational Group classifications, which classifications are reviewed every ten years to take account of changes in the composition of the workforce. The relevant classification for our case, which is found in ASHE 6115, is care assistants and home carers who, in the words of the SOG classification

“Assist[s] residents to dress, undress, wash, and bathe; serve[s] meals to residents at tables or in bed; accompany residents on outings and assist recreational activities; undertake light cleaning and domestic duties as required.”

Since ASHE 6115 is not in itself an index, a further step was necessary before it could be used in the indexation of the claimants’ care needs. Dr Wass, the labour economist instructed by the claimants, proposed that, starting from the care package in terms of number and types of carer agreed by the care experts, there should be calculated a weighted average of the claimant’s hourly current cost of care. For example, as described by Swift J in paragraph 108 of her judgment, that exercise on the facts of Thompstone yielded a figure of £8.50 per hour. There was then identified where on the list of incomes arranged in order of value identified by ASHE 6115 (as it was put, at what percentile of the total list) that figure of £8.50 appeared. Whatever figure thereafter appeared at that percentile of ASHE 6115 was then to be used for the indexation of the original figure of £8.50 per hour. By that means ASHE 6115, though not itself strictly speaking an index, could be used for indexation purposes.

The appellants expressed this part of their case by issues 7- 9 as follows:

7.

Whether as a matter of law a measure which is not an index and which can only be made to function as an index by means of a notionally derived “weighted average rate” can operate in substitution for the presumed Retail Prices Index

8.

Whether as a matter of law irrespective of the answer to Issues 5-7 above, there is any sufficient or cogent evidence that the measure provided by ONS (ASHE 6115 at the relevant percentile) bears any sufficiently proximate relationship to the facts of the care and case management multiplicands

9.

Whether as a matter of law irrespective of the answer to Issues 5-7 above, the features associated with ASHE 6115 at the relevant percentile, namely reclassification, movement on the distribution, volatility, and workability are individually or in combination sufficient to disqualify the use of this measure.

Two preliminary points must be made. First, the appellants express these issues as matters of law. If that contention were to be taken seriously, it fails in limine. That is because the only issue of law as to the content or nature of ASHE 6115 is whether ASHE 6115 can ever be even considered as a measure of indexation. That issue was decided against the appellants in Flora. It is of course not difficult to see why the appellants present what are issues of assessment and judgement as matters of law. On all of the issues there are findings against them of the trial judges. Those findings in themselves have not been appealed, nor has it been argued, nor could it have been, that the judges who heard the four cases acted irrationally, perversely or in disregard of relevant evidence. The exercise urged on us by the appellants is therefore highly artificial, in that it is in effect a complete rehearing of issues already decided by the trial judges. With no little hesitation we nonetheless embark on it, not least because it is of general importance that there should be a definitive statement on these issues at appellate level.

Second, although the statement of issues 7- 9, set out above, makes reference to most of the complaints that the appellants seek to raise, it became clear during argument that there were other objections, either assumed in other parts of the list of issues, or not mentioned at all, that the appellants wished to raise. In the same spirit as inspires paragraph 73 above we seek to address the whole of the appellants’ case.

In paragraph 71 of his judgment in RH, Mackay J reported the criteria for the suitability of an index that had been agreed by the experts in that case. No-one before us sought to gainsay that guidance, the terms of which have already been set out in paragraph 56 above but which it is convenient to repeat here as the context of the enquiry:

This is a question, in the terms of issue (i) set out in paragraph 75, of accuracy of match. It will be noted that in issue 8 this question is stated in terms of whether there is “sufficient or cogent evidence” of a sufficiently proximate relationship. That harks back to the appellants’ contentions as to burden of proof that are discussed in paragraphs 65-67 above. The short answer to a question posed in those terms is that there is indeed evidence, and the trial judges have found it to be sufficient and cogent. But in the spirit indicated above we go further and investigate the substance of the complaint.

The complaint has two, rather different, limbs. First, the multiplicand to which ASHE 6115 is to be applied does not contain exclusively care costs, so an index based on the wages element in care will cause distortion. Second, ASHE 6115 does not sufficiently accurately target the particular carers in issue in this case. There is a certain irony in these objections being raised by a party who urges the adoption of the RPI, which contains almost no element that relates to care costs, either of the present claimants or generally; but we let that pass.

First, the composition of the multiplicand. This issue was pursued in some detail before Swift J, who made a finding of fact, at paragraph 107 of her judgment in Thompstone, that the non-earnings-related elements of the annual multiplicands amount to no more that about 4-5% of the total care costs to age 19 and 2-3% thereafter. That finding was not appealed. That is the end of this point.

Second, the carers included in ASHE 6115. There were various issues, not always very clearly distinguished from each other, that were pursued under this head. These complaints were based on the undoubted fact that ASHE 6115 is a general survey, not specifically related to the carers who will be looking after the claimants. From that it was said to follow that it could not be assumed, and certainly had not been proved, that the wages of those workers would move in the same way as the wage movements reflected in successive editions of ASHE 6115; and it could not be assumed, and certainly could not be proved, that home carers were not in a discrete employment market, to which the sort of general trends reflected in ASHE (or in any other index) would not apply. Mr Rees pointed out that, on the other hand, its national generality was not a relevant criticism in this context of the RPI, because indexation with reference to the RPI did not even purport to reflect the wages paid to the claimant’s carers. Its status was as a general, in the present context effectively abstract, index, but approved for that purpose by Parliament.

We have to say that the unreasonableness of the conclusions to which these arguments led the appellants amply demonstrates why they saw their main hope of success as lying in arguments about the burden of proof. The very nature of the use of an index means that one cannot prove, in the way in which the law requires facts to be proved, that a particular situation is specifically represented in the index. But that does not disqualify the index. All that has to be established is that the composition of the index is sufficiently close to the subject-area in which it is to be used to render that use reasonable and representative.

So viewed, the complaints made by the appellants are counter-intuitive, and cannot withstand the application of the commonsense that the judges below brought to them. As Swift J put it at paragraph 139 of her judgment in Thompstone:

“I do not accept the Defendants’ contention that the absence of a measure specific to the local labour market with which this case is concerned must necessarily be fatal to the Claimants’ case on indexation. Provided that I can be satisfied that there is an alternative measure which would provide a reliable indicator of growth in the earnings of carers such as those whom the Claimant will employ, it seems to me that that will suffice.”

These findings were fatal at trial to the suggestion at trial, and it was no more than that, of a mis-match between ASHE 6115 and the wages-cost of the claimants’ care; and they are equally fatal in this court.

How the weighted average was used in order to apply ASHE 6115 has been described in paragraph 71 above. That process is put in issue by the appellants’ issue 7. The appellants did not criticise the calculations in themselves but, as explained by Swift J in paragraphs 108-110 of Thompstone, objected to any use of a weighted average, as an artificial construct, not a real earnings rate, and impossible to calculate with absolute accuracy.

These are completely irrelevant objections, and were treated as such in the courts below. One of the leading requirements of an index, as agreed in the extract from RH that is set out in paragraph 75 above, is that it should combine simplicity in application with accuracy of match. To take a weighted average of the wages of the people making up the agreed care package, rather than (as seems to be suggested) assessing each worker individually, is an obviously sensible way of achieving those two objectives. The complaint that the result does not set out the position of any single individual only demonstrates a misunderstanding of the nature and purpose of indexation, as did the complaint that ASHE 6115 does not specifically set out the wage-rates of domiciliary carers. The whole point of an index is that it enables general trends to be applied to a particular situation that is sufficiently accurately related to the matter on which the index is based. And as to accuracy, it was for the defendants to produce their own calculations if they were able to do so.

Swift J summarised the position thus in paragraph 47 of Thompstone:

“I am satisfied that the weighted average hourly rate, as calculated by Dr Wass, provides a fair and reasonable estimate of the average wage to be paid to the Claimant’s carers. It is based on the pay rates agreed by the care experts. Any imprecision is likely to be negligible. I find that it is appropriate to use it in order to match the carers’ earnings level to the appropriate ASHE 6115 percentage.”