No doubt it was this passage which the Court of Appeal had in mind when they said that it was necessary "to hold the balance evenly between both sides." I have to say that I do not find Lord Scarman's reasoning persuasive. If the object of an award of damages is to put the plaintiff in the same position as he would have been in if he had not been injured by the negligence of the defendant (as was common ground) then one ought, in principle, to get as near as one can to the wages which he would actually have earned but for the injury and the cost of the needs which he will actually incur. In other words, one ought so far as possible to take account of inflation, as Lord Reid had said.

      What then did Lord Scarman mean by saying that this would put the plaintiff in a privileged position in comparison with others who have to rely on capital for their support? Once the lump sum has been calculated and paid, he is in exactly the same position as others, such as those who have saved or inherited a lump sum. But in calculating the sum his position is in no way comparable. For the plaintiff is entitled to be protected against future inflation at the expense of the tortfeasor; otherwise he does not receive full compensation. The others are not so entitled. It is only in that sense that the plaintiff is in a privileged position. I cannot for my part see anything unjust in requiring the defendant to compensate the plaintiff in full, however burdensome that may prove. Lord Scarman recognised this himself when he said, at p. 187:

 "There is no room here for considering the consequences of a high award upon the wrongdoer or those who finance him. And, if there were room for any such consideration, upon what principle, or by what criterion, is the judge to determine the extent to which he is to diminish upon this ground the compensation payable?"

Conclusion

      My conclusion is that the judges in these three cases were right to assume for the purpose of their calculations that the plaintiffs would invest their damages in I.L.G.S. for the following reasons:

      (1)  Investment in I.L.G.S. is the most accurate way of calculating the present value of the loss which the plaintiffs will actually suffer in real terms.

      (2)  Although this will result in a heavier burden on these defendants, and, if the principle is applied across the board, on the insurance industry in general, I can see nothing unjust. It is true that insurance premiums may have been fixed on the basis of the 4 to 5 per cent. discount rate indicated in Cookson v. Knowles and the earlier authorities. But this was only because there was then no better way of allowing for future inflation. The objective was always the same. No doubt insurance premiums will have to increase in order to take account of the new lower rate of discount. Whether this is something which the country can afford is not a subject on which your Lordships were addressed. So we are not in a position to form any view as to the wider consequences.

      (3)  The search for a prudent investment will always depend on the circumstances of the particular investor. Some are able to take a measure of risk, others are not. For a plaintiff who is not in a position to take risks, and who wishes to protect himself against inflation in the short term of up to 10 years, it is clearly prudent to invest in I.L.G.S. It cannot therefore be assumed that he will invest in equities and gilts. Still less is it his duty to invest in equities and gilts in order to mitigate his loss.

      (4)  Logically the same applies to a plaintiff investing for the long term. In any event it is desirable to have a single rate applying across the board, in order to facilitate settlements and to save the expense of expert evidence at the trial. I take this view even though it is open to the Lord Chancellor under section 1(3) of the Damages Act to prescribe different rates of return for different classes of case. Mr. Leighton Williams conceded that it is not desirable in practice to distinguish between different classes of plaintiff when assessing the multiplier.

      (5)  How the plaintiff, or the majority of plaintiffs, in fact invest their money is irrelevant. The research carried out by the Law Commission does not suggest that the majority of plaintiffs in fact invest in equities and gilts, but rather in a building society or a bank deposit.

      (6)  There was no agreement between the parties as to how much greater, if at all, the return on equities is likely to be in the short or long term. But it is at least clear that an investment in I.L.G.S. will save up to 1 per cent. per annum by obviating the need for continuing investment advice.

      (7)  The practice of the Court of Protection when investing for the long term affords little guidance. In any event the policy may change when lump sums are calculated at a lower rate of return.

      (8)  The views of the Ogden Working Party, the Law Commission and the author of Kemp and Kemp in favour of an investment in I.L.G.S. are entitled to great weight.

      (9)  There is nothing in the previous decisions of the House which inhibits a new approach. It is therefore unnecessary to have resort to the 1966 Practice Statement.

Consequences

      Once it is accepted that the lump sum should be calculated on the basis of the rate of return available on I.L.G.S., then an assessment of the average rate of return at the relevant date presents no problem. The rates are published daily in the Financial Times. A table of average rates for the period June 1990 to December 1994 is included in Kemp and Kemp at para. 8-068. No doubt the table will be brought up to date from time to time.

      The average gross redemption yield in June 1995 when Mr. Prevett gave his evidence was 3.78 per cent. If one takes the average over the previous six months it was 3.8 per cent., and if over the previous 12 months it was 3.83 per cent. The equivalent figures for November and December 1995 when Collins J. and Dyson J. gave judgment were marginally lower at 3.53 per cent. and 3.52 per cent. There must then be a deduction for tax on income. In his valuable appendix to the judgment below, Thorpe L.J. scorns the assumption of a 25 per cent. flat rate of tax as "crude, unrealistic and favourable to plaintiffs." I agree. In the first place it ignores the impact of allowances and tax bands. Secondly, it assumes a constant rate of income throughout the period to be covered, whereas in reality the income element in the annual draw-down will reduce and the tax-free capital element will increase as time goes by. The Duxbury Tables attempt a much more accurate calculation of the incidence of tax. Figures put before us show that on a fund of £1 million invested to produce 3 per cent. over 20 years the actual incidence of tax would be no more that 15.37 per cent.

      It is not altogether clear how Judge Wilcox arrived at his 2.5 per cent. as the appropriate discount on an average gross return of 3.8 per cent. If he deducted tax at 25 per cent., he would have arrived at 2.8 per cent. net, not 2.5 per cent. But for reasons already mentioned 25 per cent. is certainly too high, quite apart from the fact that it is no longer the standard rate of tax. Judge Wilcox's figure of 2.5 per cent. cannot stand.

      In its place I would substitute 3 per cent., which is the net discount rate adopted by Collins J. and Dyson J., representing a deduction of 14 per cent. for the impact of taxation on a gross return of 3.5 per cent. This sounds about right. I appreciate that such an approach is less precise than what is available by using the Duxbury Tables, which was Thorpe L.J.'s preferred approach. On the other hand it is important to keep the calculations simple as well as accurate, as Thorpe L.J. was the first to recognise. So far as the three appeals currently before the House are concerned I would regard 3 per cent. as the appropriate net return. It follows that the award in Wells v. Wells will have to be recalculated on that basis.

Guidelines

Section 1 of the Damages Act 1996 provides:

 "(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."

      The section came into force on 24 September 1996, but no rate has yet been prescribed. Lord Mackay of Clashfern, the previous Lord Chancellor, was said to be awaiting the decision of the Court of Appeal in the instant cases. It goes without saying that the sooner the Lord Chancellor sets the rate the better. The present uncertainty does not make the settling of claims any easier.

      In the meantime it is for your Lordships to set guidelines to replace the old 4 to 5 per cent. bracket. There is something to be said for a bracket, since it allows some flexibility in exceptional cases, as where, for example, the impact of higher-rate tax would result in substantial under-compensation. Thus on an award of £2 million higher rate tax payable over the first half of a 20-year period would alone amount to nearly £75,000. But the majority of your Lordships prefer a single figure. I do not disagree provided it is subject to the same flexibility as is to be found in section 1(2) of the Damages Act.

      What then should the figure be? The average gross redemption yield on I.L.G.S. has fallen steadily over the last year. In May 1997 it was 3.68 per cent., by May 1998 it was only 2.8 per cent. Less tax at, say, 15 per cent., this would give a net return of 2.38 per cent. Logically, therefore, we should take 2.5 per cent. as the guideline figure, since the assumption is that the plaintiff will purchase in the market at that price. The higher-yielding stock is no longer available. If therefore the calculation is done at 3 per cent. instead of 2.5 per cent., he would be substantially under-compensated.

      But since it is undesirable that the guidelines should be changed too often, it may be better that the average gross return should be ascertained over a period of months rather than on a particular day; and since, as I have said, the average return has been falling over the last year, one would expect the average return over that period to be higher than the current return. Such proves to be the case. Over the last six months and 12 months to March 1998 the average return has been 3.02 per cent. and 3.28 per cent. respectively. These figures justify a guideline rate of return of 3 per cent. net rather than 2.5 per cent., and this is the rate which I would propose for general use until the Lord Chancellor has specified a new rate under section 1 of the Damages Act.

      I would not, however, accept that the average should be taken over as long a period as three years. For if the rate of return had been falling steadily over the whole period (in fact this has not been the case) it would work very unfavourably to the plaintiffs; and vice versa if it had been rising steadily over three years. A year would seem to be the best compromise period. Once the net return has been established to the nearest 0.5 per cent., it is a simple enough matter to find the correct multiplier from the Ogden Tables.

Wells v. Wells

      I come now to the miscellaneous points. They do not call for extensive discussion, since they were argued only briefly. The underlying question is whether the defendants in each case succeeded in showing that damages awarded by the judges at first instance in respect of any particular head of damage (see George v. Pinnock [1973] 1 W.L.R. 118 per Sachs L.J. at 126) are outside the appropriate bracket (see Every v. Miles [1964] C.A. Trans. No. 261 per Diplock L.J.; Kemp and Kemp para. 19-006), or else represented a "wholly erroneous estimate", whether due to mistake of law or a misapprehension of the facts: see Pickett v. British Rail Engineering Ltd. [1980] A.C. 136 per Lord Wilberforce at 151, and Lord Scarman at 172.

      As already mentioned Judge Wilcox awarded Mrs. Wells £120,000 for pain and suffering. The Court of Appeal reduced the figure to £100,000. On behalf of the defendant it is said that the case falls at the upper end of the moderately severe brain-damage bracket as described in the Judicial Studies Board Guidelines 2nd ed. (1994) (£77,500 to £95,000), and not within the most severe brain-damage bracket (£105,000 to £125,000). Mrs. Wells is severely disabled, and will remain so for the rest of her life. But I agree with the Court of Appeal that the sum of £120,000 awarded by the judge falls well above the bracket for her type of case. The Court of Appeal was therefore right to intervene for the reasons set out convincingly in their judgment, which I need not repeat. As for Cunningham v. Camberwell Health Authority [1990] 2 Med.L.R. 49, on which the plaintiff relied, Mr. Leighton Williams pointed out that the plaintiff in that case was 12 years younger than Mrs. Wells, and it looks as though her injuries may have been rather more serious. In any event the award in that case may have been on the high side. I would uphold the Court of Appeal's figure of £100,000.

      On the other hand I am unable to agree with the Court of Appeal's reduction in Mrs. Wells's life expectancy from 15 years to 10 years. Dr. Peter Harvey's evidence was that her life expectancy is not affected by her present condition in view of the intensive care which she now receives. He would therefore predict a normal life expectancy for a woman of her age, namely, 20 years. Mr. Alan Richardson, on the other hand, predicted a life expectancy of 10 years, or 13 at the most. The judge expressed his conclusion as follows:

 "I prefer the considered and more closely reasoned approach of Mr. Alan Richardson and doing the best I can upon the medical evidence and all the other evidence I am persuaded that the life expectancy in this case more likely than not is 15 years."

      The Court of Appeal fastened on the judge's preference for Mr. Richardson's evidence, and held that he was therefore wrong to split the difference (if that is what he did) by taking 15 years. But this is to misunderstand the judge's approach. Even though he preferred Mr. Richardson's evidence, he was not obliged to accept the lower of his two figures uncritically. Nor was he obliged to reject Dr. Harvey's views out of hand. In arriving at a life expectancy of 15 years the judge took all the medical evidence into account, and all the other evidence besides. In my view the Court of Appeal should not have interfered. I would therefore restore the judge's figure of 15 years' life expectancy.

      The third point relates to the multiplicand. The Court of Appeal rightly rejected a number of Mr. Leighton Williams's criticisms of the judge's findings. But in one respect they found his criticism justified. At the time of the trial the care regime consisted of a residential carer and a day carer by day, and the residential carer and a night carer by night. Miss Teresa Gough, who gave evidence at the trial for the defendant, urged strongly that a night sleeper could be substituted for the night carer, with a saving of £10,539 per year. But the judge preferred the evidence of Mrs. Statham and Mrs. Clarke-Wilson, who had, as he put it, "the direct hands-on experience." If, as they said, Mrs. Wells needed to be seen three times a night, then Miss Gough accepted that that was a task for a night carer rather than a night sleeper.

      In my view the Court of Appeal was right to scrutinise the individual items which went to make up the multiplicand. Since the effect of reducing the rate of discount will be to increase the multiplier in every case, it is all the more important to keep firm control of the multiplicand. Plaintiffs are entitled to a reasonable standard of care to meet their requirements, but that is all. Having said that, however, and having heard all that Mr. Leighton Williams had to say, I am not persuaded that the Court of Appeal was entitled to substitute their own view of the evidence to that formed by the judge. They gave no reason other than it was another instance of the judge over-providing. I would therefore restore the judge's finding under this head.

Thomas v. Brighton Health Authority

      The agreed medical evidence was that the plaintiff has a life expectancy to the age of 60. Collins J. held, however, that he ought to reduce the arithmetical multiplier by about 20 per cent. "to cater for the hazards of life in such cases." In the result he took a multiplier of 23. The Court of Appeal agreed with the judge's approach but started from a different starting-point. With a 4.5 per cent. discount rate the arithmetical multiplier came to 20. Reduced by 15 per cent., rather than 20 per cent., they arrived at a multiplier of 17.