High Court of Australia Barwick, C.J. Gibbs, Stephen, Mason, Murphy, Aickin and Wilson JJ. Pennant Hills Restaurants Pty Ltd v Barrell Insurances Pty Ltd [1981] HCA 3 ORDER Appeal by Pennant Hills Restaurants Pty. Ltd. dismissed with costs. Cross-appeal by Barrell Insurances Pty. Ltd. allowed with costs. Appeal by Barrell Insurances Pty. Ltd. allowed with costs. Cross-appeal by Pennant Hills Restaurants Pty. Ltd. dismissed with costs. Order of the Supreme Court of New South Wales (Court of Appeal) entered on 5th March 1979 varied as follows: (a) Delete par. 2 of that order and substitute the following: 2. Verdict in the sum of one hundred and eighty seven thousand one hundred and forty two dollars sixty five cents ($187,142.65); (b) Delete par. 4 of that order and substitute the following: 4. In lieu of the interest which would otherwise be payable under s. 95 of the Supreme Court Act 1970, interest shall be payable at a rate equivalent to the prescribed rate on so much of the sum of $187,142.65 as was and is from time to time unpaid as from 16th December 1977 until payment. Cur. adv. vult. The following written reasons for judgment were delivered:— 1981, Feb. 10 Barwick C.J. The respondent in this case is to be compensated now by a lump sum against its statutory liability to pay workers' compensation in the future, under the Workers' Compensation Act, 1926 N.S.W., as amended. Had the respondent been insured by the due efforts of the appellant to which it was entitled, that liability would have fallen on an insurance company. The case is in its facts not one of a similar kind to the case dealt with in O'Brien v. McKean [1] (O'Brien's Case), i.e. it is not a case of the assessment of compensation for personal injuries. Rather, it is a case in which compensation is to be assessed for the liability to make future disbursements the amount of which is statutorily indexed to a designated index, which is variable according to future wage movements. In so far as that index has already operated, no particular problem arises. The question is whether the possibility of future changes in the index increasing the amount of the workers' compensation payable can be reflected in the assessment of the damages for which the appellant is liable. 1. (1968) 118 C.L.R. 540. When the amount which the respondent will be called upon in the future to pay by way of workers' compensation has been determined, a second question arises as to the manner in which a sum is to be presently assessed to cover the amount of the respondent's future liability. This can be regarded as involving two questions: first, whether, in assessing the amount of the damages payable, account is to be taken of the taxation of the income which may be earned by the investment of the amount of assessed damages which is referable to that future liability and, secondly, and subject to the answer to that question, what rate of discount should be employed in determining the amount now to be awarded to cover the amount of that future liability. I have had the advantage of reading the reasons prepared in this appeal by my brother Mason. I have no need to supplement his recital of the facts or of the curial history of the case. I agree with my brother's view as to the result of this Court's decision in O'Brien's Case and of the decision of the House of Lords in Lim Poh Choo v. Camden and Islington Area Health Authority [2] . I should add that I agree with my brother's conclusion that the amount allowed for domestic help did not constitute an amount paid or payable for "nursing" within s. 10 of the Workers' Compensation Act. 1. [1980] A.C. 174. I remain of the opinion I expressed in the former of the cases to which I have referred, namely, that the possibility or, for that matter, the certainty of future changes in the value of money ought not to be reflected in the assessment of damages to cover tortious breaches. I would also maintain my then expressed view as to the assessment of damages in respect of future disbursements, a view which left room for the inclusion in an assessment of damages for future expenditure of future changes in the amount of that expenditure, i.e. an amount which was proved to be certain or at least probable to be payable. I remain of the opinion then expressed that, to be reflected in the assessed damages, such changes in the amount to be expended must be established by evidentiary proof. I do not think that the fact since the decision in O'Brien's Case we have become more conscious of the operation of "inflation" or of its extent and its possible or even certain duration can make any difference to the principle on which I thought O'Brien's Case should be decided. Indeed, the fluctuation of the extent of the loss of value in money and the uncertainty of the duration of that fluctuation formed part of my own thinking when arriving at my expressed conclusion in that case. I see no reason to rethink the matter because of subsequent and present experience. Basic to any consideration of the principal question in this case is that it rests upon a plaintiff seeking damages, whether in tort or in contract, to prove his loss or damage, past or future, by admissible evidence. Mere speculation cannot satisfy what such a plaintiff must prove. It is perhaps unfortunate that a plaintiff, who by another's fault or breach has to sue to obtain a present sum to compensate for a loss, including the future liability to make payments, which has or have necessarily to be measured wholly or in part by the events of the future, may find it impossible to prove by admissible evidence the full extent of the loss or damage. This, however unfortunate, is but an inevitable consequence of being placed in the situation of a suitor at law. The situation should not be met by the substitution of speculation for admissible prolative evidence. The view I formerly expressed in O'Brien's Case [3] laid some weight on the fact that the successful plaintiff being in possession of a capital sum was able by his investment and perhaps by what for brevity I might call his "taxation policy" to buffer himself against the effect of changes in the value of money. It is said that an investor cannot now wholly protect his funds from loss of value due to change in the value of money. If this be so, and I am not convinced that it is necessarily so, the remedy cannot be in the augmentation of the assessed damages upon some basis not the subject of what I called in O'Brien's Case [4] "solid proof". If by a policy of investment the effect of inflation cannot be met, it is scarcely to be said that the courts by some means of their own devising can successfully overcome that effect. 1. (1968) 118 C.L.R. 540. 2. (1968) 118 C.L.R., at p. 550. It is perhaps not a very satisfying answer to say that damages are not in every case a perfect compensation but in many cases no more than an approximation lacking in mathematical or economic accuracy or sufficiency. But, however unsatisfying, that answer, in my opinion, must be accepted. That the payment of workers' compensation to be made in the future may be increased in amount because of upward wage movements may be a possibility: some might say a probability. But it is the amount of that increase, if it occurs, which must be established: it is the amount of the increase and not the possibility or probability that there will be some increase with which the assessment of the damages is concerned. That amount has not, in my opinion, now been proved. It can only be the subject of speculation. I would therefore not include in this case in the calculation of the future liability of the respondent for the statutorily indexed payments of workers' compensation any sum to represent the possible effects of the changes in the index in the future. The respondent has not established the amount of the anticipated increase. I turn then to the question whether the amount of tax which may become payable on the income of the investment of the sum awarded to represent future liability for workers' compensation ought to be taken into account in assessing the present sum to be awarded. The Court has decided not to depart from British Transport Commission v. Gourley [5] (Gourley's Case). But I do not regard that decision as carrying the conclusion that liability to taxation must always and in every case be taken into account when assessments of damages are being made. In that case, the receipt of emoluments by reference to which assessment was to be measured was assumed to be reduced in the hands of the recipient by an amount of tax which it was assumed would be established, if not otherwise, then by agreement. Here, clearly, we have no comparable situation. 1. [1956] A.C. 185. In this case, I do not think that the amount to be paid in tax on the investment of the whole or part of the assessed damages can be established to that degree of certainty which would warrant its recognition in the assessment of the damages themselves. To take a "base" rate and assume it to be payable on the amount of the investment income is not, in my opinion, acceptable. For one thing, the successful party may not invest in income producing assets. He may opt for capital gain. If investment does take place the amount of tax payable on the investment income may be affected by so many circumstances as to involve a view as to its amount no better than highly speculative and of a kind not satisfying the requirements of proof. I would therefore not attempt to reflect in the assessment of damages the amount of tax which might possibly become payable by reason of the investment of the whole or part of the amount to be assessed. There remains the question of the rate of discount to be applied to the sum of the future liability in order to determine the amount of damages presently assessed. I agree that the percentage figure currently to be obtained for investment of money ought not to be used for this purpose. Nor do I think that an endeavour should be made to work out what might be called a "real" rate of interest, i.e. an endeavour to determine the extent of the inflationary element in the current rates of interest obtainable. I favour adhering to an artificial rate of discount which, when compared with the current or "going" rate of interest leaves some room for the successful plaintiff to some extent to offset the effect of declining value in money. After all, the discount rate traditionally used has in truth always, or at least for a considerable time, been to a degree artificial. On balance, I would continue the use of 5 per cent as the appropriate rate of discount. In the result, I would allow the appeal and restore the verdict proposed by the Full Court. Gibbs J. I have had the advantage of reading the reasons for judgment prepared by my brother Mason. I agree with the conclusions which my brother Mason has reached, and subject to what I shall say hereafter am in general agreement with the reasons which he has expressed. The plaintiff in these proceedings, Pennant Hills Restaurants Pty. Ltd. ("Pennant Hills"), is entitled to damages to compensate it for the failure by the defendant, Barrell Insurances Pty. Ltd. ("Barrell"), to arrange insurance cover in respect of the liability of Pennant Hills under the Workers' Compensation Act, 1926 N.S.W., as amended ("the Act"). The learned trial judge found that this failure was due to the negligence of Barrell and further that it constituted a breach of contract between the parties. It seems to me immaterial whether damages are assessed in tort or in contract. In both cases the general principle is that "the injured party should receive compensation in a sum which, so far as money can do, will put him in the same position as he would have been in if the contract had been performed or the tort had not been committed":Butler v. Egg & Egg Pulp Marketing Board [6] . If the insurance had been effected, Pennant Hills would have been indemnified against its liability to the injured workman under the Act. Because the insurance was not effected, Pennant Hills is required by s. 18C of the Act to pay the compensation. It will thus be liable, during the lifetime of the injured workman, to make the periodic payments of compensation for which the Act provides. It is therefore entitled to receive as damages a sum that will put it in the same position as if it did not have to make those payments. The Court must therefore determine what lump sum, if paid at the date of the judgment, would represent the equivalent of all the periodic sums to be paid during the lifetime of the injured workman. 1. (1966) 114 C.L.R. 185, at p. 191. The assessment of such a sum of course requires a judgment to be made as to each of two matters, neither of which can be predicted with certainty. The first matter, the likely duration of the life of the workman, is not seriously in dispute, and it has been accepted that his expectation of life at the date of the trial was about 34 years. As to the second matter, the amount of compensation payable under the Act, it is clear that this amount was $64 per week at the time of the trial, and $83 per week at the date of the judgment given by the Court of Appeal on 16th December 1977. However s. 9A of the Act provides for the adjustment, twice a year, of the weekly rate of compensation in the same proportion as the "latest weekly rate" bears to the "base rate". The "base rate" within the meaning of that section is 144.57 and the "latest weekly rate" is a number shown in a specified statistical table as being "the weighted average minimum weekly rate, in dollars, payable for all industry groups to adult males for a full week's work (excluding overtime), as prescribed in awards, determinations and collective agreements". The law in force at the relevant date accordingly prescribed that the amounts payable under the Act to an injured workman are to vary in proportion to changes in the average weekly wage of all adult males employed in industry. It was submitted on behalf of Barrell that damages should be fixed on the basis that future payments of compensation under the Act will be made at the rate of $83 per week. In other words damages, it is said, are to be the same as if s. 9A did not appear in the Act. This anomalous result is said to be required by the principle that in the assessment of damages for personal injuries no account is to be taken of the possibility that future inflation will cause an increase in the nominal amount of wages. The submission was that the variations in the average weekly wage of all adult males, which provide the basis for indexation under s. 9A, are themselves largely the product of inflation. In my opinion the courts should, generally speaking, adhere to the rule, established by O'Brien v. McKean [7] , that in the assessment of damages for personal injuries no allowance should be made for inflation. To attempt to ensure that damages provide protection for the plaintiff against future inflation is, as Lord Scarman said in Lim Poh Choo v. Camden and Islington Area Health Authority [8] "seeking after a perfection which is beyond the inherent limitations of the system". It is unreasonable to suppose that any economist will be able to predict with accuracy the nature and extent of changes in the purchasing power of money during a period extending for several decades ahead. Whether inflation increases or is brought under control depends upon political and economic events and decisions at home and abroad as to whose occurrence it is not possible to do more than conjecture. Predictions as to the economic future in thirty years time may perhaps be made by a soothsayer but expert evidence cannot rationally be given on such a subject. Testimony as to the rate of inflation at times many years in the future would prolong trials and render more difficult the assessment of damages without providing any real assistance to the court. "Solid proof, on the basis of probability" (to use the words of the Chief Justice in O'Brien v. McKean [9] is impossible to obtain on such an issue. The only practicable course is, I think, that suggested by Lord Diplock in Mallett v. McMonagle [10] : "to leave out of account the risk of further inflation, on the one hand, and the high interest rates which reflect the fear of it and capital appreciation of property and equities which are the consequence of it, on the other hand". His Lordship was there speaking of cases under the Fatal Accidents Acts, but in my opinion his remarks are applicable to the case of the assessment of damages for personal injuries. In such a case, as well as in cases under the Fatal Accidents Acts, money should (except in exceptional cases) be treated as retaining its value, but in calculating the present value of payments which would have been received in future years, interest rates appropriate to times of stable currency should be adopted. 1. (1968) 118 C.L.R. 540. 2. [1980] A.C. 174, at p. 193. 3. (1968) 118 C.L.R., at p. 550. 4. [1970] A.C. 166, at p. 176. There are in my opinion points of distinction between the present case and cases in which it is necessary to assess damages for personal injuries. In cases of the latter kind, in the very nature of things perfect compensation cannot be awarded. It is true that a full compensation, or a just and fair compensation, must be awarded in such a case, but "in assessing a just and fair compensation the purpose is not to attempt by means of money completely to insure that the plaintiff will be placed for the rest of his life in the same position as if he had not sustained the injuries": Pamment v. Pawelski [11] , per Dixon J. In a case such as the present, however, the wrong done to Pennant Hills is purely economic, and the court is required to essay the difficult task of putting Pennant Hills in the same position as if the wrong had not been committed. Moreover, although in assessing damages for personal injuries the compensation in respect of loss of earning capacity is capable to some extent of calculation, it must be remembered that in such a case damages are estimated for the present loss of a capacity or faculty, albeit a loss which will or may be productive of financial loss: see such cases as Paff v. Speed [12] ; Graham v. Baker [13] ; Parente v. Bell [14] . In the present case there is, it is true, a present wrong or breach of contract, but the loss for which damages are to be given is not a present loss, but a loss that will occur from time to time in the future when the payments of worker's compensation fall to be made. However, in O'Brien v. McKean [15] , although Barwick C.J. and Windeyer J. recognized that different considerations may apply where an assessment is made of the cost of expenditure in the future from those that apply in relation to compensation for loss of earning capacity [16] , it was held that in general the possible depreciation of the purchasing power of money is not relevant in either case. I need not consider whether the matters which I have mentioned are sufficient to distinguish O'Brien v. McKean, since it is enough to hold, as I do, for the reasons that I am about to state, that the present is a case of an exceptional kind in which the interests of justice require that regard be had to estimated future wage rates, even if those rates have resulted from inflation. I should however make it clear that the present decision does not in my opinion affect the authority of O'Brien v. McKean. 1. (1949) 79 C.L.R. 406, at pp. 410-411. 2. (1961) 105 C.L.R. 549, at p. 566. 3. (1961) 106 C.L.R. 340, at p. 347. 4. (1967) 116 C.L.R. 528, at p. 534. 5. (1968) 118 C.L.R. 540. 6. (1968) 118 C.L.R., at pp. 548, 558. It will not be possible to place Pennant Hills in the same position as it would have been in if the insurance had been effected unless Pennant Hills receives an amount which represents the present value of the cost of making the payments of worker's compensation at the rates provided by the Act. It is therefore necessary that the amounts for which the Act provides, that is amounts as indexed by s. 9A, should provide the basis of compensation. To take the amount of $83, and to refuse to recognize that it was subject to adjustment in accordance with s. 9A, would be to assess compensation upon an unreal basis and one unfair to Pennant Hills. However, the evidence given before Yeldham J. as to the future wage rates was valueless and should have been rejected. Yeldham J. himself recognized that it rested on imponderables and assumptions. I have already given reasons for suggesting that a person giving evidence directed to such an issue is attempting an impossible task. A further objection to the conclusion reached by Yeldham J. was provided by Mr. Handley in argument. Yeldham J. held that the increase in the wage rate index over a period of 34.2 years will be at the average rate of 10 per cent per annum. However, the same average arrived at on the basis that inflation would increase by 15 per cent in the first decade, 10 per cent in the second and 5 per cent in the third would give a result more than twice as large as that obtained by assuming that the increases would have been 5 per cent in the first decade, 10 per cent in the second and 15 per cent in the third. The adoption of an average rate in these circumstances further illustrates the unreliability of the method adopted. Since there is no reliable evidence as to the manner in which s. 9A of the Act will operate, it is apparent that the assessment of damages is extremely difficult. That, of course, does not relieve the court of the duty of assessing the damages as best it can. The court must give some effect to the operation of s. 9A. In the circumstances, the only feasible course, rough and ready as it is, seems to me to reduce the rate at which the amount of the future payments is discounted to arrive at their present value. In other words the court should proceed on the basis that an amount of $83 will continue to represent the amount payable by way of worker's compensation in the future, but should take into account the effect of the provisions of s. 9A by adopting an exceptionally low discount rate. As I endeavoured to explain in Cullen v. Trappell [17] when, in assessing damages, the present value of economic loss is determined by the use of tables which show the present value of a sum of money to be received over a period of time in the future, it is necessary, in order that the method followed should be consistently applied, that regard should be had to the notional tax on the income assumed to be derived from the amount awarded for the future economic loss. I must again emphasize that what is referred to is notional tax on notional income. The court is in no way concerned to enquire what tax if any will actually be payable by the plaintiff. The present value of the future loss is arrived at by the use of tables which show what sum, if invested at the given rate, would produce sufficient income to allow the given weekly sum to be obtained by recourse both to income and capital over the whole period, so that at the end of the period the fund is entirely exhausted. It is not assumed that in fact the plaintiff will invest the sum in this way. It is fundamental that the court has no concern with what the plaintiff actually does with his damages. Therefore, in the present case, it is quite irrelevant that Pennant Hills may in fact receive a tax deduction in respect of the payments of worker's compensation when it comes to make them. Pennant Hills may cease to carry on business, so that no deduction is claimable, or for some other reason a tax deduction, if allowable, may be of no value to it, but questions of this kind are collateral to the issue of damages and must be entirely disregarded. It would be a misapprehension of the method discussed in Cullen v. Trappell to set off, against notional tax, expected real tax benefits. The reason why it is necessary to consider the incidence of tax in assessing the amount payable to a plaintiff whose earning capacity has been lost or damaged is that damages are compensatory, and it would be unreal to measure the loss by reference to gross earnings when in fact the earnings would have been subject to tax. We are now concerned with a quite different question: in applying an arithmetical method to determine the present value of a future loss, is it right to have regard to the probable taxation position of the individual plaintiff? In my opinion it is not, for the method is a purely abstract one, which leads to the same result irrespective of the circumstances of the particular case. The estimated amount of the loss, and the length of time for which it will be suffered, depend entirely on the circumstances of the individual case, but the present value of such a loss is merely a matter of calculation, to which the facts of the case are not relevant. To avoid possible misapprehension, I should add that the quite different question, whether it was relevant to take into account the tax that it was said would be payable on the sum received as damages, although the subject of argument and decision the Supreme Court, was not argued on appeal to this Court, and I accordingly say nothing about it. 1. (1980) 146 C.L.R. 1, at pp. 10, 11. There is no material which would assist the court in taking into account the tax that would be notionally payable on the income produced by the notionally invested fund. In the circumstances it seems to me appropriate to adopt the expedient of fixing a discount rate which will take into account the effect, not only of s. 9A of the Act, but also of the notional tax on the notional income from the invested fund. I would adopt the rate of 2 per cent. I agree with my brother Mason that the domestic expenses in question are not expenses incurred for "nursing" within the meaning of those words in the definition of medical treatment in s. 10 of the Act. There was a duplication of process — each party appealed and cross-appealed. I would dismiss the appeal and cross-appeal by Pennant Hills and allow Barrell's appeal and cross-appeal. Stephen J. These appeals and cross appeals concern the damages recoverable by a plaintiff employer, Pennant Hills Restaurants Pty. Ltd. As a result of the defendant's default it has become liable to make weekly payments of uncertain amount to a statutory fund for an uncertain period in the future. The defendant, Barrell Insurances Pty. Ltd., is an insurance broker. It failed to secure for the plaintiff, its client, indemnity against liability for workers' compensation. An employee of the plaintiff was injured in compensable circumstances and is in consequence a paraplegic. The plaintiff, as an uninsured employer, must now periodically reimburse a statutory fund for those workers' compensation payments which the fund will make to the injured employee for the rest of his life. The common law, to which the plaintiff necessarily had recourse, knows only lump sum damages, paid once and for all. Schemes of workers' compensation, on the other hand, typically provide for periodical payments, thereby overcoming the notorious difficulties involved in matching the award of a present lump sum to the future financial needs of an injured worker. What this case has called for has been the assessment of common law damages which will compensate the plaintiff for the burden of those periodical payments for which, because of the defendant's default, the plaintiff has become liable. The terms of the relevant workers' compensation legislation, the Workers' Compensation Act, 1926 N.S.W., confronted the trial judge with the need to consider whether his award of damages should reflect the likelihood of future fluctuations in wage rates. The periodical payments to the injured worker from the statutory fund, for which the plaintiff has to reimburse the fund, are indexed payments: that is, they are subject to variation twice a year so as to accord with changes in a particular Commonwealth statistical index which reflects the weighted average minimum weekly award wages (excluding overtime) for adult males. Any future changes in such wages will, by the mechanism of indexation, alter the amount of payments to the worker and, in consequence, of the plaintiff's periodic reimbursements to the fund. The trial judge also had to consider the income tax consequences for the plaintiff of any award of damages and whether any account should be taken of them in calculating the lump sum damages to be awarded. This aspect may, for the moment, be put to one side. The curial background. At first instance in the Supreme Court of New South Wales the matter came before Woodward J. Experience suggested to his Honour that future inflation would, over time, result in considerable increases in the level of award wages and, accordingly, in the amount of the plaintiff's periodic reimbursements to the fund. His Honour thought that he could have regard to the fact that current high interest yields on fixed interest investments reflected inflationary expectations. His adoption of 5 per cent tables in discounting for present payment, regardless of the much higher yields presently offering, would, he thought, provide some hedge against inflation. However his Honour, after a careful review of authority, concluded that he was not otherwise "permitted to take inflationary trends into consideration". With the details of his Honour's assessment of damages I am not otherwise concerned. On appeal to the Full Court, Reynolds J.A., while disagreeing with aspects of the assessment made at first instance, shared the view of Woodward J. that, despite the statutory provision for automatic indexation of the worker's periodic payments and their likely effect upon the plaintiff's reimbursement obligation, the authorities required the award of damages to be computed exclusively upon the footing of the then current rate of periodic payments. To allow for possible future increases would be to take the impermissible step of paying regard to the likelihood of future inflation. Using 5 per cent tables, his Honour would have allowed only some $88,000 to the plaintiff on account of its liability to reimburse the fund in the future. Hutley J.A., on the contrary, felt able to distinguish the present case, affected as it was by the statutory provision for indexation, from those authorities which Reynolds J.A. regarded as applicable. The plaintiff, he said, was entitled to "such damages as it can prove it will suffer by reason of future increases in workers' compensation payments". He would have awarded the same sum as Reynolds J.A. "plus an allowance for the present value of the index increments". This would require the remission of the case to the Common Law Division for the taking of evidence on the issue of "the increase of the index and how the formula can be expected to work in relation to it". Mahoney J.A. was of a like view. He too was able to distinguish earlier authority and to conclude that it would be proper to take account of such increases in periodical payments as might result from the indexation provisions of the workers' compensation legislation. He concurred in the order proposed by Hutley J.A. Accordingly, by a majority the action was remitted to the Common Law Division for the determination of what further sum should be included in the verdict to provide for the present value of future increases in reimbursement payments which the employer might be called upon to pay. Pursuant to this order Yeldham J. in the Common Law Division heard further evidence and in his judgment dealt with two questions: the probable increases in the relevant index which might take place over the worker's life expectancy of 34.2 years, and the amount which should be added to the plaintiff's verdict to compensate it for its liability to reimburse the fund in respect of such increases. His Honour accepted the evidence of an economist that it was reasonable to assume that over the worker's life expectancy "the average annual increase in the wage rate index would be 10 per cent, or within a range of 8 per cent to 12 per cent per annum". Accepting actuarial evidence before him and discounting according to 5 per cent tables, his Honour arrived at a sum of $329,600 as the further sum to be awarded to the employer so as to provide for the present value of future increases in periodical payments which it might be called upon to pay. The Court of Appeal subsequently ordered that, interest apart, there should be a verdict for the plaintiff in the sum of $475,693.65 of which $329,600 represented the present value of future increases in the employer's reimbursement liability. It is in these circumstances that this appeal now comes to this Court. The statutory indexation. It is commonplace, in any assessment of damages involving future losses or future liabilities, to encounter factors difficult of prediction, either in their incidence or their significance and quantum. In the present case a special difficulty arises due to the statutory indexation of workers' compensation benefits. Its effect is independent of any conduct of the parties, depending exclusively upon the economic circumstances of the future. It raises in principle those issues concerning future inflation of which, in recent years, common law courts have been uneasily aware and in the disposal of which no very satisfactory solution seems to have been found. Due to this indexation it is, one supposes, a practical certainty that the weekly periodic payments to the worker, and in turn the employer's liability to reimburse the statutory fund, will vary in amount over the next thirty or more years. The legislature has provided for indexation in anticipation of changes in award wages and it seems highly improbable that changes will not occur. There were, during the nineteenth century, substantial periods of relative wage stability, but the experience of the second and third quarters of this century has been one of wage rates which do not long remain at any fixed level in nominal terms. Were the experience of the past decade a reliable guide to the future, it would suggest a pattern of very large wage increases in nominal terms, accompanied in some years by not insubstantial increases in real terms. However, one needs no expert evidence to know that during this century no one decade of economic experience has been repeated in the following decade. In the long term the nation's economy is affected by many imponderables: war; the preparations lest it break out and the painful recovery from its consequences; changes in world markets and in technology; political and social changes at home and abroad; changes in prevailing economic theory and practice. All these contribute to make the past no very certain guide to what will be the economic experience of the future, at all events in the long term, although for want of any better guide past experience may continue to loom large in predictions of the future. The principle of compensation. The law requires that this plaintiff should be compensated in a lump sum for the financial detriment which it has suffered by the defendant's failure to arrange for its indemnification, a detriment involving it in liability periodically to reimburse a statutory fund over more than thirty years in the future. Lord Scarman has recently reminded us, in Lim Poh Choo v. Camden and Islington Area Health Authority [18] , that "the principle of the law is that compensation should as nearly as possible put the party who has suffered in the same position as he would have been in if he had not sustained the wrong", echoing Lord Blackburn in Livingstone v. Rawyards Coal Co. [19] . This is the principle which in The Liesbosch [20] Lord Wright described as "the dominant rule of law" and of which, in British Transport Commission v. Gourley [21] , Lord Reid spoke when he said that the general principle was not in doubt, a plaintiff was entitled to have made "good to him the financial loss which he has suffered and will probably suffer as a result of the wrong done to him". 1. [1980] A.C., at p. 187. 2. (1880) 5 App. Cas. 25, at p. 39. 3. [1933] A.C. 449, at p. 463. 4. [1956] A.C. 185, at p. 212. In an era of endemic inflation to assess the plaintiff's damages on the assumption that current award rates will remain unchanged for the next thirty-four years disregards that principle of the law. It is an assumption which few members of the community would at the present day make in planning their own affairs. It is surely impermissible for the law to adopt it as a basis of compensation designed to restore plaintiffs to the position they would have occupied but for defendants' defaults. It denies to plaintiffs that "full compensation" of which Dixon J. spoke in Pamment v. Pawelski [22] . Particularly is this so where, as here, the measure of a plaintiff's damages is directly referable to a statutory obligation which, by its indexation provisions, recognizes that, due to inflation and other causes, wage rates are likely to increase in the future. I regard the present statutory context as reinforcing the requirements of principle: that is, that if there be a likelihood of future changes in the purchasing power of money, the assessment of damages should pay regard to the future purchasing power of the amount presently awarded. What is awarded should be designed to provide an amount in each future year equal to the compensable financial detriments of that year. 1. (1949) 79 C.L.R., at p. 411. Since this will also be the case when damages are to be assessed for personal injuries involving future outgoings or the loss of future earnings, it will be important that the way in which the present appeal is resolved should not only be supportable as a matter of reasoning but should also be practicable in its application to the everyday trial of actions in our courts. Curial attitudes to inflation. There has, of late, been a heightened awareness of the impact of inflation and an increasing disillusionment concerning the ability of investment strategies to provide an effective hedge against it. In England the case of Cookson v. Knowles [23] happens to provide recent dicta reflecting each of these attitudes. In the Court of Appeal Lord Denning [24] explained the discarding by the Court of a particular rule for the assessment of damages by saying that it had been laid down at a time when "inflation did not stare us in the face. We had not in mind continuing inflation and its effect on awards". When the case was on appeal to their Lordships both Lord Salmon [25] , and Lord Fraser of Tullybelton [26] , described as now "exploded" the long-held theory that investment in equities provided a hedge against inflation. The fact that "by chance" investment in fixed interest securities now offered such a hedge might, Lord Salmon thought, be no more than a temporary situation. One may remark in passing that it will be small comfort to a successful plaintiff who has invested his damages in a form of investment regarded at the time as prudent and appropriate to learn that, however disastrous his particular investment, another form of investment exists which, could he but have predicted the future, would currently be offering that protection against inflation which his actual investments have failed to provide. 1. [1977] 1 Q.B. 913; [1979] A.C. 556. 2. [1977] 1 Q.B., at p. 921. 3. [1979] A.C., at p. 574. 4. [1979] A.C., at p. 577. There have also been occasional statements such as those of Lord Reid in Taylor v. O'Connor [27] who spoke of the courts' "nostalgic reluctance to recognize change" and said of inflation that it was "quite unrealistic to refuse to take it into account at all" at a time when "few believe that a return to the old financial stability is likely in the foreseeable future". However, the furthest that United Kingdom courts have in fact gone is to allow for future inflation "in a rough and ready way" by discounting at a relatively low rate of about 4½ per cent said to represent a rate of interest appropriate in times of stable currency, despite the ability of plaintiffs to invest their awards at "the higher rates of interest obtainable as one of the consequences of" inflation — Lord Diplock in Cookson v. Knowles [28] and earlier in Mallett v. McMonagle [29] . In Lim Poh Choo [30] , Lord Scarman gave certain reasons for otherwise disregarding the possibility of future inflation including the fact that forecasts of the economic future were necessarily speculative and that plaintiffs might, by suitable investment strategies, reduce its effect. 1. [1971] A.C. 115, at pp. 129-130. 2. [1979] A.C., at p. 571. 3. [1970] A.C., at p. 176. 4. [1980] A.C., at p. 193 These two reasons of his Lordship are very much those relied upon by such of the jurisdictions in the United States as still fail to allow for future inflation. In Schnebly v. Baker [31] Uhlenhopp J. of the Iowa Supreme Court described them as follows: The principal opposing argument is that future inflation or deflation is speculative, and that the injured person can in any event invest his recovery in equities such as land and thus have the benefit of both inflation and the income. United States decisions reflect the whole spectrum of possible views. There are jurisdictions which resolutely disregard future inflation — Williams v. United States [32] (United States Court of Appeals, 1st Circuit) and Johnson v. Penrod Drilling Co. [33] (5th Circuit). There have been intimations that "if inflation should continue at its present pace, courts may have to reconsider the propriety" of existing practice — Yodice v. Koninklijke Nederlandsche S.M. [34] (2nd Circuit) per Judge Friendly, a prediction to which the Second Circuit gave effect in Feldman v. Allegheny Airlines [35] (United States District Court). There have been carefully reasoned decisions to follow a discerned "trend among a number of courts to allow the jury to consider the impact of future inflation" in personal injuries cases where loss of future income is in question, at least where expert evidence of the impact of future inflation, based upon a proper factual foundation has been tendered — District of Columbia v. Barriteau [36] (D.C. Court of Appeal) and see United States v. English [37] (9th Circuit). There have been seemingly more radical solutions in which discounting for present payment is simply abandoned so as to make allowance for future inflation—Beaulieu v. Elliott [38] (Alaska) and Leavitt v. Gillaspie [39] (Alaska). In general, there appears to be a distinct movement away from disregard of future inflation and towards the adoption of varying methods of taking it into account. 1. (1974) 217 N.W. 2d 708, at p. 727. 2. (1970) 435 F. 2d 804. 3. (1975) 510 F. 2d 234. 4. (1971) 443 F. 2d 76, at p. 79. 5. (1975) 524 F. 2d 384; (1974) 382 F. Supp. 1271. 6. (1979) 399 A. 2d 563, at p. 567. 7. (1975) 521 F. 2d 63. 8. (1967) 434 P. 2d 665. 9. (1968) 443 P. 2d 61. In Canada, in a much discussed trilogy of cases, the Supreme Court recently determined that account should be taken of the effect of future inflation in assessing damages. This it did quite explicitly by adjusting the rate of discount for present payment. It arrived at what it regarded as an appropriate rate by subtracting from current market rates of interest what it took to be the predicted rate of long-term inflation. Detailed critical analyses of these Canadian decisions have appeared in a number of Canadian journals of the time—see for example C.J. Bruce, The Calculation of Foregone Lifetime Earnings, Canadian Public Policy, Spring 1979, p. 155: B. Feldhusen and K. McNair, General Damages in Personal Injuries Suits, University of Toronto Law Journal, vol. 28 (1978), p. 381: S. A. Rea, Inflation, Taxation and Damage Assessment Canadian Bar Review, vol. 58 (1980), p. 280. Their criticism is not at all directed to the Supreme Court's recognition that the prospect of future inflation should be allowed for in awards of damages; it is concerned, rather, with details of the method adopted by the court in making that allowance and with the factual material concerning future inflation upon which it relied. The attitude of Australian courts has hitherto been substantially to disregard questions of the impact of future inflation upon awards of damages. They have been influenced by the obvious difficulty of predicting the future, by the speculative nature of the evidence upon which predictions of the economic future must rest, and by the added complexity which accounting for inflation would introduce into trials. They have drawn solace from the opportunity which a plaintiff is said to have to so invest his award of damages as somewhat to guard against inflationary erosion of its real value. In addition, the characterization of lost earning capacity as a present loss, despite its element of futurity, has been said to avoid the need to take future inflation into account. Again, because a plaintiff may do what he will with his damages, it is sometimes said to be unjustified to assume that they will be used to acquire investments which will be subject to inflationary erosion of real values. That the process of assessment of fair compensation is a matter of judgment rather than of calculation has also been relied on; to allow for inflationary factors by applying correcting formulae has been said to give to the whole process a spurious air of exactitude and certainty. A modern authority in this Court, O'Brien v. McKean [40] , gathers together many of these diverse currents of thought, some of which also find expression in the judgment of Windeyer J. sitting at first instance in Parente v. Bell [41] . 1. (1968) 118 C.L.R. 540. 2. (1967) 116 C.L.R. 528. It is natural that in post-war years there should have been concern lest the complicating factor of future inflation endanger the whole system of assessment of awards of damages for personal injuries, a system already under considerable strain in a society exposed, since the last war, to the consequences of widespread road trauma. Moreover, the 1960s marked the closing years of a very long period during which inflation, although well-known and occasionally significant for short periods, was not endemic at high rates, as it is today; against it appropriate investment strategies did seem to offer an effective hedge. The changed economic circumstances of the present day need no elaboration, but the fact of those changes is of importance. As Lord Scarman said in Lim Poh Choo [42] , the courts' treatment of future inflation involves no rule of law, only a rule of practice. In Admiralty Commissioners v. S.S. Chekiang [43] Lord Sumner said of such rules of practice: "The measure of damages ought never to be governed by mere rules of practice, nor can such rules override the principles of the law on this subject." To this Lord Wright, in The Liesbosch [44] , added that "subsidiary rules can only be justified if they give effect to" the dominant rule, the principle of restitutio in integrum. The considerations referred to in O'Brien v. McKean [45] are essentially practical: they are not concerned with principles of law so much as with the mechanics of applying principle to fact; of converting into money sums a plaintiff's entitlement to compensation for personal injuries. The mechanics of doing so must depend upon the quality of the currency which is the medium of compensation; it must necessarily differ when what is seen as a relatively stable currency becomes a currency likely to be subject to long-term depreciation in real value. What was said in a different context by Lord Wilberforce in Miliangos v. George Frank (Textiles) Ltd. [46] , when he cited with approval what had then recently been said by Lord Reid, is much in point. Lord Wilberforce said [47] : A recent example of the House changing a very old established rule is West Midland Baptist (Trust) Association (Inc.) v. Birmingham Corporation [48] . Lord Reid thought that it was proper to re-examine a judge-made rule of law based on an assumption of fact (as to the stability of money) when the rule was formulated but which was no longer true and which in many cases caused serious injustice. It is noteworthy that in the United Kingdom the recent report of the Royal Commission on Civil Liability, the Pearson Report, sees a need, in this inflationary age, for a new approach to the assessment of lump sum damages—see vol. I, pars. 646-708. It points out, in par. 648, that to discount at about 4½ per cent, assuming the plaintiffs can obtain a net rate of return of some 4½ per cent (after the effects of tax and inflation have been experienced) is "not now realistic. The net real rates of return have over a number of years been considerably lower than 4½ per cent". 1. [1980] A.C., at p. 193. 2. [1926] A.C. 637, at p. 643. 3. [1933] A.C., at p. 463. 4. (1968) 118 C.L.R. 540. 5. [1976] A.C. 443. 6. [1976] A.C., at pp. 469-470. 7. [1970] A.C. 874. To sacrifice the fundamental principle of proper compensation by adhering to methods of assessment which, in changed circumstances, no longer obtain it cannot be a proper outcome of the doctrine of precedent. It seems clear that the conventional Australian basis of assessment, which substantially disregards the effect of future inflation, no longer provides a satisfactory means of compensating plaintiffs for future economic loss or liability. Inflation at currently prevailing rates and those thought likely to exist in the future will quickly render grossly inadequate an award of damages calculated upon such a basis. A striking example of just such a case, no doubt far from atypical, is provided in Appendix 5 to vol. 1 of the Report of the National Committee of Inquiry into Compensation and Rehabilitation in Australia, 1974, the Woodhouse Report. It is recounted in detail in P. S. Atijah's Accidents, Compensation and the Law, 2nd ed. (1975), pp. 182-183. Accordingly, even if the present case were a simple personal injuries claim I would not regard adherence to the method of assessment discussed in O'Brien v. McKean as mandatory. But in fact this is a very different case. In this respect I share, with respect, the views expressed by the majority in the Court of Appeal. As was said by Mahoney J.A. in the present case [49] "Whatever be the ambit of O'Brien v. McKean [50] , it does not, in my opinion, require that increases resulting from statutory provisions should not here be taken into account." The problem in this case is simply how best to translate into a present lump sum award a stream of future outgoings whose future volume is, by statute, made to depend upon future economic events as reflected in future average minimum award wages. 1. [1977] 2 N.S.W.L.R. 827, at p. 863. 2. (1968) 118 C.L.R. 540. Possible methods of allowing for future inflation. There are two well recognized methods by which courts have sought to take account of future inflation. The first, the "real interest rate" approach, seeks, in the assessment process, to make use of those same economic forces as will, in the present case, cause periodic payments to fluctuate in the future. It involves no direct prediction of rates of future inflation but operates exclusively through manipulation of the discounting rate. The second, the predictive approach, would involve the prediction, with the aid of expert evidence, of the future pattern of changes in award wages, thus estimating the future trends in the relevant index and hence the changes in the periodic reimbursements for which the plaintiff is to be liable. This done, a lump sum is calculated which, with suitable discounting for present payment, will from capital and income provide a fund sufficient to meet the total reimbursements. The real interest rate approach. The real interest rate approach has obvious attractions since it involves no predictions as to the economic future: it does not endeavour by such predictions to pitch the amount awarded at a level appropriate to predicted conditions prevailing in the future. Instead it is concerned directly with the sum to be awarded. While it recognizes that the future will involve changes in economic conditions and acknowledges that their accurate prediction is impossible, it seeks, by using what is said to be a recognized economic fact, to award such a present sum as will enable the plaintiff, from income and capital, to meet future compensable commitments. It rests upon the assumption that interest rates have two principal components: the market's own estimation of likely rates of inflation during the term of a particular fixed interest investment, and a "real interest" component, being the rate of return which, in the absence of all inflation, a lender will demand and a borrower will be prepared to pay for the use of borrowed funds. It also relies upon the alleged economic fact that this "real interest" rate, of about 2 per cent, will always be much the same and that fluctuations in nominal rates of interest are due to the other main component of interest rates, the inflationary expectation. Upon these assumptions future outgoings or future losses of earnings may be calculated upon the footing that they will continue into the future at current rates. The amount so computed is then discounted not by any market rate of interest but only by a rate representing the "real interest" rate, say 2 per cent. Professor Fleming describes this method in an article "The Impact of Inflation on Tort Compensation", American Jo. of Comp. Law, vol. 26 (1978), pp. 51-78; he speaks of discounting by the real interest rate, of which he says, at p. 66, "The real (or "opportunity") cost of money is relatively stable but the actual cost is increased by the inflation component." This real interest rate he describes, at p. 68, as "inflation-free", so that when used as the discount rate it "assures that the plaintiff will remain least affected by whatever turns out to be the rate of inflation in future years and therefore provides a reliable automatic adjustment". The real interest rate is to be calculated by subtracting the average yearly inflation rate from the current interest rate on an appropriately risk free investment. See also the article by T. E. Malone, Considering Inflation in Calculating Lost Future Earnings in Washburn Law Journal, vol 18 (1979), p. 499, especially at p. 511, and that by S. A. Rea to which I have earlier referred and R. Posner's Economic Analysis of Law, 1st ed. (1973), p. 81 (but compare 2nd ed. (1978), pp. 147-148). The case of Feldman v. Allegheny Airlines [51] to which I have also earlier referred provides one of numerous instances of the application of this approach. 1. (1974) 382 F. Supp. 1271. Because its proponents hold that high interest rates constitute an opposing response to future inflation, or the expectation of it, this method need neither take account of inflation in the future nor discount by prevailing high nominal rates of interest. Each is omitted from the calculation because the omission of the one offsets that of the other. The need for evidence of the economic future is avoided, as is any difficulty in choosing a suitable rate of discount. By the use of this method the effect of future inflation is, so it is claimed, "factored out" of the calculations. The attraction of the method is, of course, its simplicity of application. It also promises to make automatic allowance for inflation in the future without the need for the prediction of changes in the economic environment. Its application would lead to a marked increase in the quantum of awards compared with those at present awarded by Australian courts; instead of discounting for present payment at 5 per cent or 6 per cent, the appropriate "real interest" discount would be only 2 per cent or perhaps 3 per cent. This consequence is said to do no more than reflect existing gross undercompensation of plaintiffs due to the impact of inflation. However, founded as it is upon the assumed existence of a relatively constant rate of "real interest", if that assumption be shown to be false, the method loses the rationale upon which it depends. Past Australian economic experience appears to provide little support for the concept of a relatively constant rate of "real interest". Year by year a figure for "real interest" can of course be calculated, simply by subtracting from nominal interest rates the rate of inflation. But these figures are no more than a series of numbers bearing no resemblance to any relatively constant rate of interest which lenders are supposed to demand and borrowers to pay after allowing for estimated inflation. If official statistics for the past twelve calendar years are consulted, the Reserve Bank of Australia's Statistical Bulletins supply interest rates on two-year Australian government bonds (non-rebatable) and the O.E.C.D. Economic Outlook — July 1980, p. 105 and p. 143, supplies annual percentage changes in consumer prices, which gives a measure of inflation. The difference figure year by year, which should represent the "real interest" rate, averages out at a negative average rate of interest of -1.46, the widest fluctuations found in particular years being a positive rate of 2.58 per cent and a negative rate of -6.61 per cent. Nothing resembling a relatively constant positive rate of 2 per cent-3 per cent emerges. An equally random series of numbers, showing no steady rate of "real interest", appears as Table 9.1 in the recent Interim Report of the Campbell Committee of Inquiry (Australian Government Publication Service (1980)). For the period of thirty years which that Table covers, from 1950 to 1979, the average "implicit real interest rate" is a negative rate of -.7 per cent, with 4 per cent as the greatest positive rate in any year and -20.2 per cent as the greatest negative annual rate. If the material in Table 9.1 is confined to approximately the same time span as the twelve year period earlier referred to it generally confirms the earlier results. It shows an average negative rate of -.6 per cent with variations up to a positive 3.3 per cent and down to a negative -8.2 per cent. Discrepancies between the two sets of results are due to differences in the base periods and in the statistical sources of information. Table 156 in Ch. 28 of the Pearson Report, showing "real rates of return on capital" provides a convenient comparison of United Kingdom figures. Mr. Bruce's article in Canadian Public Policy, earlier referred to, contains on p. 162, a similar tabulation of Canadian experience of real interest rates. Academic writings confirm the absence, in recent Australian experience, of any relatively steady rate of real interest. Those sources cited earlier as lending support to this "real interest rate" approach take as their starting point the teachings of the economist Irving Fisher, and describe the behaviour of interest rates in terms of inflation as being that, as prices rise in response to inflationary pressures, the nominal interest rate rises proportionately, the real interest component remaining constant. In the case of Australia it has, however, been said of empirical tests that "They do not confirm Irying Fisher's hypothesis that the nominal interest rate is equal to the real interest rate plus the anticipated rate of change in the price level": "Interest Rates and Price Expectations", Juttner and Bartels, Macquarie University School of Economic and Financial Studies, Research Paper No. 30 (1973). Steindl, writing in Journal of Money, Credit and Banking, vol. 5 (1973), says, in his article "Price Expectations and Interest Rates" at p. 948, that only where, the expectation of inflation having reduced the real demand for money, increased real demands in both commodity and bond markets are equal and offsetting will the Fisher effect hold good. In Feldman v. Allegheny Airlines Inc. itself, the Court [52] , found, in the United States economic context, that there occurred fluctuations in real interest rates no less startling than those extracted above from Australian statistics. In discussingFeldman's Case [53] in Freeport Sulphur Co. v. S.S. Hermosa [54] Judge Wisdom of the Fifth Circuit Court of Appeals drew attention to this and spoke of the "speculation inherent in factoring the inflation component out of the market rate of interest" [55] . A detailed criticism of this approach appears in W. T. McGough's article, "Future Inflation, Prospective Damages and the Circuit Courts" in the Virginia Law Review, vol. 63 (1977), p. 105. 1. (1974) 382 F. Supp., at pp. 1293-1294. 2. (1974) 382 F. Supp. 1271. 3. (1976) 526 F. 2d 300. 4. (1976) 526 F. 2d, at p. 310. The predictive approach. The alternative approach, that in fact adopted in the present case, involves the reception of evidence concerning the economic future. Although, unlike the "real interest rate" approach, it does not rest upon any dubious economic theory, it suffers from other serious defects. The most marked is its reliance, case by case, upon expert evidence regarding an inherently uncertain fact, the economic future. This has given rise to the familiar and substantial objection not only to the speculative nature of such evidence but also to the added complexities which the introduction of such evidence will bring to trials. Objections such as these lie at the root of much past Australian reluctance to take account of inflation in the assessment of damages. There are two particular aspects of the predictive approach which merit special mention. The first is a consequence of the introduction of expert evidence; in a number of cases there is likely to be conflicting expert evidence and judges, and in some States juries, will have to choose between those conflicting predictions of the economic future. The evidence accepted may well vary from case to case, the conclusion of the trier of fact concerning one and the same subject matter, the likely economic future, accordingly differing from case to case. This may result in wide disparities between awards and any amounts awarded will be difficult to predict in advance. This in turn will make more difficult the negotiation of settlements out of court, which at present accounts for the resolution of the vast majority of claims. The evidence in the present case illustrates a second point: that it will not be enough that the expert evidence should essay the formidable task of predicting an average rate of inflation for the relevant period in the future; its prognostication must also extend to the distribution of inflationary swings within the period. In the present case an economist testified at the trial that in the long run, over the worker's life expectancy of some three and a half decades, the index to which the periodic payments, and hence the plaintiff's reimbursement obligation, are tied, was likely to reflect an increase in wage rates averaging about 10 per cent per annum, or within a range of 8 per cent to 12 per cent per annum. But this rate of increase would not be constant: the witness said that he could "make with great confidence the forecast that the rate will vary considerably". His average rate was composed of two components, an inflation factor ranging from 6 per cent to 8 per cent per annum and an annual increase in the real value of wages, related to increases in Australia's gross domestic product, ranging from 2 per cent to 3 per cent. The inflation component might, he thought, fluctuate greatly throughout the three decades. To illustrate his point the witness cited, merely as examples, "fairly heavy swings" of 8 per cent, 3 per cent and 9 per cent in successive decades. Even were the predicted overall rate of 10 per cent to prove precisely accurate, the failure correctly to predict the distribution of such swings within the thirty year period will nevertheless lead to gross error in the amount awarded. The extent of possible gross error was illustrated in the respondent's written submissions which were supported by actuarial evidence in the trial: a constant rate of 10 per cent averaged over thirty years would, on a given hypothesis, produce $329,000; on that same hypothesis that same average "arrived at by decades of 15 per cent, 10 per cent and 5 per cent inflation would give a figure of $481,000 and on the basis of 5 per cent, 10 per cent and 15 per cent would give $220,000". The distribution pattern of the swings thus becomes of overwhelming importance. Yet of any likely distribution pattern the witness could say nothing. Instead he was, very properly, at pains to emphasize that it was only because a long period, of over thirty years, was in question that he felt able to give his evidence of an average of 10 per cent per annum, relying upon the averaging out of fluctuations over the years. Feldthusen and McNair in their article in the University of Toronto Law Journal earlier referred to, in the course of their review of the trilogy of personal injury suits decided by the Supreme Court of Canada, refer to the dangers of averaging inflation rates over a period, saying, at p. 394, that "reliance upon an average rate creates further and financially significant problems. The annual fluctuation from the average may substantially affect the adequacy of the award": even if the long term average rate proves, with hindsight, to have been entirely accurate, the distribution of the short term highs and lows which go to make up the average will greatly affect the outcome. If higher than average rates of inflation occur in the early years the distortion of result will cause the amount awarded to be inadequate; if they occur in the later years the award will be excessive. These considerations provide a strong disincentive to the adoption of any method of assessment of damages involving specific forecasts of future inflation. In the United Kingdom both the majority and the minority of the Pearson Committee avoided such a method in that restricted range of cases in which they recommended that lump sum awards of damages should be retained; the majority was conscious of the "considerable uncertainty" to which any calculations of lump sum damages for future loss of income was subject: par. 646. The undiscounted approach. If neither of these approaches offers a satisfactory solution, a third suggests itself. If gross undercompensation is to be avoided without suffering the disadvantages inherent in these two approaches, the solution may lie simply in the abandonment of discounting for present payment. The only reason for discounting for present payment is to offset the unintended advantage which a plaintiff gets from present payment, namely, the chance to earn income from its investment. But against that should be weighed the disadvantage he suffers by only being compensated at present rates of wages (or at present rates of outgoings) when his lost future wages (or future outgoings) would have proved much greater in money terms. If the advantage and disadvantage can be seen approximately to offset each other, neither need figure in the process of assessment and the occasion for discounting disappears. During periods of sustained inflation such as we have known in recent years and such as the expert evidence in this case predicts for the future, there does appear to occur something very like such a cancelling out of advantage and disadvantage. This is demonstrated by the Australian statistics to which I earlier referred in considering the real interest rate approach. Of course, as those figures reveal, no precise set-off occurs between the rates of interest and of inflation. But, continuing, as I do, to ignore for the moment the impact of income tax upon investment income, it does appear that so long as the future holds promise of continued and relatively high rates of inflation substantial justice would be done both to plaintiffs and to defendants by simply awarding as damages the total estimated loss of future earnings (or future outgoings) calculated at current rates and without making any discount for present payment. Such an approach to the assessment of damages does not depend upon the theoretical existence of a relatively constant real rate of interest nor upon any other economic theory of perhaps questionable validity in the Australian context. It relies instead upon recent Australian experience of movements in interest rates and rates of inflation which suggest that the advantage to a plaintiff of present payment is offset by likely future erosion of the value of his award due to the effects of inflation. The approach is both simple and pragmatic. The reason why something like set-off occurs is that, although there is no precise correlation between interest rates and inflation sufficient to support a strict real interest rate theory, nevertheless over time interest rates do tend to reflect inflationary expectations. This phenomenon was noted by J. A. Carlson in his article "The Present Value of Future Earnings" in the American Bar Association Journal, vol. 62, p. 628, in which he provides an economist's account of the causes. In Beaulieu v. Elliott [56] (Alaska) the Court treated the discount rate as fully offset by the future rate of inflation and accordingly awarded damages which were undiscounted, and see Leavitt v. Gillaspie [57] . In the article by W. T. McGough already referred to, "Future Inflation, Prospective Damages and the Circuit Courts", the author's lengthy examination of different approaches concludes that te method adopted in Beaulieu is "particularly appealing when compared with other approaches in the contexts of predictability, efficiency and accuracy". The judgment of Judge Wisdom in Freeport Sulphur Co. v. S.S. Hermosa [58] , to which I have already made some reference, provides closely reasoned support for the adoption of the Beaulieu approach. 1. (1967) 434 P. 2d 665. 2. (1968) 443 P. 2d, at p. 69. 3. (1976) 526 F. 2d 300. To discard discounting for present payment of course runs counter to long practice in the assessment of damages in Australia. But such discounting was adopted as no more than a mathematical aid in arriving at the true measure of compensation. When, due to changed circumstances, it no longer serves the end for which it was designed but instead ensures that that end will not be attained, its retention is unjustified. This undiscounted approach lays no claim to precise accuracy. But the whole process of assessing damages, not only where personal injuries are in question but also, to a lesser degree, in a case such as the present, is necessarily very much a matter of estimate and approximation. Uncertain factors abound, the most obvious of which are, perhaps, the estimated future lifespan of a plaintiff and what would have been his pattern of earnings, but for the accident: the case of young plaintiffs raises the latter in a particularly acute form. In personal injuries cases the question of future contingencies and vicissitudes is also essentially uncertain. In such an imprecise process, where the working assumptions are no more than estimates, it is no good ground for objection that the step of converting those estimates into a money sum is itself very much a matter of approximation only. As Lord Reid observed in British Transport Commission v. Gourley [59] , "such damages can only be an estimate, often a very rough estimate, of the present value of his prospective loss". 1. [1956] A.C., at p. 212. Obvious virtues of this approach are its simplicity and its predictability, the latter assisting in the out-of-court settlement of claims. Its validity depends, of course, upon something like the existing conditions of inflation continuing on into the future. It is those conditions which provide the justification for regarding the advantage to a plaintiff of present payment as offset by his inability, in the face of inflation, to maintain over the years the purchasing power of the awarded capital sum and its interest-earning potential. However, in the present case the expert testimony, open to detailed criticism as it may be, at least provides an evidentiary basis that satisfies this requirement. Should the general climate of economic opinion as to the future change, some new approach may then be called for. But until that occurs it seems improbable that there will be any contest in individual cases as to the general prospect for the future being one of continued inflation at substantial rates. Accordingly there will in the general run of cases be no call for any expert economic evidence. This undiscounted approach will result in a substantial increase in the amount of damages to be awarded in individual cases, but if its result is not to overcompensate the plaintiff, this is in itself no objection to its adoption. It is no part of the judicial function to depress the level of awards on policy grounds: the courts have no mandate to entertain any such policy. If proper compensation is productive of what are seen as socially undesirable consequences, as by unduly increasing insurance premiums, this may be doing no more than redistributing the financial burden of road traumas and industrial accidents from the victims and the community at large to road users and industry. If such a redistribution is unacceptable, overseas models exist for legislative remedies in the shape of schemes of compensation which do not entail lump sum payments. So long as the question of compensation is left to the common law and to its remedy of a once and for all lump sum award the aim of such awards must remain fair compensation, leaving it to the legislature, if it sees fit, to intervene on policy grounds. A defect inherent in common law awards of lump sum damages affects this undiscounted approach no less than any other. It is due to the once and for all nature of such awards. Their assessment necessarily involves some prediction of the future and, once awarded, they remain unalterable however wrong that prediction may prove to be. No existing method of assessment can overcome this; only radical legislative intervention will suffice. It is appreciated that wage rates will not necessarily vary in step with changes in the purchasing power of money. They may reflect real increases in wages, outdistancing inflation, or may lag behind inflation, perhaps because of statutory controls or the consequences of a system of industrial arbitration. Yet in the course of these reasons I have tended to draw no distinction between future growth in nominal wages and general price inflation. Were such a distinction to be made it would require a different approach to loss of future earnings from that adopted where, for example, the cost of future medication or of the replacement of a prosthesis in the future was in question. These latter outgoings would respond only to general price inflation. The complexity which any such a distinction would introduce outweighs, I think, any loss of accuracy which is entailed in ignoring it. Taxation aspects in the light of Cullen v. Trappell. I have so far disregarded all question of income tax. As a consequence of Cullen v. Trappell [60] , and in some States as a result of prior legislative initiative, there has been a return to the situation as it had existed for some twenty years until altered by the majority decision in Atlas Tiles Ltd. v. Briers [61] : it is once more upon net after-tax loss of income that awards of damages for future economic loss are to be expressed. In addition, regard is to be had to the income tax notionally payable on the income which would be produced were the sum awarded in respect of future economic loss to be invested. 1. (1980) 146 C.L.R. 1. 2. (1978) 144 C.L.R. 202. The first of these two consequences of Cullen v. Trappell [60] is of no relevance in the present case because the plaintiff makes no claim to loss of future income. In other cases it will occasion no particular difficulty in any assessment of damages which allows for future inflation: all it means is that the relevant lost earnings will be those after rather than before tax. However, the position is otherwise in the case of the burden of income tax notionally payable on income notionally produced by investment of damages for future economic loss. 1. (1980) 146 C.L.R. 1. Taxation aspects as they affect the present appellant. There exist in the present case quite special considerations affecting any allowance that might otherwise properly be made on account of tax liability on income notionally derived from the investment of damages. They arise from the corporate character of the plaintiff, from the deductible nature of the instalment payments which it will have to make over the years and from the very different position in which the plaintiff finds itself from that of the usual plaintiff obtaining damages in a personal injuries claim. While it is true that, as was emphasized by Gibbs J. in Cullen v. Trappell [63] , it is a quite notional tax obligation that is to be taken into account, so that it will ordinarily be inappropriate to look at the actual post-trial circumstances of the plaintiff, the present is a special case in which the legal character of the plaintiff results in consequences which are out of the ordinary and which should receive consideration. 1. (1980) 146 C.L.R., at pp. 14-15. Because the plaintiff is a company the rates of income tax applicable to it are radically different from those applicable to an individual and, being a private company, they will be affected by whatever course it adopts over the years in relation to the distribution of its profits. Its position also differs very much from that of the ordinary plaintiff who is injured in an accident and accordingly recovers damages for future economic loss. Unlike most such individuals, the present plaintiff's income earning activities will remain unaffected; its business will presumably continue, and perhaps expand, in the future and may provide opportunities for investment of its award of damages in ways which make particularly unreal the ordinary process of calculating nominal tax on nominal income from the assumed investment of damages. Perhaps the most significant feature which sets apart the present plaintiff's position is that the whole of the instalment payments which it will be obliged to make over the years will constitute allowable deductions in its hands. Thus, while it is common ground on this appeal that the lump sum award will not form part of its assessable income, the future disbursements which this award is designed to meet will be wholly deductible. It is true that these disbursements will be spread over the remaining lifetime of the injured worker whereas it is in the early years of that period that income from the award of damages, if treated as invested, would attract substantial tax. Tax liability will diminish as, at an accelerating rate, capital is resorted to and income from it is reduced. However, there remains the substantial tax advantage to the plaintiff due to the deductible nature of the future payments which it will be obliged to make. The aggregation of these several circumstances not only makes it particularly difficult to arrive at any sum as an appropriate allowance for the tax consequences to the plaintiff; it casts very real doubts upon whether any allowance for income tax liability on notional income from invested damages is properly called for in this case. I would in the circumstances make no such allowance on account of that liability. Taxation aspects, more generally considered. In Cullen v. Trappell [64] Gibbs J. said, that because the process of assessment of damages for future economic loss assumes the investment of damages and the consequent receipt of income, a failure to take account of the liability to tax on that income "will produce a somewhat misleading result". In the undiscounted approach the assumption is retained that what is awarded will be invested and will produce income. It is the very capacity of the invested capital to produce income roughly equivalent to and offsetting the effects of inflation and increased productivity that is the justification for the undiscounted approach. Accordingly, notional tax liability on that income cannot be ignored. But whatever allowance is to be made for it should not be the occasion for the introduction of refined calculations such as have so troubled the assessment process in the past, while at the same time giving a spurious air of precision to a process founded upon a variety of imprecise assumptions or estimates. 1. (1980) 146 C.L.R., at p. 15. Such nice calculations are particularly inappropriate in the case of this notional liability to tax upon the income of invested damages. Quite apart from the fact that the plaintiff may never invest his damages, or may invest them with a view to capital gain rather than income potential, their unreality is emphasized by the inherent uncertainty which affects both future tax rates and the level of a plaintiff's future taxable income after due allowance for deductions to which he may be entitled, especially for medical and nursing expenses and the like. The possibility of some radical shift in tax policies reveals the shifting sands upon which such a calculation is built. If in the future direct taxes were to be reduced and indirect taxes increased the entire basis of calculation would be invalidated. In Cullen v. Trappell [65] Gibbs J. was at some pains to point out why it was that "the court cannot attain complete mathematical accuracy, and is not bound to engage in complicated exercises in an attempt to do so". To make all the assumptions necessary for such exercises, any one or more of which may prove wrong, each instance of error perhaps being cumulative upon others, and then to undertake precise calculations of tax consequences seems inherently unsound. I would repeat what I said, in a somewhat different context, in Atlas Tiles Ltd. v. Briers [66] : that it is "obviously illusory to seek for precise accuracy in an ancillary factor such as tax, itself dependent for its quantum upon an uncertain estimate of earnings". 1. (1980) 146 C.L.R., at p. 16. 2. (1978) 144 C.L.R., at p. 231. In Taylor v. O'Connor [67] their Lordships in effect took account of tax on invested income by adjustment of the discount for present payment, although this was somewhat obscured by the operation of the traditional English multiplier system of assessment. It is just such a broad brush approach that appears to me to be necessary in ordinary personal injuries cases. 1. [1971] A.C. 115. In the United Kingdom, in the context of a very different income tax system, the Pearson Report, in pars. 646 to 708, recommends a system of several distinct tables of multipliers based upon varying discount rates. The need for these varying rates of discount was seen by the Commission to lie in the differing tax situations of particular plaintiffs, affected both by the levels of their incomes and their marital status. To apply the same discount rates regardless of the varying tax situations of plaintiffs was regarded as leading to marked inequalities: pars. 678-680. The rates of discount adopted sought to reflect "the real rate of return after tax, that is, approximately the net rate of return after tax less the rate of inflation" (par. 677). Illustrative discounts appearing in Table 11 (par 686), range from a positive discount of 1 per cent to a negative discount (having the effect of actually increasing the damages awarded) of — 2 per cent. Without expert evidence and the aid of appropriate calculations this Court is in no position to formulate any such comprehensive regimen for the adjustment of awards of damages to take account of tax. But the general approach of the Pearson Report appears to be suitable in Australian conditions. This suggests that tables might be developed by which the initial undiscounted approach to the assessment of damages may be appropriately adjusted to take account, albeit only in broad terms, of tax on notionally invested awards. It would, initially, be for plaintiffs, in whose favour such adjustments would operate, to introduce such tables into evidence. Once accepted by courts as a basis for adjusting the basic undiscounted approach in the direction of modest negative discount rates, such tables would thereafter provide an efficient tool, leading both to predictability of outcome and to such accuracy in attaining proper levels of compenstion as the subject matter permits. "Nursing" in s. 10 of the Act. There remains one question of statutory interpretation, a short point but one which carries with it important financial consequences. It concerns the meaning of nursing in s. 10 (2) of the Workers' Compensation Act. I set out below the terms of s. 10 (1), of par. (c) of the definition of Medical treatment in s. 10 (2) and of the definition of Hospital treatment in that sub-section: (1) Where as a result of an injury received by a worker — (a) it is reasonably necessary that any medical or hospital treatment be afforded, or any ambulance service rendered, to him; or (b) it is appropriate that any such treatment, being treatment by way of rehabilitation, be afforded to him, his employer shall, subject to and to the extent provided by this section, be liable to pay, in addition to any compensation otherwise provided, the cost of that treatment or service. (2) "Medical treatment" includes — (c) any nursing, medicines, medical or surgical supplies or curative apparatus, supplied or provided for him otherwise than as a patient at a hospital; and but does not include ambulance service or hospital treatment. "Hospital treatment" means treatment, including treatment by way of rehabilitation, at any hospital or at any rehabilitation centre conducted by a hospital, and includes the maintenance of the worker as a patient at the hospital or rehabilitation centre, and the provision or supply by the hospital, at the hospital or rehabilitation centre, of nursing attendance, medicines, medical or surgical supplies or other curative apparatus, and any other ancillary service, and also includes the cost to the worker of any fares and travelling expenses necessarily and reasonably incurred by him in obtaining any such treatment, but does not include ambulance service. By a majority the Court of Appeal held that the provision to the injured worker of assistance in the form of bed-making, cleaning, laundering and cooking did not constitute "nursing" and was therefore not within the definition of Medical treatment in s. 10 (2). Accordingly the agreed cost of this assistance, $60 per week, was not payable out of the Fund, was not recoverable by it from the plaintiff and should not be included in any assessment of the plaintiff's damages. Only if that assistance ought in fact to be regarded as "nursing" within par. (c) of the definition of "Medical treatment" will the weekly sum of $60 enter into the assessment of damages. "Nursing", like the verb "to nurse", may be used with various shades of meaning. All appear to involve the attending to the personal needs of another who lacks the capacity to care for himself, whether because of illness, extreme age or youth, or other physical or mental handicap. While often applied to the care of the young and of the sick, "nursing" is clearly not to be so confined; it includes the care of the healthy but incapacitated: Public Trustee v. McKay [68] , per Turner J. and Wallbridge v. Dorset County Council [69] . Nor can it be confined by reference to any particular training or qualifications possessed by the person who undertakes it. It is, however, confined to caring for the person of the recipient of such care and for his immediate surroundings. This may itself involve activities carried on at some distance from him, such as food preparation and laundering. Its extent in any given case will be determined by the extent of incapacity. For example, for the bedridden it will include washing, feeding, dressing and undressing him and caring for the state of his immediate surroundings, his bed and bedroom and all that is necessary for those purposes, including food preparation and laundering. 1. [1969] N.Z.L.R. 995, at p. 1006. 2. [1954] 1 Ch. 659, at pp. 667-668. That the person being cared for is not bed-ridden but, like the present injured worker, can move freely about in his wheel-chair may not diminish the extent of his need for nursing but will extend the area of what I have called his immediate surroundings. This may permit it to be said that general house-cleaning, rather than the mere cleaning of his bedroom, forms a part of his nursing. While relevant evidence in this case was far from detailed it would, I think, support the view that his incapacity prevents him from undertaking general housekeeping, such as the cooking of his food or the washing of his bed linen and clothing; he at least needs some help in cooking, cleaning and washing even if these tasks are not wholly beyond him. I would regard all such help as included in the nursing of him. A particular statutory context may, of course, involve placing some particular restriction upon the meaning of "nursing". It will usually do so only by reference to the particular cause of incapacity, infancy, old age and so on, rather than to the particular services to be performed, which will vary with the extent of incapacity of each person being cared for. Here such a restrictive context exists; s. 10 (1) will confine "nursing" to caring for the needs occasioned by the "result of an injury" to the worker which makes it "reasonably necessary" that nursing be afforded. The more immediate context of par. (c) of the definition of "Medical treatment" does not appear to me to involve any further restrictions upon meaning. It contains no other verbal substantives like "nursing" but only a series of nouns, "medicines, medical or surgical supplies or curative apparatus", which do no more than reflect that restriction of meaning already referred to and which flows from the terms of s. 10 (1). They say nothing as to the range of services which "nursing" may encompass. The definition of Hospital treatment in s. 10 (2) also contains the word "nursing", although it there appears as part of the composite phrase "nursing attendance". Because preceded by a reference to "the maintenance of the worker or a patient at the hospital", it suggests that "nursing attendance" is distinct from a patient's "maintenance". But this does not, I think, throw any light upon the terms of the definition of "Medical treatment". The definition of "Hospital treatment" is drafted in terms of the various services which a hospital may provide and hence, no doubt, the use of the term "nursing attendance". The element of what might be called board and lodging which is provided to a hospitalised patient accordingly receives specific mention, which would be inappropriate in the context of the definition of "Medical treatment". Given, then, the general meaning of nursing which I have described, there do not appear to me to be any implications arising from immediate statutory context which would confine "nursing" to some only of the particular nursing services which might be rendered to a paraplegic. It follows that, with due respect to the views of the Court of Appeal expressed both in this case and again in Thomas v. Ferguson Transformers Pty. Ltd. [70] , I am led to a contrary conclusion. I regard those services described in the evidence as required by the injured worker as falling within "nursing" in the definition of Medical treatment in s. 10 (2). 1. [1979] 1 N.S.W.L.R. 216, at p. 221. Conclusions. The award of damages in this case should, for the reasons indicated above, proceed upon what I have described as the undiscounted approach. It will, in the particular circumstances of this case, make no allowance for income tax upon notional income from invested damages. It should include an allowance for nursing services, at the agreed rate of $60 per week. It will otherwise proceed upon the same footing as that adopted by the majority in the Court of Appeal as, for instance, in allowing for a possible increase in the rate of payment to the injured worker should he marry and his wife prove to be wholly dependent upon him. The appropriate calculation will be as follows: 1. Amount paid or liable to be paid by the plaintiff to the Uninsured Liability Fund up to the date of the conclusion of the hearing at first instance. $54,528.39 2. Legal costs and fees awarded by the trial judge. 3,614.26 3. Amounts paid or liable to be paid between the hearing date and the date of judgment, 4th April 1977, being 23 weeks. Compensation $64.00 per week Nursing 60.00 per week Hospital and medical expenses 8.40 per week $132.40 per week 3,045.20 4. Amounts paid or liable to be paid between the date of judgment, 4th April 1977, and the coming into force of Workers' Compensation (Rates) Amendment Act 1977 on 9th December 1977, being 36 weeks. Compensation $64.00 per week Nursing 60.00 per week Hospital and medical expenses 8.40 per week $132.40 per week 4,766.40 5. Pursuant to the Workers' Compensation (Rates) Amendment Act 1977, from 9th December 1977, compensation was payable to the injured worker, at a rate of $83.00 per week. Yeldham J. found that from 16th December 1977, the injured worker had a life expectancy of 34.2 years. (a) Amounts payable from 16th December 1977 and for the remainder of the injured worker's estimated life: Compensation $83.00 per week Nursing 60.00 per week Hospital and medical expenses 8.40 per week $151.40 per week for a period of 34.2 years 269,249.79 (b) Amounts paid or payable between 9th December 1977 and 16th December 1977. 189.14 6. The injured worker may marry and his wife may become wholly dependent on him. This contingency may be broadly met by adopting the assessment for it made by Reynolds J.A. in the Court of Appeal. Consistent with the undiscounted approach however, his discounting by five per cent tables is eliminated and the allowance for this contingency proportionately increased. 6,236.25 7. Interest awarded by the Court of Appeal 3,500.00 $345,129.40 This sum of $345,129.40 is considerably less than the total amount awarded as a result of the order of the Court of Appeal, $479,193.65, and this despite the fact that I have formed a view favourable to the plaintiff on the question of "nursing" expenses. This is, in essence, due to the fact that although the Court of Appeal's order took account of future inflation in reckoning the amount of future loss it applied a discount rate of only 5 per cent, a rate set well below inflation-affected market rates of interest and which, as Lord Diplock has pointed out, will of itself take some rough and ready account of inflation. The award, so calculated, accordingly involved a substantial measure of double allowance for the effects of future inflation. The outcome of this appeal is that the plaintiff-appellant succeeds as to the issue of nursing expenses. The defendant-cross appellant fails in its arguments as to the proper mode of assessment; however in the outcome the amount awarded is reduced and to that extent the cross-appeal succeeds. Mason J. These appeals and cross-appeals arise originally out of an action brought by Pennant Hills Restaurants Pty. Ltd. ("Pennant") against its insurance broker, Barrell Insurances Pty. Ltd. ("Barrell") for damages for professional negligence in failing to keep current a workers' compensation policy for Pennant. In the period the policy was not current a worker employed by Pennant was seriously injured on his way to work. Woodward J. found for Pennant and awarded damages of $231,187.85, on 4th April 1977. Both Pennant and Barrell appealed to the Court of Appeal on the question of damages only. The Court of Appeal delivered an interlocutory judgment on 16th December 1977 and by majority entered a judgment for Pennant for $146,093.65 with $3,500 interest, together with a further sum to be assessed to provide for the effect of the indexing of workers' compensation payments. This additional determination was remitted to Yeldham J. in the Common Law Division who on 7th September 1978 assessed the additional sum to be awarded to Pennant at $329,600. On 22nd September 1978 the Court of Appeal varied its previous order of 16th December 1977 with reference to interest and costs and both parties appealed to this Court on the basis that the assessment by Yeldham J. had taken effect without any further adoption of it by the Court of Appeal. Pennant also sought leave pursuant to s. 101 (2) (e) of the Supreme Court Act to appeal to the Court of Appeal against the assessment of Yeldham J. on the footing that the finding of Yeldham J. was an interlocutory order. On 3rd October 1978 the Court of Appeal refused leave. Thereafter both Pennant and Barrell appealed to this Court on the basis that the decision of the Court of Appeal on 3rd October 1978 was a final judgment or rendered the earlier decision of the Court of Appeal, together with the findings of Yeldham J., a final judgment of the Court of Appeal. The injured worker had claimed compensation in the Workers' Compensation Commission and on 19th December 1972 an award was made by the Commission in favour of the worker against Pennant for periodic payments and medical and hospital expenses under the Workers' Compensation Act 1926 N.S.W., as amended, ("the Act"). Section 18C of the Act establishes an Uninsured Liability Scheme and the payments to the worker were made under this Scheme. Pursuant to s. 18C (6), Pennant is liable to reimburse the Uninsured Liability Fund for its periodic expenditure under the Scheme in relation to the worker. The damages Pennant is seeking to recover from Barrell are damages in a lump sum to compensate it for the liability to periodically reimburse the Workers' Compensation Commission for its payments to the worker. These payments are to continue for the rest of the life of the worker. Pennant would not be under this liability if it had not been for the failure of Barrell to keep current the worker's compensation policy. At the time of the trial it was found that the worker had a life expectancy of between 33 and 34 years. This finding is not challenged. The amount of $231,187.85 awarded by the trial judge was made up as follows: 1. Amounts paid or liable to be paid by the plaintiff to the Commission up to the conclusion of the hearing $58,142.65 2. Amounts paid or liable to be paid between the hearing and the date of judgment, namely 4th April 1977: Compensation $64.00 per week Nursing attention in the form of domestic assistance 60.00 per week Hospital and medical expenses 8.40 per week $132.40 per week 23 weeks at $132.40 3,045.20 3.39 weeks at $132.40 from 4th April 1977 to 31st December 1977 $5,163.60 5,000.00 Discounted to 4. For the future on the basis of periodical payments of $85.00 per week (in lieu of $64.00), plus $60.00 per week for nursing and $8.40 per week for hospital and medical expenses, i.e. a total of $153.40 per week. Adopting a discount rate of 5 per cent for 30 years $122,566.00 for 35 years $132,543.00 Say 125,000.00 5. Taxation allowance for incidence of tax due to investments made as a hedge against inflation 25,000.00 6. Pre-judgment interest 15,000.00 Total $231,187.85 At the time, the maximum weekly compensation sum (which the worker is admittedly entitled to) under s. 9 of the Act was $64. On the basis of evidence of movements in weekly payments, Woodward J. estimated that the average weekly rate for the next thirty to thirty-five years would be $85 per week and the total amount per week for thirty to thirty-five years was then discounted at a rate of 5 per cent. The Court of Appeal varied this award. Just before the decision of the Court of Appeal was handed down the Act was amended by Act No. 124 of 1977 ("the Rates Act") to provide for indexation of weekly compensation payments. Section 9A provides that the various specified rates of compensation, which are termed "adjustment amounts" and which include the amount of $64 per week, should be adjusted on 1st April and 1st October each year according to rises in the period in the "weighted average minimum weekly rate, in dollars, payable for all industry groups to adult males for a full week's work (excluding overtime), as prescribed in awards, determinations and collective agreements" (s. 9A (1) (ii)). The section thus picks up a particular index which is published by the Australian Statistician. A "base rate" of $144.57 is specified (s. 9A (1)) and the adjustment amounts are adjusted by the same proportion each time as the proportion the latest weekly rate in the index bears to the base rate (s. 9A (2)). This figure is to be published in the Gazette (s. 9A (4)). The Court of Appeal held that the trial judge had not been entitled to make any assumption that the maximum weekly compensation rate payable under the Act would be altered in any particular manner. However, the Court of Appeal was of the opinion that the Rates Act provided for indexation of weekly compensation rates and that it was operative on the award. Under the 1977 indexation the weekly figure of $64 had been increased to $83. On this basis Reynolds J.A. assessed Pennant's future economic loss at $76,951, being the figure of $83 per week plus $8.40 per week for hospital and medical expenses (the figure selected by the trial judge) for a period of thirty-four years (the remaining life expectancy of the worker) discounted at a rate of 5 per cent (the discount rate adopted by the trial judge). Hutley and Mahoney JJ.A. agreed with the assessment of future economic loss at $76,951 based on the weekly figure of $83, but were of the view that some additional allowance should be made for the effect of future indexation taking the base figure beyond $83 per week in calculating Pennant's future economic liability, and that the issue should be remitted to the Common Law Division for assessment. The Court of Appeal sustained the use of a discount rate of 5 per cent. By majority it set aside the allowance of $60 per week for nursing. The award of $25,000 for income tax compensation was disallowed and the award of pre-judgment interest was reduced to $3,500. Amounts of $6,000 and $5,000 were included to cover the contingency of marriage and future nursing expenditure. The Court of Appeal therefore arrived at a verdict of $146,093.65 together with an additional sum to be assessed to provide for increases in the weekly sum of $83 which Pennant might be called on to pay pursuant to s. 9A and pre-judgment interest in the amount of $3,500. The sum of $146,093.65 was arrived at as follows: Loss to date of trial $58,142.65 Future loss based at weekly payments of $83 and hospital expenses of $8.40 76,951.00 Contingency of marriage 6,000.00 Future nursing 5,000.00 $146,093.65 The question of what further sum should be allowed to compensate for the effect of indexation was heard by Yeldham J. Expert economic and actuarial evidence was placed before his Honour. Dr. R. V. Horn from the School of Economics, University of N.S.W., presented a report to the Court on future economic trends and Dr. Horn's conclusions were accepted by Yeldham J. His Honour concluded that "the wage rate index over the period of 34.2 years from 16th December 1977 will be at the average rate of ten per cent per annum". Using a discount rate of 4 per cent per annum his Honour calculated the "further sum which should be included in [Pennant's] verdict to provide for the present value of increases in the weekly sum of $83 which [Pennant] may be called upon to pay pursuant to s. 9A of the Workers' Compensation Act, 1926 " as $329,600. The sum awarded to Pennant was thus $479,193.65 made up as follows: (i) Amount awarded by trial judge for loss to date of trial $58,142.65 (ii) Future loss based on weekly compensation payments of $83 and hospital expenses of $8.40 above 76,951.00 (iii) Contingency of marriage 6,000.00 (iv) Future nursing 5,000.00 (v) Further sum for future increases 329,600.00 (vi) Pre-judgment interest 3,500.00 $479,193.65 The primary issue before this Court is whether the Court of Appeal was correct in awarding an additional sum, based on projections of the weighted average weekly minimum rate, to compensate Pennant for a liability to reimburse the Uninsured Liability Fund for future increases in weekly compensation payments. If the Court of Appeal was correct in principle in awarding such an additional amount, the question is, how should this amount best be calculated? It was agreed by the parties that the question of damages to be awarded to Pennant should be considered on a post Rates Act basis. Barrell submits that such projections involve predictions as to the future rate of inflation and that this is inconsistent with the reasoning and this Court's decision in O'Brien v. McKean [71] . According to Barrell, O'Brien is authority for the proposition that in the assessment of damages no account is to be taken of future inflation. The correctness of this proposition, Barrell submits, is confirmed by recent decisions of the House of Lords which, with one exception, are cases like O'Brien , dealing with the assessment of damages for personal injury. 1. (1968) 118 C.L.R. 540. Whether the courts in assessing damages for personal injury should take account of inflation and, if so, by what means, has been a vexed question. The recent resurgence of high rates of inflation has given the question a new vitality. Essentially the problem is one of arriving at a presently payable lump sum award measured by reference to the future earnings which the plaintiff might be expected to receive but for his injury, when the amount of the future earnings will probably be affected by inflation and there is a need to discount that amount so as to allow for the immediate receipt of the lump sum award. There is a separate, but related, problem in connexion with future payments or expenses. Different solutions have been proposed, some legislative (see J. G. Fleming, The Impact of Inflation on Tort Compensation Amerian Jo. of Comparative Law, vol. 26 (1978), p. 51; the Report of the Royal Commission on Civil Liability and Compensation for Personal Injury ("the Pearson Report") (1978)), some judicial. Of the judicial solutions, apart from those adopted in Australia and the United Kingdom, the principal alternatives are two methods designed to give effect to what has been described as the real rate of interest. They are: (a) the application of a discount rate which reflects the difference between the long-term investment rate and the predicted inflation rate, the difference throwing up the real rate of interest. This approach has been adopted by the Supreme Court of Canada (see Andrews v. Grand & Toy Alberta Ltd. [72] ;Thornton v. Board of School Trustees of School District No. 57 (Prince George) [73] ; Arnold v. Teno [74] ; Keizer v. Hanna [75] ; and (b) the suggested application of such a discount rate as represents the steady difference between the nominal interest rate and the inflation rate as established by research based on past experience. 1. [1978] 1 W.W.R. 577. 2. [1978] 1 W.W.R. 607. 3. (1978) 83 D.L.R. (3d) 609. 4. (1978) 82 D.L.R. (3d) 449. However, my immediate concern is to examine the effect of the decision in O'Brien [76] and the effect of the recent English decisions. I have a problem in identifying the precise principle on which O'Brien was decided. There the Court reviewed a lump sum award of general damages involving a substantial loss of earning capacity in consequence of personal injury. Taylor and Menzies JJ. took the broad view that for most purposes the law disregards changes in the purchasing power of money and that this general principle applies to the assessment of damages for personal injury [77] . McTiernan J. thought a forecast of the trend of the purchasing power of the currency is too speculative and difficult to justify in its introduction into the assessment of damages for personal injury [78] . Windeyer J. seems to have been of the same opinion, though acknowledging that it was legitimate for juries to have regard to inflation [1] . 1. (1968) 118 C.L.R. 540. 2. (1968) 118 C.L.R., at p. 553. 3. (1968) 118 C.L.R., at p. 552. 4. (1968) 118 C.L.R., at p. 560. Barwick C.J. drew a distinction between compensation for loss of future earning capacity and compensation for future expenditure. He thought that changes in the amounts paid for the exercise of lost or impaired earning capacity which would be due solely to changes in the purchasing power of money ought not to be considered relevant to the assessment of compensation for lost earning capacity. His Honour based this conclusion on the fact that the plaintiff exchanged his earning capacity for a capital fund with which he was free to protect himself from the consequences of inflation by such means as he might choose. He said [2] : I do not think the question concerns the discount rate which is commonly used in determining the present value of a periodic sum paid regularly over a period of years. In my opinion, so far as the lost or diminished capacity to earn is concerned, the exclusion of any consideration of further depreciation in the purchasing power of money in the future should be founded directly upon the reason that the injured person is being presently compensated for the loss which has already occurred by the payment of money which he is free to employ or invest in any way he may see fit. It would follow that, in my opinion, evidence directed to establishing the further decline in purchasing power of the currency in the future in relation to compensation for lost or diminished earning capacity is not admissible. 1. (1968) 118 C.L.R., at pp. 547-548. The Chief Justice thought that different considerations apply to the assessment of compensation for future expenditure in relation to personal injuries. He conceded that, if it could be established that the cost of the goods and services will probably increase in the relevant period, an allowance could or should be made for the increased cost, including the diminished purchasing power of money, though it would be necessary to take account of the fact that the plaintiff was in possession of a fund investible in "securities which could provide a hedge against future inflation". However, the Chief Justice considered that no such allowance should be made for two reasons. First, the ascertainment of the cost of goods and services to be provided in the future was speculative and not susceptible of "solid proof". Secondly, historical experience did not justify the conclusion that the increase in the cost of goods and services over a substantial period would be as much as 3 per cent p.a. and that the investment return could very well exceed the rate of increase in the cost [3] . 1. (1968) 118 C.L.R., at p. 550. I regard O'Brien [4] as deciding that inflation should not be taken into account in assessing damages for personal injury, whether in respect of lost earning capacity or future expenditure. Whether the decision prevents the application in such cases of a low discount rate equal to the interest rate appropriate to a stable currency in lieu of the high rates of interest which prevail in an era of inflation is another question which I leave for later consideration. 1. (1968) 118 C.L.R. 540. At this stage it is worth making the point that the O'Brien principle was based in some respects on factors which have been falsified by our experience since 1968. First, the case for saying that there will be a significant decline in the purchasing power of money over a given period in the future is very much stronger now than it was then. We cannot ignore the fact that wage rates have increased continuously for almost fifty years, the rate of increase having become more pronounced in recent times. The assumption that the rate of increase in the cost of goods and services will not be as much as 3 per cent per annum has been exploded, as has the conviction that investment in "equities" will render a capital fund immune to inflation. Secondly, the fact that we now live in an era of rapid inflation is in itself a reason for departing from the so-called principle that the law disregards changes in the purchasing power of money. If it ever was a principle, it was very much the product of an age in which currency had a stable value in terms of purchasing power. Although certain remarks of Barwick C.J. in O'Brien have been cited and applied (see Mitchell v. Mulholland [No. 2] [5] ), the basic English approach involves one element which was not discussed in O'Brien . The United Kingdom method of assessing damages for personal injury differs from that adopted here, though it is intended to lead to a similar arithmetical result, as Glass J.A. explained in Beneke v. Franklin [6] . Both methods seek to make allowance for the present receipt in a lump sum award of compensation for loss which is measured by reference to earnings which would, but for the plaintiff's injury, have been received in the future. In both methods the allowance is made by reference to the interest-earning capacity of the sum to be awarded to the plaintiff. In the United Kingdom this factor is taken into account in the selection of an appropriate multiplier. Here, it is the subject of a discount rate applied to the amount which would otherwise be awarded. 1. [1972] 1 Q.B. 65, at p. 79. 2. [1975] 1 N.S.W.L.R. 571, at p. 592. The development of the current approach taken by the House of Lords to the element of inflation in the calculation of damages for personal injury may be traced through such cases as Mallett v. McMonagle [7] ; Taylor v. O'Connor [8] ; Cookson v. Knowles [9] to Lim Poh Choo v. Camden and Islington Area Health Authority [10] . It will be sufficient if I refer to two passages from the speeches in Cookson and to some observations of Lord Scarman in Lim. In Cookson [11] , Lord Diplock said: In times of stable currency the multipliers that were used by judges were appropriate to interest rates of 4 per cent to 5 per cent. the likelihood of continuing inflation after the date of trial should not affect either the figure for the dependency or the multiplier used. Lord Fraser of Tullybelton was of the same opinion and concluded that it would be wrong to increase damages by making "a further specific allowance for future inflation" [12] . 1. [1970] A.C. 166. 2. [1971] A.C. 115. 3. [1979] A.C. 556. 4. [1980] A.C. 174. 5. [1979] A.C., at p. 571. 6. [1979] A.C., at p. 577. In Lim [13] , Lord Scarman (with whom the other members of the House agreed), after referring to these comments, said: The law appears to me to be now settled that only in exceptional cases, where justice can be shown to require it, will the risk of future inflation be brought into account in the assessment of damages for future loss. Lump sum compensation cannot be a perfect compensation for the future. An attempt to build into it a protection against future inflation is seeking after a perfection which is beyond the inherent limitations of the system. While there is wisdom in Lord Reid's comment (Taylor v. O'Connor [14] ) that it would be unrealistic to refuse to take inflation into account at all, the better course in the great majority of cases is to disregard it.— 1. [1980] A.C., at p. 193. 2. [1971] A.C., at p. 130. I do not regard Lord Scarman's remarks as excluding the selection of an appropriate multiplier by reference to a real or low interest rate which reflects the rate payable in times of stable currency. His Lordship did not express any disagreement with the conventional English practice as explained in Cookson . His remarks were directed against the specific inclusion of an allowance for inflation in other than exceptional cases. In substance the United Kingdom approach to the assessment of damages for personal injury is to ignore the element of inflation and to assume the existence of two of the characteristics of a stable economy, (a) a continuation of existing nominal wage rates, and (b) low interest rates appropriate to such an economy. It has been thought that by taking an approach which is adapted to the fiction that we have a stable economy the plaintiff's damages will be fairly assessed, it being for the plaintiff to counter the effects of inflation as best he can by pursuing a suitable investment policy. The use of a low interest rate in the selection of the multiplier excludes one of the principal characteristics of inflation, the prevalence of high rates of interest. The consequence of this is that, if inflation continues, the plaintiff will be able to invest his verdict at a higher rate of interest than that on which the verdict was based. Conversely, if inflation does not continue, the verdict will have been calculated on an interest rate which will approximate the prevailing rate. Because the high rates of interest are themselves a product of inflation, it would not be right to take them into account and thereby diminish the amount of the plaintiff's verdict. To a substantial extent they compensate the lender for the erosion of the value of his capital. Certainly they significantly exceed the real rate of interest. In Beneke [15] the New South Wales Court of Appeal took a view similar to that expressed in Cookson [16] in deciding that it was wrong to apply a discount rate based on prevailing high interest rates which reflected the factor of inflation. Moffitt P. said that to apply the current rate of interest would be to "bring future inflation to account and not disregard it" [17] . Neither Moffitt P. nor Glass J.A. thought that O'Brien [18] decided otherwise. I agree. 1. [1975] 1 N.S.W.L.R. 571. 2. [1979] A.C. 556. 3. [1975] 1 N.S.W.L.R. at p. 587. 4. (1968) 118 C.L.R. 540. Thus, in both Australia and the United Kingdom, damages for personal injury are assessed without taking inflation into account. Instead, the characteristics of a stable economy are assumed, thereby enabling the plaintiff to counter inflation by pursuing a suitable investment policy and investing at current higher rates of interest if he so chooses. In calculating an amount which was designed to reflect estimated future increases in average weekly earnings, Yeldham J., having found that there would be an increase of the order of 10 per cent per annum, applied a discount rate of 5 per cent. He considered that he was bound to apply this rate because he thought that each member of the Court of Appeal had held it to be appropriate. In this he was mistaken. Only Mahoney J.A. had held this rate to be appropriate in relation to future increases. The question was not examined by the other members of the Court. Mahoney J.A. qualified his acceptance of the rate by saying that no convincing reason had been advanced for adopting any other rate. With respect, I should have thought that there was a convincing reason for adopting a higher discount rate once it was decided to take account directly of future increases in average weekly earnings. As we have seen, the adoption of a low discount rate, say 5 per cent, is an element in the English and Australian approach which assumes the existence of a stable non-inflationary economy and offers the plaintiff the compensatory advantage of investing his verdict at higher prevailing rates of interest if inflation should continue. The adoption of an artificially low discount rate is quite inappropriate once the court proceeds to take inflation directly into account by estimating future wage rates. To apply an artificially low rate then is to give the plaintiff an additional advantage which in my view necessarily results in over-compensation. Even so, the fundamental question remains: Should the damages in this case be assessed by estimating future increases in wages or should the likelihood of such increases be ignored and a low discount rate applied? The reasons which underlie the decisions in O'Brien and Lim [19] are opposed to the first approach. The only distinction which this case presents is that here we have a statute which increases the amount of the liability at regular six-monthly intervals by reference to a statistical index. But the inevitable problems of determining whether inflation will continue, and at what rate, over a long future period remain with us. It was the presence of those problems and the difficulties in answering them that led to the decisions in O'Brien and Lim. 1. [1980] A.C. 174. Some of the assumptions on which O'Brien [20] was based were, as we have seen, unsound. None the less it is still correct to say that prediction of the inflation rate over the next thirty to forty years is a hazardous and uncertain enterprise and that the nature of the topic is such that it is incapable of being rendered certain by expert evidence or existing statistical materials. So much is, I think, demonstrated by the evidence of Dr. Horn, which was not contradicted. Moreover, it is undesirable that the assessment of damages for personal injury, already a complex undertaking, should be further complicated by an inquiry into future rates of inflation. Essentially it is for these reasons that I regard the attempt to calculate future wage increases as being too speculative and difficult to justify. Instead, I favour the calculation of damages, without attempting any such projection, by applying a low discount rate, if the application of such a rate be appropriate to the case. 1. (1968) 118 C.L.R. 540. The critical question, then, is as to the selection of the discount rate. I do not favour the adoption of a rate which reflects the difference between the long-term investment rate and the projected inflation rate (the wage rate in this case) because once again it is an exercise which calls for prediction of the unpredictable. Consequently, I would not follow the recent decisions of the Supreme Court of Canada. It has been suggested by some who support the adoption of a low discount rate that we should have regard to current movements in the purchasing power of money and that some speculation should be made about future inflation — see, for example, Beneke [21] . But this is to admit through another door the speculative elements whose direct admission has been prohibited. 1. [1975] 1 N.S.W.L.R., at pp. 593, 608. We are left then with the question whether the selection of a discount rate should be based on rates of interest appropriate to a stable economy or on the real rate of interest as established by evidence of past experience. For the moment I put to one side the element of taxation on the investment income produced by the verdict. The adoption of a low rate of interest reflective of a stable economy is supported by the recent United Kingdom decisions. However, its weakness is that it also reflects an element of inflation for even in times of a stable economy inflation has generally proceeded at 2 to 3 per cent. The adoption of a 4 to 5 per cent interest rate appropriate to such an economy therefore throws up an element of inflation, though not at such an unacceptable level as that reflected by current interest rates. The adoption of the real interest rate would have an obvious advantage. Unfortunately it also has disadvantages. One such disadvantage is that the statistical and other information relating to past experience which is available to me does not establish that there is a steady real rate of interest in Australia or that, if there be such a rate, what it happens to be. The parties did not direct their attention to this question and the Court has therefore not had the benefit of evidence and expert opinion upon the matter. In this unsatisfactory situation I would adopt a discount rate of 2 per cent as a fair approach to the problem raised by this case—one which does more justice to the plaintiff than the adoption of a 4 per cent or 5 per cent rate appropriate to a stable economy reflecting a moderate level of inflation. In expressing this view I am not to be taken as saying that it should necessarily apply to all personal injury cases. I am conscious of the special nature of this case and the imperfect materials which have been made available to us. Accordingly, subject to an examination of the question of taxation, I would apply a discount rate of 2 per cent. Pennant does not argue that the verdict itself attracts income tax. But it does contend that we should take account of the circumstance that the income produced by the verdict, when invested, will be subject to tax and this in itself justifies the application of no discount rate or of a lower discount rate than might otherwise be chosen. The recent majority decision of this Court in Cullen v. Trappell [1] establishes authoritatively that it is right to take into account the tax which will be payable on the income produced by the verdict, if it is invested. Quite plainly the award of damages will be inadequate in this case if, after tax on the investment income, the capital sum provided does not fully indemnify Pennant against the payments which it is bound to make. After all, the amount has been calculated on the footing that Pennant will be able to have recourse to the income as well as the corpus of the verdict. 1. (1980) 146 C.L.R. 1. However, there are difficulties in estimating the tax which Pennant will bear on the investment income. Although the ordinary rate of company tax is 47½ per cent, Pennant might so conduct its affairs or arrange its investments that its investment income attracts no tax. On the other hand, Pennant, by deciding not to distribute profits by way of dividend, may expose itself to a liability to pay undistributed profits tax. It is impossible to evaluate accurately all these contingencies. Moreover, there is the added complication that Pennant may be entitled to treat the payments which it makes under s. 18C as allowable deductions from its assessable income. In these circumstances I would make no adjustment to the discount rate of 2 per cent on account of taxation. On the existing materials I would fix the amount of damages to cover future loss for 34.2 years at $118,000. The final question is whether domestic assistance constitutes "nursing" within the meaning of s. 10 of the Act. Woodward J. held that it did. The majority in the Court of Appeal disagreed, Mahoney J.A. dissenting. Section 10 (2) (c) provides: For the purposes of this section— "Medical treatment" includes— (c) Any nursing, medicines, medical or surgical supplies or curative apparatus, supplied or provided for him [an injured worker] otherwise than as a patient at a hospital. The assistance here in question was described by the worker as assistance in activities such as bed-making, cleaning, washing, laundry and cooking. In a suitable context "nursing" may denote a wider range of activities than those undertaken by a nurse in caring for a patient who suffers from illness or injury (Wallbridge v. Dorset County Council [2] ). There, s. 206 of the Public Health Act, 1936 referred to "the nursing and maintenance of a child". Here, we have a very different context. "Nursing" occurs in a setting of medical services, rendered outside a hospital, not in a context of general maintenance. Consequently, in s. 10 (2) (c) the word denotes care and attention to a patient which is designed to relieve or remedy the illness or injury from which he suffers. The domestic assistance needed by the injured worker here stands outside this conception. 1. [1954] Ch. 659. In the result I would dismiss the appeal by Pennant and allow the cross-appeal by Barrell. The sum of $118,000 should be substituted for the sums of $76,951 (being part of the amount of $146,093.65 mentioned in the order of the Court of Appeal of 22nd September 1978) and $329,600 (mentioned in the same order). Murphy J. This case concerns the failure of the defendant insurance broker to effect the proper insurance against worker's compensation liability of the plaintiff. A worker employed by the plaintiff became totally and permanently incapacitated and the plaintiff is liable under the Workers' Compensation Act 1926 N.S.W., as amended, which provides for periodic payments of sums which are to be adjusted in accordance with the index prepared by the Commonwealth Statistician of the movement of wages. The problem is to assess the damage suffered by the plaintiff. The obvious way to assess the damage would be to ascertain the amount it would cost the plaintiff to relieve it of this liability. There is a large insurance industry in Australia. The defendant was an insurance broker and, no doubt, it would be possible to obtain expert evidence on what those in the industry would regard as an appropriate sum for the assumption of such a liability. This would include an element for the administrative expenses and the profits of an insurer who would undertake to assume such a liability, but, as I see it, these would properly be included. The computation of such liabilities is an every day task for those who are concerned with the transfer of such liabilities on purchase or merger of insurance businesses. However, no attempt was made by the plaintiff or the defendant to ad??uce such evidence. This case emphasises the injustices suffered by plaintiffs in serious personal injury cases over recent years (see Kaufmann v. Van Rymenant [3] ; Jacobs v. Varley [4] ; Sharman v. Evans [5] ). The effect of monetary inflation on increasing wages has been ignored in the assessment of future economic loss, but taken into account in setting high discount rates (which reflect the high interest rates that are a product of inflation) when determining the present value of future wage losses. The general increase in the real value of wages has also been ignored. In Australia, over the period of twenty-five years prior to 1975, the real value of wages rose by about 3 per cent per year because of increases in productivity. Although these increases occurred in some and not other industries and to a greater extent in some than in others, their effect was distributed throughout the community of wage earners by the arbitration system (see Report of the Working Party on the Measurement of Labour Productivity, Department of Employment and Industrial Relations, Nov. 1975). In view of this long experience of increasing productivity being reflected in increased real wages, it is reasonable to expect that over a similar period of decades (with which this case is concerned) the same trend will occur. However the evidence is that, since 1975, average real wages have decreased. In what must be a crude process of assessment under present circumstances this factor of increase or decrease should be left out. 1. (1975) 6 A.L.R. 153. 2. (1976) 50 A.L.J.R. 519. 3. (1977) 138 C.L.R. 563. Also, although it is common experience that workers undergo natural age progression and promotion, the practice has been to assume that what are described as the "vicissitudes" will be adverse to their plaintiff's economic prospects. A further factor has been the practice of taking into account income taxation on earnings when assessing future economic loss but ignoring the effect of income taxation in assessing the sum necessary to recoup that loss. After British Transport Commission v. Gourley [6] , tax was considered (by way of reducing the assessment of the loss) but was ignored in assessing the value of the sum awarded as restitution. Gourley's Case did not decide that it should be ignored. This practice continued until Atlas Tiles Ltd. v. Briers [7] (which decided that tax should be ignored altogether). In Cullen v. Trappell [8] , it was held that tax should be considered in assessing the loss and the restitution. 1. [1956] A.C. 185. 2. (1978) 144 C.L.R. 202. 3. (1980) 146 C.L.R. 1. The purpose, in arriving at a present valuation, of discounting moneys to be received in the future, is to take account of the fact that use can be made of the moneys if they are received now. When dealing with money, and ignoring the possibility of turning it into non-monetary hedges against inflation, a justifiable approach is to adopt a discount figure which represents the difference between the inflation rate and the return upon a supposedly safe monetary investment, that is, bonds (gilt edged) or loans on mortgage. These are safe in the sense that the capital is safe although unsafe in the sense that the real value is susceptible to erosion by inflation. If inflation is running at 10 per cent and safe interest rates at 12 per cent, then the appropriate discount rate is 2 per cent; in times of what has been called "stable currency", that is, when inflation is running at about 3 per cent, and interest rates at 5 to 6 per cent, the discount rate also logically should be about 2 to 3 per cent. In England, according to Lord Diplock, it seems that the margin between inflation, if any, and return on safe monetary investment has been about 4 to 5 per cent (see Cookson v. Knowles [9] ; Mallett v. McMonagle [10] ). In Australia, the use of high discount rates, such as 5, 6 7 and even 8 per cent represented a grave injustice to seriously injured claimants and widows and children under Lord Campbell's Acts. The result has been that ridiculously low sums have been awarded for very substantial economic losses. 1. [1979] A.C., at p. 571. 2. [1970] A.C., at p. 176. The proper way to assess the present value of the amounts to be paid under the indexing machinery of the Workers' Compensation Act, is to ignore the effects of monetary inflation on increasing the payments and if any discount is applied, to adopt a rate, if there is one, which prevails as the difference between the rate of inflation and the return upon safe investment, which may be described as a "constant" or "real" interest. Otherwise, one is making an assessment in which monetary inflation, instead of being cancelled out, affects both the payments and the offsetting discount rate. But as Stephen J. shows, there is no "constant" or "real" interest. The variations above and below zero justify the adoption of a rough zero rate, in other words, no discount. The merits of such a convergent approach are simplicity and absence of wild fluctuations in times of sudden inflation or deflation. Logically, taxation on the earnings on the moneys awarded to the plaintiff should be taken into account by way of increasing the award. This is because the no discount approach is an offset of high interest on such moneys against inflation of the sums to be paid. But the high interest (or other earnings) must still be obtained in order to justify the offset. The interest or other earnings will be subject to income taxation, and to fail to take this into account is to fail to achieve complete restitution. As the majority does not share my view, and because the task would be extremely difficult on the material in evidence, I will refrain from attempting to assess this factor. With this reservation (leaving aside the question of "nursing") the most appropriate course is to agree with amounts arrived at by Stephen J. This case has been a very long and expensive exercise. A more efficient approach would have been to measure the damages as the cost to the plaintiff of having an insurer assume the liability. I suspect that result would have been much higher than the award the plaintiff will now receive. I agree with the reasons of Aickin J. for holding that no allowance should be made for "nursing" under s. 10 of the Workers' Compensation Act, so that the appeal on that ground should be dismissed. Otherwise I agree with the order proposed by Stephen J. Aickin J. In the appeal by Barrell Insurances Pty. Ltd. ("Barrell"), I agree with the reasons for judgment prepared by my brother Stephen and with his conclusion that the appeal should be allowed and the amount of damages as awarded by the Court of Appeal of the Supreme Court of New South Wales should be reduced in the manner indicated by him, subject to the observations below as to the liability in respect of "nursing". In the appeal by Pennant Hills Restaurants Pty. Ltd. ("Pennant"), I am of the opinion that on the facts of this case the domestic assistance in question does not constitute "nursing" within the definition of Medical treatment in s. 10 of the Workers' Compensation Act, 1926 N.S.W., as amended. This is not to say that some of the activities now in question could not form part of "nursing" in a different context. Some of them may be capable of falling within that term if they were incidental to or directly associated with "nursing" in the strict sense. In the present case however they were not so associated. Accordingly I am of opinion that the injured worker would not have been entitled to an award of $60 per week in respect of the services in question. The majority of the Court of Appeal took the view that the services in question were not "nursing" but added a sum of $5,000 to the award in respect of Pennant's liability to pay for future "home nursing" and that figure does not appear to have been challenged. The Workers' Compensation Commission did not in its order against Pennant direct that it should pay any amount in respect of nursing services. According to the trial judge, the Commission on 19th December 1972 made an award in favour of the injured worker that Pennant pay to him a weekly sum of $43 (the then maximum weekly payment) as well as "s. 10 expenses". The latter item was not quantified or particularized but it plainly meant expenses referred to under s. 10 as and when they might properly arise. There is nothing in the material before us to suggest that any claim in respect of nursing services was made by the injured worker before the Workers' Compensation Commission or against Pennant. It appears from the material available to this Court that it was only in the present proceedings that Pennant made a claim against Barrell on the basis that its damages included an amount based on its liability to pay for nursing services. If however the services in question are not "nursing", then no liability in respect of the payment has or will arise. I am therefore of opinion that Pennant's appeal on this point should be dismissed. Accordingly the damages awarded were currently reduced by the Court of Appeal, which excluded the amount of $60 per week which was included in the trial judge's calculation. Wilson J. Section 9A of the Workers' Compensation Act 1926 N.S.W., as amended, ("the Act") provides for the periodic adjustment of compensation payments in proportion to changes in the average weekly wage of all adult males employed in industry. This consideration, together with the nature of the liability of Pennant Hills Restaurants Pty. Ltd. ("Pennant") in respect of which compensation is to be assessed serve, in my opinion, to make this an exceptional case in the field of assessment of damages. I have had the advantage of reading the reasons for judgment prepared by Mason J. I agree with his Honour's conclusions, both in respect of the choice of a discount figure of 2 per cent in the assessment of compensation for future expenditure, and in respect of the question whether domestic assistance constitutes "nursing" within the meaning of s. 10 of the Act. I am also in general agreement with the reasoning advanced by his Honour in support of those conclusions. I would therefore allow the appeal and cross-appeal of Barrell Insurances Pty. Ltd. and dismiss the appeal and cross-appeal by Pennant.