[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	The Law Commission
You are here: BAILII >> Databases >> The Law Commission >> Pre-Judgment Interest on Debts and Damages (Report) [2004] EWLC 287(7) (23 February 2004) URL: http://www.bailii.org/ew/other/EWLC/2004/287(7).html Cite as: [2004] EWLC 287(7)

[New search] [Help]



PART VII

COMPOUND INTEREST IN PERSONAL

INJURY CLAIMS

7.1     In broad terms, personal injury damages can be divided into three types: nonpecuniary loss; past pecuniary loss (which often arises over time); and future loss. The courts have evolved particular rules for how interest should be awarded on non-pecuniary loss and continuing loss, which we consider below. Future loss does not carry interest, and does not need to be considered here.

INTEREST ON DAMAGES FOR NON-PECUNIARY LOSS

7.2     Non-pecuniary damages include damages for pain and suffering and loss of amenity. In personal injury claims they carry interest on the whole sum at the rate of 2% from the date of service of the claim until the date of trial or payment (whichever is the earlier). This rule was laid down by the Court of Appeal in 1982 in Birkett v Hayes, ^[1] and was later approved by the House of Lords in Wright v British Railways Board. ^[2] It does not apply to non-pecuniary damages for fatal accidents, which do not increase with inflation and which usually carry interest at the special investment account rate. ^[3]
7.3     The subject has proved controversial. As discussed in Part V, in many areas of law non-pecuniary damages do not carry interest. ^[4] The Pearson Commission, among others, argued that no interest should be payable on non-pecuniary loss in personal injury claims. ^[5] They put forward three main arguments. First, awards are made at current day prices and are therefore protected against the effects of inflation. Often an investor cannot do more than protect the value of their money, after inflation and tax are taken into account. Second, the figures relating to nonpecuniary loss are essentially arbitrary and it would be inappropriate to apply detailed financial calculations to them. Third, awards cover both the pain that a claimant has already suffered and the pain that they will suffer in the future. In theory, only the award for past suffering should carry interest, but it would be too difficult and artificial to distinguish between the two.
7.4     The 2% rate has been justified on the grounds that it represents the rate a standard investor could expect to receive over and above inflation. The limitation that interest only runs from the issue of the claim is justified on the grounds that pain and suffering takes place over time. Some will occur at the time of the injury; some between injury and settlement; some in the future. ^[6]
7.5     Both justifications can be described as rough and ready. First, the rate investors can expect over and above inflation also arises in the context of future pecuniary loss, where the full loss is discounted to reflect the fact that a claimant will invest the money at little or no risk. Currently, the discount rate is set at 2.5% by statutory instrument ^[7] to reflect returns above inflation from index-linked government stock, as suggested by the House of Lords in Wells v Wells. ^[8] Second, pain and suffering will occur at different times for different people – and bears no relationship to the date of the service of the claim. Under the civil justice reforms, the pre-action protocol procedure means that claims are now being served later, ^[9] which has reduced the amount of interest defendants are required to pay.
7.6     We examined the issue at length in our 1999 report on Damages for Non- Pecuniary Loss. ^[10] We noted that views were divided, with some arguing in favour of the 2% rate, some arguing that the discount rate for future loss should be used, and some arguing for an even higher rate to discourage delay. In the end, we recommended no change. We did not accept that the low interest rate led to delay. We felt that the rate based on index-linked government securities should be discounted to cover the fact that interest was being awarded for future suffering. Injustice could be done to defendants in particularly large and catastrophic cases if no allowance was made for the fact that some of the damages related to the future. ^[11]
7.7     In the Consultation Paper we did not seek to reopen the debate on interest on non-pecuniary damages. We accepted that the current rule would remain, and observed that given that both market interest rates and inflation operate in a compound fashion, "it would make sense for the inflation-free rate… also to be compounded". This observation generated little discussion in the responses we received.
7.8     This is not the place to reopen the difficult question of interest on non-pecuniary loss in personal injury claims. We have already consulted on this issue extensively and do not feel that we can usefully add to the full discussion set out in our 1999 report. The only question we need to consider here is whether the 2% rate should always be simple, or whether in some cases it should be compounded.
7.9     As a matter of principle, there is no reason to strive for precision in the calculation of interest on non-pecuniary loss. The loss figure itself is a global sum, aimed broadly to compensate for a loss that is difficult to pin down in monetary terms. The interest rule is also imprecise, intended as broad compromise between providing no interest and providing interest at market rates.
7.10     In practice, compounding a rate as low as 2% would make little difference. Compound interest works exponentially, which means that the difference between simple and compound rates is much greater for high rates of interest than for low ones. It would take 22 years for a monthly-compounded rate of 2% to overtake a simple rate of 2.5%. As interest runs only from the date of service of the claim, the period to which it applies is short. This means that the amount of money at stake will be extremely low – too low to justify the costs of calculation.
7.11     In our view the "2% rate" of interest should continue to be simple, rather than compound. We say this partly because it would be unwise to attempt to apply commercial precision to such an imprecise award and partly because the amount of money at stake does not justify the calculations.

7.12 We recommend that the "2% interest rate" applying to non-pecuniary personal injury damages from the date of service of the claim should continue to be simple, rather than compound.

7.13     The draft Bill permits rules of court to exclude certain heads of damages from the compound interest regime ^[12] and we recommend that this power should be used to exclude non-pecuniary personal injury damages. The question of compounding could be reconsidered if, in the future, the courts were to reconsider the 2% rule and provide for a higher rate of interest for a longer period.

INTEREST ON PAST PECUNIARY LOSS

7.14     Loss of earnings and care costs often arise steadily over a period of time. In Jefford v Gee, ^[13] Lord Denning MR argued that it would be sensible to take a "broad brush" approach to continuing losses, ignoring small discrete items of loss and minor fluctuations in earnings. On this basis, he suggested that past pecuniary loss should carry interest for the full period between injury and trial, but at half the normal rate. The "half rate" approach has since developed into a general rule, although it is subject to exceptions.
7.15     In Jefford v Gee, Lord Denning linked the normal rate to the rate paid on money paid into court (now called the "special investment account rate"). For many years, this was 8%, which led the "half-rate" approach to be labelled the "4% rule". As discussed in Part III, the special investment account rate was reduced to 7% in August 1999 and to 6% in February 2002. However, we were told that the 8% rate has become so much part of the court culture that it is not unusual for claimants to continue to receive interest on continuing losses at 4%.
7.16     We examined the "half-rate" rule in detail in our 1999 report on medical, nursing and other expenses. ^[14] We thought that the half-rate approach to past losses was a useful starting point, but could often work injustice. We approved of case law suggesting that there should be wide-ranging exceptions: The use of a half-rate calculation is only an approximation, and is accurate only if the loss has accrued at a constant rate between the injury and the trial. Where losses have occurred over a period which has ended before trial or are discrete items of expenditure, the halfrate approach becomes inaccurate, and this inaccuracy is exacerbated where the loss occurs either very shortly after the injury, or not long before the trial. ^[15]
7.17     We welcomed the fact that practitioners were increasingly using computer programmes to work out the full interest on discrete individual losses from the date they occurred.
7.18     A further problem with the "half-rate approach" is that it only works accurately if the interest rate remains constant throughout the period of the loss. When the interest rate changes, the approach breaks down. If one calculates interest for each period during which the interest rate applies at half the rate on the whole sum, one ends up giving too much prominence to the early rates and too little prominence to the later rates. ^[16] As interest rates were much higher in the early 1990s than they are now, a crude application of the half rate rule tends to over-calculate interest.
7.19     Despite the problems with the half rate approach, several respondents supported it on the grounds that it was well known, easy to use and roughly fair. ^[17] We accept that on many occasions it is appropriate to treat continuous losses as arising evenly, and ignore slight variations in earnings and expenses. This will continue. However, we are concerned that the "half-rate" rule of thumb already causes serious inaccuracies when the rate varies over time, and ceases to apply when interest is compounded. It should be possible to devise computer programmes or tables that reach a more accurate figure, using the same basic principles.
7.20     For present purposes, the first question is whether interest on past pecuniary loss should be simple or compound. Logic suggests that the rates should be compound. Pecuniary losses are real losses. Victims who have lost income or incurred increased care costs will have either increased their borrowing or lost the opportunity to save or pay off their mortgages. All these possibilities would have been at compound rates. In some cases, borrowing can be substantial. Hazel Genn's survey into the needs of accident victims found that the more serious the accident, the longer the wait, and the greater the victim's indebtedness. Among those receiving compensation of £100,000 or more, a fifth had borrowed £5,000 or more and 7% had borrowed £10,000 or more. ^[18]
7.21     Defendant organisations argued that many families will have simply foregone expenditure – which is not necessarily a compound loss. If one assumes, however, that individuals are economically rational they would only spend if they assessed the benefits to them of spending to be more than the benefits of saving at compound rates. Thus the subjective loss represented by forgoing expenditure is, by definition, higher than the loss represented by not investing (at compound rates).

7.22 We recommend that past pecuniary losses should be subject to the general scheme for compound interest outlined in Part V.

7.23     We accept that in smaller cases, it may not be worth calculating compound interest. In Part V we recommended that in claims of less than £15,000 there should be a presumption in favour of simple rates. We recommend that for the purposes of applying the £15,000 figure to personal injury cases, non-pecuniary loss, past pecuniary loss and future loss will each be looked at separately. Thus in a case involving £5,000 non-pecuniary damages, £10,000 past loss of earnings and £20,000 future loss, the court would start with the (rebuttable) presumption that interest on the £10,000 past pecuniary loss should be simple. In many cases past losses are substantial and endure over long periods, so the amount of money at stake justifies the calculation.
7.24     This leaves the practical problem of how the calculations will be carried out. We endorse the policy we reached in 1999. Where there are significant discrete elements (such as home adaptations), these should be treated separately. In most cases, though, a broad brush approach should be taken by assuming that the losses arose evenly over time. Many past pecuniary losses arise continuously, or approximate sufficiently closely to continuous loss for the assumption to be appropriate. We also accept that the basic principle is correct. Continuous loss should bear less interest than would be paid on a one-off loss, because not all the loss has endured for the whole period. The difficulty is that the "half-rate calculation" does not work for compound interest. ^[19] Nor does it allow rates to be tracked over time. ^[20] More sophisticated tools will be necessary. We think that if lawyers are given the right tools in terms of computer programmes and tables they will be able to reach more precise interest figures without any increase in legal costs or disputes.
7.25     Personal injury lawyers told us that they already routinely use computer programmes to calculate interest (including separate calculations for several discrete items of expenditure). There is no reason why such a computer programme should not use compound rates.

7.26 We recommend that the Court Service's prescribed computer programme should be able to calculate compound interest on losses that occur evenly over time. It should also be able to deal with compound interest on one or more discrete items of expenditure, arising at different dates.

7.27     Judges told us that it would also be helpful if interest on continuous loss could be calculated from tables, available for settlements at the door of the court and in court itself. This would not only be useful for personal injury claims, but may be used in several types of claim – including unpaid rent or breach of patent. As with the "one-off" loss tables they would provide a multiplier based on the month in which the award started and the month in which it concluded. ^[21]
7.28     We recommend that the Court Service should consult practitioners on whether there is a demand for published tables to cover compound interest at the specified rate on continuous losses that occur evenly over time. The Court Service should also consider whether there is a demand for simple interest tables to track the specified rate in past years.
7.29     These tables would be in addition to a computer programme able to carry out such calculations. They would also be in addition to the published tables to cover "one-off" losses recommended in Part VI. ^[22] Where losses cannot be treated as arising evenly, or where interest does not track the specified rate, the parties would need to use the computer programme.

THE COST TO THE TAXPAYER

7.30     The final argument made against compound interest is that it would lead to additional costs for the taxpayer. The Medical Protection Society observed that the Government (and thereby the taxpayer) is probably the largest single personal injury compensator in England and Wales (including compensation for acts or omissions by employees in the National Health Service, the Ministry of Defence, Home Office, Police and Local Authorities).
7.31     The NHS litigation authority (NHSLA) told us that their estimate for future payments of all cases already on their database exceeded £4 billion: so "even an increase in expenditure of 10% would constitute a huge additional burden for the NHS". They feared that money spent on paying compound interest would take money away from new hospitals, doctors and nurses.

The exponential effect of compounding

7.32     In order to assess the cost effect of our proposals, it is important to understand the exponential effect of compound interest. Table 7.1 takes a hypothetical case with total damages of £1 million, of which £150,000 represents past pecuniary loss arising continuously since the cause of action. We were told that in large clinical negligence cases of this type, this was typical in that the past pecuniary element averaged around 15% of damages.
7.33     The first line of the table shows the interest that would be payable if one tracked the special investment account rate applying each year since the cause of action arose. The second line shows the effect of changing to a simple interest rate set at base +1% over the same period. It demonstrates that for recent cases (of ten years or less), this change would produce net savings. For the last decade the special investment account rate has been higher than base +1%.
7.34     The subsequent lines show the effect of compound interest. For cases of five years duration there are net cost savings. For cases of ten years duration the effect is minimal: compound interest adds less than half of one percent to total damages. After 15 years, damages increase by 2%; after 20 years, by 8%; and after 25 years, compound interest adds over a fifth to the total damages bill.
7.35     There are two main factors that contribute to this steeply rising curve. The first is the exponential effect of compounding. The second is that between 1979 and 1993 interest rates were much higher than they are now. With interest rates of 15%, there was more to compound.

TABLE 7.1: INTEREST PAYABLE IN A £1 MILLION CASE, WITH CONTINUOUS PAST PECUNIARY LOSSES OF £150,000

	5 Years	10 Years	15 Years	20 Years	25 Years
Special investment interest rate	£24,900	£54,915	£97,530	£141,525	£191,398
Simple interest at base +1%	£20,850	£47,100	£78,550	£118,594	£135,462
Compound interest at base +1%	£22,994	£59,309	£119,905	£238,451	£451,044
% increase in interest payments	-7.6%	+8.0%	+22.9%	+68.5%	+135.7%
Total payment including simple interest at special investment rate	£1,024,900	£1,054,915	£1,097,530	£1,141,525	£1,191,398
Total payment including compound interest at base + 1%	£1,022,994	£1,059,309	£1,119,905	£1,238,451	£1,451,044
% increase in total payments caused by compound interest at base +1% over simple special investment rate	-0.19%	+0.42%	+2.04%	+8.49%	+21.79%

7.36     The effect of these figures is that where an insurer defendant is settling claims that last 10 years or less, the effect of our proposals will be broadly neutral. Any small increases in the ten-year cases will be offset by interest reductions in the five-year cases. However, if a defendant pays large sums for claims in which more than 15 years have past from the loss arising to payment, then the effect of compound interest can be significant.

Clinical negligence claims

7.37     It is rare for personal injury claims to take more than 15 years to resolve. As discussed in Appendix D, most claims are resolved within three years. ^[23] Surveys have highlighted that clinical negligence claims take longer than other types of claim, but even so most are resolved well within 10 years. A survey of High Court litigation conducted before the Woolf reforms found that the average clinical negligence case took 65 months from when solicitors were first instructed to the conclusion of the case. ^[24]
7.38     The problem comes in cases of severe injuries to children, usually at birth. These represent only a small number of cases, but a high proportion of the money paid. The NHSLA told us that around two-thirds of all the compensation they pay is paid to children. In children's cases the limitation period does not begin to run until adulthood. If the child is under disability, there is presently no limitation period at all. ^[25] This means that there may be lengthy delays from the incident to when the claim is first reported.
7.39     Interest claims are concerned with the time from loss to payment (rather than from the incident). A particular feature of children's cases is that losses may start to arise immediately following the incident, which means that there may be very lengthy delays between the loss and payment. This contrasts with industrial disease claims, where the incident may have been many years ago, but where the loss has only started to show recently.
7.40     In Appendix E we analyse the information provided by the main medical defence organisations. This shows that in 2002, the NHSLA spent just over £63 million on claims where the cause of action had arisen 15 or more years previously. ^[26] Over £32 million was spent on claims that were more than 21 years old. Old claims are also reported to the Medical Defence Union and Medical Protection Society, though not in such quantities. On the basis of the figures in Appendix E, we estimate that in 2002, between them, the MPS and MDU paid around £5.5 million on claims that were 15 or more years old. On the data we were given, it would appear that around 16% of total damages represent past pecuniary loss.
7.41     The NHSLA only covers England. We were not able to gather information about claims paid by NHS Wales. However, published data suggest that the cost of clinical negligence in Wales is just over a tenth of that in England. ^[27] We estimated Welsh expenditure on this basis.
7.42     In Appendix E we estimate that compound interest would add between £20 million and £25 million per year to the cost of clinical negligence claims. This figure consists of £19.1 million to the cost of NHSLA claims, £2.3 million a year to the cost of MPS and MDU claims combined, and £2 million in respect of Wales. These costs would, directly or indirectly, be borne by the taxpayer.
7.43     In arriving at these figures, we have assumed that interest is granted in full, rather than being reduced to penalise claimants for their delay in bringing the claim. The courts have shown themselves increasingly willing to reduce interest payments where there has been considerable delay in bringing infant claims. ^[28] For example, in Corbett v Barking, Havering and Brentwood Health Authority ^[29] the Court of Appeal upheld a reduction of four years' interest, where an infant whose mother had died at birth did not bring a claim for 11.5 years. They rejected the argument that this would penalise an innocent child rather than the adults and lawyers responsible for the delay. Clearly, there would be some reductions in interest payable to penalise claimants for delay, but it is difficult to know how much this would be.

Is compound interest for clinical negligence claims a political priority?

7.44     The medical defence organisations argued that it was not a political priority to compensate clinical negligence claimants for the cost of being kept out of their money for many years. First, they suggested that the losses were often not tangible in terms of money spent, but hypothetical constructs of the opportunity that has been forgone. Second, they pointed out that delay in reporting the incident was the claimant's responsibility rather than the defendant's. Defendants should not be penalised for that delay. Third, they suggested that there were better uses to which the money could be put in improving the health service now rather than attempting to remedy old wrongs.
7.45     As a matter of principle we think it right for the courts to have a power to award compound interest in clinical negligence claims. Many losses are far from hypothetical. Parents have often lost income, leading to increased borrowing or the loss of the opportunity to invest in pensions or other savings, to buy a property or to pay off the mortgage. Where a claimant has been guilty of unwarranted delay the courts already have the power to deny interest for this period. That does not mean that all delay is always unwarranted.
7.46     On the other hand, we cannot say how far granting compound interest to clinical negligence victims represents a political priority, compared with all the other calls on government expenditure. That must be a matter for the Government. There are arguments that the Government should wait until the current backlog is cleared before introducing compound interest for clinical negligence claims. Compound interest for clinical negligence might, for example, be delayed for five years after its introduction in other areas. If needs be, compound interest for clinical negligence claims could be introduced only prospectively, for causes of action that arise after the introduction date. We return to this issue when considering the transitional arrangements in Part IX.

SUMMARY

7.47     We recommend that compound interest should not be available on non-pecuniary damages for personal injury. The current rule should continue, by which nonpecuniary loss carries interest at 2% from the date of service of the claim. It is unnecessary to apply commercial precision to such an imprecise award, and the amount of money at stake does not justify the calculations.
7.48     Damages for past pecuniary loss should be subject to the general scheme for compound interest. The Court Service will need to ensure that the prescribed computer programme is able to calculate compound interest both on losses that occur evenly over time, and on discrete items of expenditure arising at different dates. The Court Service should also consult practitioners on whether there is a need for published tables to cover compound interest on continuous losses that arise evenly over time.
7.49     For most types of personal injury claim, the cost will be broadly neutral. Moderate increases in the interest paid on claims for between five and ten years will be offset by reductions in the interest paid in claims of less than five years duration. Compound interest makes a significant difference in claims that have been outstanding for 15 years or more. In the field of clinical negligence, significant sums are paid on a few very long-running claims – which sometimes arose 25 years or more before payment. We estimate that our proposals will result in net costs to clinical negligence defendants of between £20 million and £25 million per year. How far this represents a priority in terms of public expenditure is a matter for the Government to determine.