. 6
( 18)


5.4 Conclusion
In tort law, a person cannot be held liable to compensate for harm su¬ered by
another unless there was a causal connection between the harm and that person™s
conduct. We have seen that the legal concept of ˜cause™ is a complex and detailed
one which, although related to ideas of causation and responsibility utilized in non-
legal contexts, has distinctive features which are explicable in terms of the purposes
of tort law. It is important to distinguish between the concept of factual causation,
which is concerned with the way things happen in the world, and the concepts of
legal causation, foreseeability and remoteness of damage, which are concerned with
allocating responsibility for life™s misfortunes.

Damages for personal injury and death

6.1 The lump sum: predicting the future
6.1.1 Personal injury cases
Damages for personal injury and death typically take the form of a lump sum. The
award or settlement is made once for all, and there is “ except in rare cases “ no pos-
sibility of increasing it or decreasing it later because of changes in the claimant™s situ-
ation. In the great majority of instances where the injuries are relatively minor, this
raises no real problem because the injured person is likely to be completely recov-
ered long before the damages are assessed, and the whole episode is by then past
However, the lump-sum remedy does raise acute problems wherever a person
su¬ers serious injuries, the e¬ects of which may still be felt long after the damages
are assessed. The Pearson Commission estimated that about 7.5% of all tort claims
(including claims in fatal cases) involved future earnings losses after the trial or
settlement of the claim;1 and this is the type of claim that raises problems with
lump sums. In cases of continuing income loss, or where the injured person will
have a continuing need for hospital, medical or nursing care, two sets of predic-
tions have to be made at the date of trial or settlement in order to calculate an
appropriate sum. First, it is necessary to predict what would have happened to
the injured person if they had not been injured, a prediction which obviously
cannot be veri¬ed or falsi¬ed by subsequent events. Secondly, it is necessary to
predict what is now likely to happen to the injured person. For example, will they
ever make a complete recovery? If so, how long will it take? If not, what residual
degree of disability will there be? How will this a¬ect the injured person™s earning
capacity? Will they su¬er further pain and discomfort? Or die sooner than might
otherwise have been the case? In the case of certain types of injury there is always
a risk of complications in the future; for example, epilepsy is almost always a
risk in brain damage cases, and arthritis is a common risk wherever bones are
severely fractured. These risks are often low (e.g. the risk of epilepsy following
brain damage is often put at one in ten), but lump-sum damage awards must take

1 Pearson Report, vol. 2, paras. 43“4.

Damages for personal injury and death 131

account of the risk. This is done by calculating what sum would be appropriate if
the risk materialized, and then giving as damages a fraction of this sum propor-
tional to the risk occurring.
Obviously the task of predicting the future is extremely di¬cult, and mistakes
can occur even when best e¬orts are made. Still, it is highly unsatisfactory that given
the extreme di¬culty of the task, there is hardly ever any opportunity for making
a subsequent correction. If it is predicted that the injured person will make a com-
plete recovery and this does not happen, they will have been awarded less damages
than they should have got. Conversely, if a claimant recovers more quickly than pre-
dicted, too much compensation will have been awarded. In ˜chance™ cases (such as
those involving a risk of epilepsy) the problem is even worse because the lump-sum
award is bound to be wrong. If the risk eventuates, the amount awarded will be too
little; if it does not, too much. There is no possibility here of making an inspired
guess and hitting the right sum.
These di¬culties may be aggravated by the phenomenon of ˜compensation neu-
rosis™.2 This psychological condition “ which may be distinguished from conscious
˜malingering™ “ is said to have the e¬ect of prolonging the period of recovery and
rehabilitation until after trial or settlement of a tort claim. Anxiety over the likely
outcome of the claim may postpone complete recovery. Besides being an undesir-
able by-product of the once-for-all lump-sum damages system, compensation neu-
rosis can cause problems for assessing damages. If it is assumed that disabilities are
permanent whereas they are, in fact, a symptom of compensation neurosis that will
disappear once the claim has been resolved, the injured person will have been over-
compensated; conversely, if the case is wrongly thought to be one of ˜compensation
neurosis™, the injured person may be under-compensated.
To some extent these di¬culties of prediction can be, and are, mitigated by post-
poning the trial or settlement of the action until a prediction about the ultimate
outcome can be made with greater con¬dence.3 Indeed, in many cases of serious
injury it is essential to do this because no satisfactory predictions can be made at
all until many months at least after the injuries were su¬ered. Although in strict law
the assessment of damages in most cases should be related to the time of injury

2 This is a very complex and controversial topic. See e.g. T.G. Ison, ˜The Therapeutic Signi¬cance of
Compensation Structures™ (1986) 64 Canadian Bar R. 605, 610“29; C.Vincent and I.H. Robertson,
˜Recovering From a Medical Accident: the Consequences For Patients and Their Families™ in C.
Vincent, M. Ennis and R.J. Audley eds., Medical Accidents (Oxford, 1993), 163; G. Mendelson,
˜ “Compensation Neurosis” Revisited: Outcome Studies of the E¬ects of Litigation™ (1995) 39 J. of
Psychosomatic Research 695; P.W. Halligan, C. Bass and D.A. Oakley eds., Malingering and Illness
Deception (Oxford, 2003), esp. chs. 1, 13, 16, 17 and 18. In suggesting that a person either has or
does not have compensation neurosis, the text may be simplistic. There is evidence that injured
people who make compensation claims may su¬er worse long-term health outcomes than people
who do not. The reasons are ill understood, but it appears that the compensation process is only
one. See e.g. Royal Australasian College of Physicians, Compensable Injuries and Health Outcomes
(Sydney, 2001); R. Mayou, ˜Psychiatric Outcome Following a Road Tra¬c Accident™ [2004] JPIL
3 A variant is to decide the issue of liability ¬rst but leave the assessment of damages until later.
132 Chapter 6

because that is when the ˜cause of action™ vests in the claimant, the courts have
always sensibly insisted that what happens between the date of the harm-causing
incident and the date of the trial must be taken into account in assessing the
damages. Thus, lost earnings su¬ered between date the harm is su¬ered and the
trial or settlement will be calculated, not guessed; and increases in wage rates during
this period will become the basis of the assessment of damages for expected future
earnings losses.
On the other hand, postponement of the trial or settlement of cases brings its
own evils; indeed, delay in actually securing payment under the tort system is one
of the major causes of dissatisfaction with it. It may be possible to reduce these
delays, but plainly it would not be in the claimant™s interest to require the claim to
be determined before the exact nature and extent of the injuries could be pre-
dicted. Postponement will not, in any event, solve all the problems that may arise.
There are many cases in which, even when a reasonably ¬rm medical prognosis
can be given, the e¬ect of a person™s injuries on their future working prospects
must remain problematical until long after the time at which a claim must be tried
or settled.

6.1.2 Fatal cases
There are two quite di¬erent types of actions that may be brought in respect of
the death of a person as the result of a tortious act. The older type is the depen-
dency4 action under the Fatal Accidents Act 1976 (a descendant of the original
(Lord Campbell™s) Act of 1846).5 This action is primarily designed to provide
compensation for the lost income6 of a person who was formerly maintaining
members of their family, normally a spouse or cohabiting partner and children.
An action of this kind is brought by the dependants in their own name, in
respect of their own loss of ¬nancial support resulting from the death. It is still
necessary for the dependants to prove that the deceased died as a result of a tort,
and any damages awarded will be reduced if the deceased was personally guilty of
contributory negligence. The action provides compensation not only for an
actual dependant but also for a prospective dependant, so long as the claimant
falls within the list of persons entitled to sue under the Act. Thus, a parent may be
able to sue in respect of the death (say) of a child of 16 who has not yet contribu-
ted anything to the parent™s support but who might have been expected to do so
in future.
The second type of claim lies under the Law Reform (Miscellaneous Provisions)
Act 1934. Under this Act a claim for damages (called a ˜survival action™) lies on behalf
of the estate of a person killed, and such a claim may be brought whether or not the

4 The words ˜dependency™ and ˜dependant™ are misleading because the Act allows members of a
de¬ned class of persons to recover damages for ¬nancial loss su¬ered by them as a result of the
death. Such persons will usually be dependants in the ordinary sense, but not always.
5 See generally Law Com. No. 263, Claims for Wrongful Death (1999).
6 Including income in the form of social security bene¬ts: Cox v. Hockenhull [1999] 3 All ER 577.
Damages for personal injury and death 133

deceased had any dependants. The only damages recoverable in such an action are
for pecuniary and non-pecuniary losses su¬ered by the deceased between the acci-
dent and death, plus funeral expenses. In most cases such an action would be
brought concurrently with an action under the Fatal Accidents Act 1976 because a
person™s dependants are commonly also the bene¬ciaries under their will.
Assessment of compensation in dependency actions involves the same two sets
of predictions as must be made in cases of long-term injuries. It is necessary to start
by predicting what would have happened to the deceased if they had not been
killed. In particular an assessment must be made of what the deceased™s earning
prospects were. As Lord Diplock said in Cookson v. Knowles,7 the court is required
to make assumptions:

. . . as to the hypothetical degree of likelihood that all sorts of things might happen in
an imaginary future in which the deceased lived on and did not die. What in the event
would have been the likelihood of his continuing in work until the usual retiring age?
Would his earnings have been terminated by death or disability before the usual retir-
ing age or interrupted by unemployment or ill-health? Would they have increased, and
if so, when and by how much? To what extent, if any, would he have passed on the
bene¬t of any increases to his wife and dependent children? Would she have gone to
work when the children had grown older and made her own contribution to the family
expenses in relief of his?

And so on.
The second set of predictions (i.e. about what will now happen in the future)
generally causes less di¬culty in fatal cases: obviously a lot of predictions required
about the prospects of a living claimant are not relevant in a fatal accident claim.
But over the years there has been a great deal of di¬culty about the problem of
remarriage by a widow. Most fatal accident claims are brought by widows, with or
without additional claims by dependent children. Prior to 1971 it was the accepted
rule that the damages had to be reduced to take account of the remarriage prospects
of the widow;8 and, of course, it followed that they had to be reduced where the
widow had actually remarried before the assessment of damages. The application
of this rule had a considerable e¬ect on claims by young widows, especially widows
without children, but much less e¬ect on the claims of older widows, especially
those with several children. The rule was based on the simple idea that damages in
a fatal accident case, as in all other tort claims, are designed to compensate for a
loss. The damages given to a widow are designed to replace the share of the income
of her former husband that was devoted to her maintenance. When the widow
remarries she will often make good her loss; and, accordingly, the fact or prospect
of remarriage must be taken into account in reducing the damages.

7 [1979] AC 556, 568“9.
8 Goodburn v. Thomas Cotton Ltd [1968] 1 QB 845; the rule was much older than this case, which
rea¬rmed it.
134 Chapter 6

However, many people found the rule distasteful.9 It was argued that there was
a ˜cattle market™ element in the valuation of remarriage prospects,10 and in 1971 a
provision was enacted to the e¬ect that the fact or prospects of remarriage of a
widow in a fatal accidents claim were to be wholly ignored.11 This must be one of
the most irrational pieces of law ˜reform™ ever passed by Parliament. It would be as
sensible to require a divorced husband to maintain his wife after she has remarried,
or for the State to pay pensions to widows after remarriage. An extreme example of
the situation thus created occurred in 1974 when a young widow of 25 who had
remarried an ˜oil man with a ¬ve-¬gure salary™ was awarded £65,000 in damages for
the death of her ¬rst husband 2 years earlier.12 The real complaint against this pro-
vision is that it involves extraordinary generosity to one group of accident victims
without regard to the needs of others. In the case just referred to, for example, the
deceased was killed in a motor collision with a car driven by a man who was also
killed. Since the latter was clearly at fault, his dependants (if any) would have recov-
ered no damages at all, even if they had been in much greater need than the
claimant. Another objection to the provision is that it only deals with remarriage
and not with the fact or prospects of cohabitation, which are taken into account.
This is all the more extraordinary given that a limited class of cohabiting partners
can make claims under the Fatal Accidents Act 1976.
The obvious answer to problems raised by changes in the circumstances of
claimants, including the remarriage of widows, is to pay compensation in the form
of periodical payments rather than a lump sum.13 Such payments could then be
reviewed as circumstances changed, and could be ended if, for example, a widow
remarried.14 But even this would not wholly dispose of the problems arising from
the remarriage of widows. Obviously, if widows™ compensation took the form of
periodical payments which came to an end on remarriage, there would be a temp-
tation for widows to avoid remarriage and enter into less formal relationships.
Social security legislation (where analogous problems have to be dealt with) meets
this danger by providing that widows™ bene¬ts are not payable during periods of
cohabitation. The cohabitation rule has been much vili¬ed as involving an intru-
sion into the private lives of widows, but such an attitude is surely outmoded today

9 See Buckley v. John Allen & Ford Ltd [1967] 1 QB 637, and the Report of the Committee on Personal
Injuries Litigation (Cmnd 3691, 1968) (Winn Committee Report), paras. 378“9.
10 Is there not a similar ˜cattle-market™ element in valuing the loss of marriage prospects of an
unmarried woman?
11 See now s. 3(3) of the Fatal Accidents Act 1976. This provision does not apply to men (Regan v.
Williamson [1976] 1 WLR 305); nor to the assessment of damages for a child whose father has
been killed (Thompson v. Price [1973] QB 838). The Law Commission recommended repeal of
this provision and its replacement with a much more complex set of provisions.
12 The Times, 15 May 1974. This was no doubt an exceptional case, but the fact is that a signi¬cant
number of widows, especially young widows, do remarry. In 1990 the remarriage rate for widows
aged 25“9 was 98 per 1,000, and for those aged 30“4, 86 per 1,000: Marriage and Divorce Statistics
1990 (OPCS, 1992), table 3.3.c (1990 is the most recent year for which this ¬gure is available).
13 Contrast Law Com. No. 263, 4.30“4.34.
14 But see Pearson Report, vol. 1, paras. 409“17.
Damages for personal injury and death 135

when many people openly choose cohabitation as a socially acceptable alternative
to marriage, and when legal discrimination against children born out of wedlock
has been largely removed. Indeed, under the Fatal Accidents Act 1976 itself, chil-
dren born to parents who are not married to each other are to be treated as if their
parents had been married at the time of the birth.
One of the strangest things about the di¬culties encountered by the law in
dealing with compensation for widows is that the law is already perfectly well
acquainted with the system of periodical payments in the family jurisdiction where
amounts payable by way of maintenance can be varied as the situations of the
parties change. If a person receiving maintenance remarries or secures a better job
or inherits a large fortune, the amount of maintenance payable can be reduced;
conversely, if the payer™s income goes up, it can be increased. There is no need for
the court to guess or make predictions about the future.
Death of her husband may not only make it possible for a woman to remarry
but also to go (back) to work. The law requires ˜gains™ resulting from the death to
be set o¬ against losses, but it is not clear in England to what extent actual earnings
or the prospect of earnings are in practice set o¬ against the damages awarded. In
Australia the courts take the view that the death of a husband does not revive the
wife™s ability to work since marriage does not prevent a wife working. And if the
fact that there were children prevented the wife working, the death of the husband
does not alter this. So neither actual nor potential post-death earnings are taken
into account in assessing the widow™s damages.15 This approach, too, can clearly be
attacked as overly generous, although the disincentive to work that would be gen-
erated by the opposite approach might be thought undesirable (even in times of
high unemployment).

6.1.3 Variation of awards after trial
In the small proportion of cases which go to trial, it may very occasionally be pos-
sible to vary a lump-sum award to take account of changes in circumstances occur-
ring after the trial “ at least where they occur soon after the trial. Since, at trial, it is
the facts as they are known at that date that are relevant, appellate courts have not
shrunk from saying that, in the event of an appeal, it is the facts as they are known
at the date of the appeal that are relevant. Notice of appeal must normally be given
within 4 weeks; and if the appeal is heard reasonably soon thereafter, there will be
little opportunity in the ordinary case for taking new facts into account on appeal.
However, the Court of Appeal does have power to extend the time within which an
appeal may be entered by granting ˜leave to appeal out of time™. This power is dis-
cretionary and is not often exercised, but it can be used to increase (or reduce) an
award of damages where new facts come to light very soon after the trial. In one
case, for example,16 where damages had been assessed on the assumption that the

15 See F.A. Trindade and P. Cane, Law of Torts in Australia, 3rd edn (Melbourne, 1999), 544.
16 Murphy v. Stone Wallwork Ltd [1969] 2 All ER 949.
136 Chapter 6

claimant was still capable of continuing in his former employment, the award was
re-opened and increased when he was dismissed as soon as the case was over. In
another case17 the House of Lords allowed evidence to be given that shortly after
the trial it became clear that the claimant would have to be maintained in a nursing
home, at substantial cost; the assessment of damages at the trial had been made on
the assumption that the injured person would remain at home.
Of greater potential importance is a provision in the Supreme Court Act 1981 (s.
32A) designed to deal with the ˜chance™ cases mentioned above. The provision applies
cases in which ˜there is proved or admitted to be a chance that at some de¬nite or
inde¬nite time in the future the injured person will . . . develop some disease or su¬er
some deterioration in his physical or mental condition™. Rules of court have been
made enabling damages to be awarded in the ¬rst instance on the assumption that
the claimant will not develop the disease or su¬er the deterioration, and allowing
further damages to be awarded at a later date if the disease or deterioration occurs.18
However, the procedure has been very little used in practice.19 In Mitchell v.
Mulholland 20 the House of Lords stressed the need for ¬nality in litigation and the
undesirability of reopening awards of damages save in very exceptional cases.
It will be noted that this provision only allows awards to be increased, not
decreased. This asymmetry in favour of the claimant is usually thought to be
required by considerations of fairness, and on the ground that a threat that dam-
ages could be reduced if the injured person™s condition improved might hinder
rehabilitation.21 In practice, too, it would be very di¬cult to secure repayment of
part of the lump sum if the claimant had spent it or invested it in a ¬xed asset such
as a house or a business; but if repayment were only required if the money had been
invested in liquid assets such as shares, claimants would have a strong incentive to
deal with their damages in other (and perhaps less prudent) ways. Another limita-
tion of both appeals and the conditional damages procedure is that they do not
apply to the vast majority of cases that are settled out of court. In 1997 a health
authority paid £700,000 to a 9-year-old who su¬ered severe injuries at birth for
which the authority accepted 75% responsibility. Eight days after the settlement,
the child unexpectedly died, and the health authority said that it intended to try to
recover the amount of the settlement referable to future care. But there is no clear

17 Mitchell v. Mulholland [1971] AC 666.
18 It is unclear whether, if the claimant dies before any further claim is made, the dependants can
bring an action under the Fatal Accidents Act 1976.
19 According to the Compensation Recovery Unit, it was used in less than 0.03% of cases between
2000 and 2003: DCA, Variation of Periodical Payment Orders and Settlements in Personal Injury
Cases: Partial Regulatory Impact Assessment (April 2004). It has been held that the provision does
not apply to cases of gradual deterioration in the claimant™s condition, but only where there is a
risk of a ˜clear-cut™ adverse event: Willson v. Ministry of Defence [1991] ICR 595. The Criminal
Injuries Compensation Authority has power to reconsider and even re-open cases to take account
of new evidence and changes in the claimant™s condition: see 12.4.4.
20 [1971] AC 666.
21 This is also a problem if the damages are awarded as periodical payments: T.G. Ison, ˜The
Calculation of Periodic Payments for Permanent Disability™ (1984) 22 Osgoode Hall LJ 735.
Damages for personal injury and death 137

legal basis on which this could have been done. On the whole, insurers want their
settlements to be ¬nal.
It can be seen, then, that such techniques as the above are of limited value in
dealing even with cases of change in the claimant™s medical condition; and they do
not deal at all with other sources of di¬culty in assessing the lump sum.

6.1.4 Suitability of lump sums
As already implied, it is highly questionable whether awarding damages for lost
income (whether earnings, in the case of a personal injury action, or support in the
case of a fatal accident) or the cost of care in a lump sum is appropriate in cases where
the loss will continue after the date when the damages are assessed. Assessing the
lump sum involves much speculation and potential inaccuracy. Just as importantly
(and perhaps surprisingly) recipients of lump-sum damages awards (except minors22
and the mentally incapable) are free to use the damages as they choose. A damages
award to compensate for future loss to be su¬ered over a period of years “ whether
loss of income or cost of care “ is, of course, designed to be used progressively to make
good those losses as and when they occur, so that at the end of the period of the award
(but not before) the damages award will have been completely used up. Lump-sum
damages are calculated on the assumptions that they will be invested in ˜gilts™ “ i.e.
Index-Linked Government Stocks (ILGS) “ a form of investment that is very secure
but yields relatively low income, and that the recipient™s needs will be met from the
combination of the capital invested and the income it generates. In other words, the
lump sum awarded takes account of the income that can be earned by investing
the damages. But recipients are not required to invest the lump sum in gilts or,
indeed, to invest it at all. Nor are recipients required to take investment advice. Indeed
this is positively discouraged by the rule that the cost of taking such advice (with a
view, perhaps, to making more risky investments that will yield higher income)
cannot be recovered from the tortfeasor as an item of damages.23 There is no legal
control over the way damages awards are managed. Nor are recipients required to use
the damages for the purposes for which they were given.
The typical recipient of a large lump-sum award of damages is, no doubt, more
or less inexperienced in investing and managing such an amount of money. There
is a danger that even assuming the amount awarded was adequate to make good the
losses su¬ered over the whole period of the award,24 it will be invested unwisely or

22 In cases under the Fatal Accidents Act 1976 brought on behalf of a widow or dependent children,
neither the income nor the capital of damages awarded for the young people will normally be paid
out of court except where this is shown to be necessary for the children™s own bene¬t, e.g. to defray
school fees, etc. Awards to children are generally paid out to them as soon as they come of age, but
in extreme cases the court may press the recipient to agree to settle the money on trust: see e.g.
Warren v. King [1963] 3 All ER 993n, where the CA pressed a woman of 20 to settle an award of
£35,000 so that she could not touch the capital until she was 31. It seems that the court has no
power to compel a young person to agree to this: Allen v. Distillers Co. (Biochemicals) Ltd [1974]
2 All ER 365.
23 Eagle v. Chambers (No. 2) [2004] 1 WLR 3081.
24 See further 6.4.
138 Chapter 6

unsuccessfully, and so be dissipated before the end of the period it was meant to
cover. As a result, the injured person (and their family) may be reduced to poverty
and made reliant on publicly funded social security payments, and health and
welfare services. If, as evidence suggests often happens in more serious cases, the
original lump sum was never going to be enough (invested in gilts) to achieve its
aims, then even if it is invested wisely and successfully in higher-yield ¬nancial prod-
ucts, it may be insu¬cient to provide an income of the order of that which has been
lost or is needed to meet expenses. In this light, the rule that recipients of damages
are free to use them as they will, seems very di¬cult to justify.
There is some evidence as to how wisely or unwisely, successfully or unsuccess-
fully, large awards of damages are actually used or invested by the recipients. The
Pearson Commission found that only about 20% of recipients made any attempt
to use their damages for investment, or treated it as capital; most spent the money
on current expenses.25 On the other hand, most of these sums were fairly small (the
average was around £250 in 1977 money values) so this would not necessarily have
been a pro¬‚igate use of the damages. Research conducted for the Law Commission
in 1992“326 found that 60% of those surveyed who received less than £20,000 saved
or invested some of their money; that among those who received £20,000“49,999,
the ¬gure was 83%; among those who received £50,000“99,999, 90%; and among
those who received more than £100,000, 97%. There was a tendency to choose safe
investments. A quarter of those who received less than £20,000 and two-thirds of
those who received £20,000 or more obtained ¬nancial advice; and one in ¬ve of
those surveyed were unhappy with their investment choices.
As to the adequacy of the amounts received, three in ¬ve respondents felt that
their damages had been su¬cient to cover past losses and expenses. A majority
thought that their standard of living had not dropped as a result of the accident, but
a signi¬cant minority thought it had. About half thought that their standard of
living in 10 years™ time would be lower than before the accident. A majority also said
that they were now less satis¬ed with the amount they had received than they had
been at the time of settlement. The researchers concluded that in many cases, this
was because the compensation received for loss of earning capacity turned out to be
inadequate. It appears, too, that in a signi¬cant number of cases, members of the
injured person™s household worked shorter hours or gave up work altogether after
the accident to care for the injured person; and that this additional income loss was
not adequately re¬‚ected in the compensation recovered.
Overseas experience of similar problems is not very encouraging. Research con-
ducted in the early 1980s for the New South Wales Law Reform Commission dis-
covered that a signi¬cant number of recipients of lump sums have inadequate
income from their awards to meet their expenses. In some cases this was the result
of mismanagement of the lump sum award by the recipient or unwise investment

25 Pearson Report, vol. 2, table 89.
26 How Much is Enough?
Damages for personal injury and death 139

(e.g. in a house which generated no income, or in low-yield but secure ¬nancial
products). But in other cases it was due to factors entirely outside the recipient™s
control, such as in¬‚ation or an unexpected deterioration in medical condition. One
survey concluded that recipients of high awards who had managed to re-establish
themselves in a comfortable and secure fashion had typically bene¬ted from some
combination of personal skill and enterprise, good fortune, good advice, and
support from family and friends.
No stranger example could be found of the fundamentally inconsistent philoso-
phies which underlie the social security system and the tort system, and which have
somehow managed to co-exist for so many years. As long ago as 1944, the govern-
ment declared that it did ˜not regard lump sum payments even if administered under
strict control as a satisfactory method of assuring an income™;27 and the payment of
bene¬ts periodically has remained one of the basic features of the modern welfare
state. This is not to say that the tort system™s preference for lump sums is without
foundation. The Law Commission survey referred to above found amongst
claimants a strong preference for lump sums: ˜. . . respondents felt that they wanted
to make their own decisions [about how to use the compensation] and to be in com-
plete control of their budget.™28 Those who expressed a preference for instalment
payments mainly sought security and protection from their own unsuitable spend-
ing patterns and investment decisions. Importantly, the small proportion of respon-
dents who received structured settlements ( generally thought them
preferable to a lump sum, even though most had not requested this form of com-
pensation. Some recipients of structured settlements complained that the lump sum
component in the settlement had not been large enough to enable them to make a
desired capital investment (e.g. in a new house).

6.1.5 Alternatives to lump sums An argument against abandoning the lump-sum system
The obvious question to which the various problems with the system of lump-sum
damages gives rise is whether some system of periodical payments would be pre-
ferable. A common argument against periodical payments is that they are incon-
sistent with the basic principle that the recipient of damages is free to use them as
they wish, and is not required to invest them to produce a stream of income to
replace what has been lost. But this argument can easily be answered by appealing
to the basic function of tort damages, namely to put the injured person back in
the position they would have been in had they not been injured. In this light, the
most accurate way of replacing a stream of income would be by providing a stream
of income, not a lump sum. Of course, damages may also be given to facilitate
capital expenditure “ for instance, on house modi¬cations. This would suggest that
ideal compensation would consist of a mix of lump sum and periodical payments.

27 Social Insurance, Part II (Cmnd 6551, 1944), para. 30.
28 How Much is Enough?, 181.
140 Chapter 6

While there are certainly pragmatic arguments in favour of allowing recipients to
use their damages as they wish “ notably, the di¬culty of monitoring and enfor-
cing restrictions on the use to which damages are put “ it is hard to think of good
reasons of principle or fairness to justify such freedom. Early proposals for alternatives
In 1973 the Law Commission reached the conclusion that a system of periodical
payments could not be ¬tted into the existing tort framework.29 However, a major-
ity of the Pearson Commission recommended that such a system should be intro-
duced for serious personal injury cases and for fatal cases.30 The Commission
proposed that the courts should have power to award damages either as a lump sum
or in the form of periodical payments; and that in the latter case, the amounts
should be in¬‚ation-proofed and variable if the injured person™s medical condition
subsequently changed. They refrained from insisting that all settlements of fatal
and serious personal injury cases should be in this form. Parties would remain free
to settle for lump sums (and, hence, free to use the damages however they wished),
though the Commission thought that claimants might increasingly become aware
of the desirability of settling for periodical payments.
One thing that emerged clearly from the Pearson Report was that very few cases
would have been covered by such a change in the law. Of the 215,000 (or so) cases per
annum in which (the Commission found) some tort compensation was payable, the
Report suggested that only some 2,200 (about 1%) were actually tried in court,31
though no doubt a substantial proportion of these are fatal and serious personal
injury cases. Moreover, the insurance survey carried out for the Commission showed
that in November 1973, only about 1% of all tort claimants received more than
£10,000 in damages (including damages for non-pecuniary loss) (say, £40,000 in
today™s money values); and only 2% received more than £5,000 (say, £20,000 in today™s
money). In other words, only about 2,000 claimants received more than £40,000 in
2006 money. Judging from other ¬gures given in the Pearson Report,32 we can esti-
mate that only about 20% of these cases (say, 400) might actually have been tried in
court. Converting lump-sum damages to periodical payments would only be worth-
while in cases where the lump-sum equivalent was reasonably large. Some idea of how
large is given by the fact that, prior to the introduction of periodical payment orders
(, a Supreme Court Practice Direction required the possibility of a structured
settlement (equivalent, for present purposes, to a periodical payment arrangement)
to be considered by the parties in any case where the amount claimed for future loss33

29 Law Com. No. 56, Personal Injury Litigation: Assessment of Damages (1973), paras. 29“30.
30 Pearson Report, vol. 1, ch. 14.
31 Pearson Report, vol. 2, table 221.
32 Pearson Report, vol. 2, table 128 shows that, of the total number of cases tried in 1974 (2,313),
about one-quarter (521) were awarded more than £5,000 (in 1977 currency values).
33 The Pearson Commission recommended that damages for non-pecuniary loss should continue
to be awarded in a lump sum: vol. 1, para. 614.
Damages for personal injury and death 141

is £500,000 or more. It would seem, therefore, that the Pearson proposals for period-
ical payments would have applied, at most, to a couple of hundred cases a year.34 We
know that the number of successful tort claims per annum has increased about three-
fold since the 1970s;35 and on that basis we might conclude that if the Pearson pro-
posals were enacted now, they would apply to perhaps 500“600 cases each year. Structured settlements
In early editions of this book the conclusion was that in the light of such calculations,
it seemed questionable whether the change proposed by the Pearson Commission
would be worth the cost and complexity it would undoubtedly entail. However,
development of the so-called ˜structured settlement™36 in the 1990s prompted recon-
sideration of this conclusion. Under such a settlement, damages for future losses
are calculated as a lump sum; but instead of the lump sum being paid to the claim-
ant, the insurer who is responsible for paying it uses it (or part of it)37 to purchase an
annuity to provide the injured person with a continuing in¬‚ation-proofed income
for as long as this is needed. The annuity may be for a ¬xed minimum period so that
it will continue to be paid to the recipient™s estate if the recipient dies sooner than
expected. Such an arrangement would provide for dependants.38 Apart from provid-
ing security for the future (the insurer bears the risk of the bene¬ciary living longer
than expected), structured settlements relieve the injured person of the need to make
di¬cult investment decisions or to employ an investment advisor, because the
insurer assumes responsibility for investing the lump sum and providing the income.
Structured settlements are also attractive by reason of the fact that the income from
a structured settlement is not taxable in the hands of the recipient, whereas if the
claimant took the lump sum and invested it, the income39 would be taxable.40 The
major disadvantage of a structured settlement is that once it has been set up, the
capital is unavailable to the bene¬ciary. This disadvantage can be partly neutralized
by leaving a lump sum out of ˜the structure™ or by purchasing a number of ˜annuities™,
one of which provides regular income and another of which provides regular but less
frequent lump sums. In short, a structured settlement involves a trade-o¬ between
¬‚exibility in the use of damages and security.
Valuable though structured settlements might be, they do not solve all the prob-
lems created by the lump-sum system, exactly because they are based on a lump sum

34 In 2004“5 the NHS Litigation Authority (which handles medical negligence claims against NHS
Trusts) made 49 structured settlements worth about £192 million in total (averaging £3.9
million). These settlements represent about 0.5% of the claims closed in 2004“5, but 38% of the
compensation paid in that year.
35 See 8.1.4.
36 See generally R. Lewis, Structured Settlements: The Law and Practice (London, 1993).
37 For instance, the parties may agree not to ˜structure™ damages for non-pecuniary loss.
38 But query whether a Fatal Accidents Act claim could be made after the early death of the
bene¬ciary of a structured settlement under which payments terminated on death.
39 But not the lump sum itself.
40 This also gives a bene¬t to the defendant™s insurer, because the lump sum needed to generate the
required annual amount is less than it would be if the amount were subject to tax.
142 Chapter 6

awarded by a court or agreed by the parties. Thus, all the di¬culties of calculation
and the problems of proof and delay associated with the present system remain. Periodical payments
As the word ˜settlement™ implies,41 ˜structures™ are voluntary. Since 2003, there has
been a statutory provision42 empowering courts to order compensation in the form
of ˜periodical payments™ in certain cases. In cases “ but only in cases “ where
damages for future loss are awarded, such an order (unlike a court order approv-
ing a structured settlement) can be made against the wishes of the parties.
Moreover, whereas a structured settlement is based on a lump sum,43 which is then
˜structured™ to provide periodical payments, the intention is that a periodical pay-
ments order might directly specify the amount to be paid periodically without ¬rst
calculating a lump sum, leaving it entirely to the defendant to decide how to satisfy
the order.44 Before making a periodical payments order, the court must be satis¬ed
that ˜continuity of payment under the order is reasonably secure™. There are also
provisions about variation of such orders along similar lines to s. 32A of the
Supreme Court Act (6.1.3).45 Like the income from a structured settlement, and
unlike the income from investment by the recipient of a lump sum, periodical pay-
ments are not taxable. A periodical payments order can provide for payments to
continue after the death of the injured person, in order to provide support for
dependants. Even if the court decides not to make a periodical payments order, the
parties may agree on some form of periodical award (i.e. a structured settlement),
and the court may con¬rm that agreement in a ˜consent order™.
The provision empowering the court to make a periodical payments order also
requires it to consider whether to make such an order in any case where damages for
future loss are awarded. However, in the light of the Practice Direction noted earlier,
it is perhaps unlikely that a court would make such an order unless the amount
claimed for future loss was at least £500,000 “ and such claims are few indeed.
Furthermore, it appears that very few insurers sell the sort of ¬nancial products
needed to secure structured settlements and periodical payments.46 It is, therefore,
unclear how popular or common periodical payment orders will become. It is also
very di¬cult to predict the likely e¬ect of the power to make periodical payment
orders on settlement of claims out of court. Relevant factors will be the cost of
funding periodical payments relative to that of funding lump sums and structured
settlements, and the willingness of the courts to make periodical payment orders.

41 A settlement is a contract between the parties.
42 Courts Act 2003, s. 100, amending Damages Act 1996, s. 2. See generally R. Lewis,˜The Politics and
Economics of Tort Law: Judicially Imposed Periodical Payments of Damages™ (forthcoming, 2006)
69(3) MLR.
43 Sometimes referred to as the ˜top-down™ method of assessment.
44 Department for Constitutional A¬airs, Guidance on Periodical Payments (2005), para. 4. This is
sometimes called the ˜bottom-up™ method of assessment.
45 The details can be found in the Damages (Variation of Periodical Payments) Order 2005.
46 P. Barrie, Personal Injury Law: Liability, Compensation and Procedure (Oxford, 2005), 536.
Damages for personal injury and death 143

Periodical payments are an advance on structured settlements in that they do not
require the calculation of a lump sum, and are to some extent variable. However, the
circumstances in which a variation order may be made are quite restricted; and
much of the speculation associated with the lump-sum system will plague the peri-
odical payments regime as well. It must also be said that however desirable struc-
tured settlements and periodical payments are compared with lump sums, their
e¬ect is to improve even more the position of a very small group of seriously dis-
abled persons who are able to claim tort damages. Furthermore, the tax advantages
of structured settlements and periodical payments mean that the additional bene¬t
to these lucky people is paid for, partly at least, by the taxpayer. The original purpose
of these tax advantages was to provide incentives for making structured settlements.
They seem unnecessary and undesirable in the light of the power to order periodi-
cal payments. Why should the injured person™s compensation-derived income be
taxed if it results from investment by the recipient personally, but tax-free if the nec-
essary investment is made by the defendant? Is it not in the public interest that recip-
ients of lump sums (who are likely to remain the majority of recipients of signi¬cant
amounts of tort compensation) should be given an incentive to invest their damages
as successfully as possible?

6.2 Full compensation
The tort system is the only compensation system that professes to provide ˜full com-
pensation™. All pecuniary losses (chie¬‚y medical expenses and loss of income, both
past and future) must be compensated for in full.47 The tort victim must also be
compensated for all possible ¬nancial ill-e¬ects of the injury; for example, the risk
of subsequent medical complications, possible reduction in marriage prospects
possible loss of employment prospects and so on. In short, the full compensation
principle requires a detailed examination of the particular situation of the individ-
ual claimant. From time to time, it has been said that ˜full compensation™ does not
mean ˜perfect™ or ˜absolute™ compensation, and that the compensation must only be
˜fair™ or ˜reasonable™. It seems that these remarks have been directed to the assess-
ment of compensation for non-pecuniary losses, where ˜full compensation™ would
be meaningless; they are not intended to suggest any quali¬cation of the principle
that the claimant is entitled to full compensation for all pecuniary losses. Indeed,
in Lim Poh Choo v. Camden Health Authority48 in which (then) record damages of
£250,000 were awarded, the majority of the English Court of Appeal and, on appeal,
the House of Lords, speci¬cally rejected Lord Denning™s argument that it would be
unfair and unreasonable to award damages for loss of earnings if the claimant was

47 But under statutory schemes of ˜strict liability™ limits on compensation may be imposed, e.g., there
is a threshold for recovery of property damage under Part I of the Consumer Protection Act 1987;
and there are limits on the liability of airlines and shipowners for injury to passengers under
various international agreements.
48 [1979] QB 196; [1980] AC 174.
144 Chapter 6

in no position to bene¬t from them (because she was unconscious), had no depen-
dants to support and had been awarded adequate damages to cover the cost of
caring for her.
An extreme example of the ˜full compensation™ principle at work is Davies v.
Whiteways Cyder 49 which was an action under the Fatal Accidents Act 1976. The
deceased in this case was a wealthy man who had made gifts of some £40,000 to a
child; his death within 7 years of the gift meant that estate duty of some £17,000
became payable. It was held that the duty was recoverable as an additional item of
damages. The decision becomes even more remarkable when it is noted that a risk of
this nature is commonly insured against by persons who have made substantial
capital donations in their lifetime, and that if in this case there had been any insur-
ance to cover the contingency, it would not have been deducted from the damages
(by virtue of what is now s. 4 of the Fatal Accidents Act 1976). This is, no doubt, an
exceptional case, but the principle for which it stands “ that the injured are in general
entitled to be fully compensated for their losses “ is applied generally throughout the
law of tort and contract, and is seen as a corollary of the fault principle.
Because the tort system “ alone amongst compensation systems and schemes “
professes to provide full compensation, every tort victim who actually succeeds in
obtaining damages is “ as compared with the great majority of injured and disabled
persons “ exceptionally well placed in a ¬nancial sense. It is against this background
that one has to judge the desirability of continuing to adhere to the ˜full compensa-
tion™ principle. And it is also against this background that proposals for the exten-
sion of the tort remedy (e.g. by expansion of strict liability) have to be considered.
Every extension of the tort system means a small increase in the proportion of per-
sonal injury victims obtaining ˜full compensation™ (estimated at some 6.5% by the
Pearson Commission). There will thus be a few more winners in the ˜forensic
lottery™; but for the losers these extensions of tort liability will, of course, do nothing.
Damages are customarily awarded under two broad heads: ˜special damages™ and
˜general damages™. This distinction is based on the di¬erence between losses that are
precisely measurable and quanti¬able and those that are not. Special damages are
con¬ned to out-of-pocket expenses and loss of earnings incurred before the trial.
Damages for expenses and loss of earnings likely to be incurred in the future, plus
damages for non-pecuniary losses “ whether incurred before or after the trial “ such
as pain and su¬ering and loss of amenities, are awarded together as general damages.
It used to be customary to award a global ¬gure for general damages, so that it was
usually impossible to say how much was intended for future loss of earnings and the
cost of care, and how much for pain and su¬ering and loss of amenities.50 Now it is
usual to itemize damages, particularly in serious cases. The court itemizes the sums
it awards under the principal headings recognized by the law, especially as between
damages for future loss of earnings and expenses, and damages for non-pecuniary.

49 [1975] QB 262.
50 Watson v. Powles [1968] 1 QB 596.
Damages for personal injury and death 145

But in the most serious cases, it is now quite common for the judges to break the
damages down into smaller sub-headings. For example, speci¬c sums may be
awarded for estimated losses of future earnings, for estimated losses of pension
rights, and for the possible contingency that the claimant may become unemployed
and su¬er yet further income losses. Moreover, on the expenses side, it has become
quite common to itemize the di¬erent sub-headings under which the damages are
assessed; for instance, so much for nursing care, so much for home adaptations, so
much for other extra expenses, and so on.

6.2.1 Interest
The entitlement to full compensation is yet further enhanced by provisions as to
interest. Unless there are speci¬c reasons to the contrary, the court is obliged to order
the payment of interest51 on damages in personal injury and fatal cases.52 Damages
for losses and expenses incurred before the trial (˜pre-trial damages™ or ˜damages for
past loss™) carry interest at half the ˜special investment account rate™, which is set peri-
odically by the Lord Chancellor.53 Interest is payable because in theory the entitle-
ment to damages arises at the date of the injuries or, at least, the date the writ is issued.
The award of interest compensates the claimant for being ˜kept out of their money™.54
Interest is awarded at half rate because special damages represent sums that have
been lost over the whole period between injury and the trial, some closer to the date
of injury and some closer to the date of trial. A rough and ready approximation of
the amount due is arrived at by awarding half the appropriate interest rate.
Damages for pain and su¬ering and loss of amenity also attract interest, cur-
rently at a rate of 2%.55 The reason why this ¬gure is so low is that, unlike damages
for pre-trial pecuniary loss, damages for pre-trial non-pecuniary losses are calcul-
ated in currency values current at the date of judgment. This means that in¬‚ation
between the date of injury and the date of assessment has been taken into account
in the basic award, and the interest rate need not include allowance for in¬‚ation, as
commercial interest rates do. Nevertheless, the ¬gure of 2% seems a bit low. The
discount rate “ in other words, the rate of return on the investment of damages for
future loss that is assumed in calculating the lump sum “ is currently 2.5%.

51 Simple, not compound: R. Bowles and C.J. Whelan, ˜The Law of Interest: Dawn of a New Era?™
(1986) 64 Canadian Bar R. 142. The Law Commission has recommended that in cases where the
amount on which interest is to be awarded exceeds £15,000, interest should be compound, not
simple. This proposal would have the greatest impact on the cases that take longest to resolve,
which tend to be medical negligence claims in relation to birth injuries, which also tend to involve
very large awards. See Law Com. No. 287, Pre-Judgment Interest on Debts and Damages (2004). The
Commission estimated that the proposal if implemented would add £20“25 million to the annual
cost of medical negligence claims.
52 Je¬ord v. Gee [1970] 2 QB 130; Cookson v. Knowles [1979] AC 556.
53 The Law Commission has recommended that the rate be 1% above Bank of England base rate:
Law Com. No. 287.
54 Or, perhaps more realistically in many cases, having to borrow money.
55 Wright v. British Railways Board [1983] 2 AC 773. The Law Commission has recommended that
this rate remain the same, and that interest on it be only simple regardless of the amount awarded
for past non-pecuniary loss.
146 Chapter 6

The award of interest acts as a discouragement to delay on the part of defen-
dants, especially in cases which are settled out of court. Damages for loss of future
earnings and for future expenses do not carry interest since, by de¬nition, these are
designed to compensate for losses that have not yet been incurred. In fatal cases
the position is much the same: damages for pre-trial losses carry interest at half the
appropriate rate, damages for bereavement would probably carry interest at 2%,
and no interest is payable on damages for future pecuniary loss.

6.2.2 Lost earnings and support
We have already seen how many predictions and guesses have to be made in assess-
ing damages for loss of future earnings (in a personal injuries claim) or loss of future
support (in a fatal accident claim). The ¬rst step in a personal injury or death claim
is to assess this probable continuing income loss, i.e. the £X per annum of earnings
or support which the claimant would have received but for the accident. This sum
(called the ˜multiplicand™) is then multiplied by a ˜multiplier™, which is a ¬gure some-
what less than the number of years for which the loss is likely to be su¬ered. The
typical starting point is the number of years for which the loss is likely to continue “
i.e. in a personal injury action until the claimant™s injuries cease to a¬ect earnings or
the injured person dies or retires. This ¬gure is then reduced partly because of ˜con-
tingencies™ (i.e. that the claimant might not have lived or worked so long or might
have lost earnings even if the accident had not occurred), and partly because the
claimant is going to receive not an income but a capital sum, which can be invested
to produce an income.56 The multiplier is not the product of precise calculation but
of estimation in the light of the facts of the particular case and of other comparable
cases. In some cases where the calculation of damages for loss of earnings requires a
more than usual amount of speculation (e.g. if the claimant is a young child) the
court may not use the ˜multiplier method™ but may decide directly on a lump sum.
Methods of this kind can produce signi¬cant variations in the pattern of awards.
The Court of Appeal does what it can to iron out the grosser deviations from the
norm, but it will not interfere with awards unless they are much too high or much
too low; appeals are often dismissed with the comment that the damages were on the
high side or the low side, and that the members of the Court of Appeal would them-
selves have awarded more or less, but that the award was not so far from the norm
that the court should interfere. It may be necessary for the Court of Appeal to take
this line, for otherwise there would be an incentive to ¬ght an appeal in every case
since there is always room for di¬erences of opinion on matters of this kind. But this
does not mean the result is satisfactory to the individual litigants, who are bound to
feel aggrieved; nor for the public as a whole, whose faith in the administration of
even-handed justice may be shaken by such an appearance of chance in the system.

56 Where the parties enter a structured settlement, the lump sum will be further reduced to take
account of the tax advantage of a structured settlement to the claimant, and also to compensate
the insurer for the costs of setting up and administering the settlement.
Damages for personal injury and death 147

Where the action is a dependency action (i.e. it is brought under the Fatal
Accidents Act 1976) a similar process is followed, except that it is necessary ¬rst to
assess the extent of the dependency. In other words, if a person earning £30,000
a year is killed, and they spent £3,000 a year on themselves and the rest on support-
ing dependants, their dependency would be £27,000 per annum. This is the ¬gure
that must then be multiplied by the ˜multiplier™. In a fatal case the starting point for
calculating the multiplier is the number of years the dependants would have been
supported by the deceased. Where there are several dependants “ typically, a surviv-
ing spouse and children “ the damages must be apportioned among the dependants.
In practice, the lion™s share of the award tends to be given to the spouse and relativ-
ely small shares to dependent children. This is because it is assumed that the spouse
will maintain the children and may therefore need the income from investment of
the lump sum (if not capital) for this purpose. It has, however, been objected that
if this is the justi¬cation, there is no reason why anything should be given to the
children at all.57 The social security system does not give grants or payments direct
to dependent children; the payments are made to the parent or person having the
care of the children, in the belief that most parents will spend the money on the
children™s maintenance.
In fatal cases, as we have seen, two types of action can be brought: a claim by
dependants under the Fatal Accidents Act 1976 and a survival action by the bene-
¬ciaries of the deceased estate under the Law Reform (Miscellaneous Provisions)
Act 1934. In the great majority of cases, the bene¬ciaries will be the same persons as
the dependants. In practice, this means that the surviving dependants can both
inherit the deceased™s property under the will and any sums paid to the estate as
damages under the 1934 Act, and also recover damages for loss of support under the
Fatal Accidents Act 1976. In most cases, the damages awarded to the estate will be
small “ mainly the deceased™s out-of-pocket expenses and lost earnings between the
accident and the death. The justi¬cation for allowing the dependants both to inherit
and recover damages for loss of support is that they would have bene¬ted from the
inheritance sooner or later even if the deceased had not been tortiously killed.
It is worth noting that damages under the Fatal Accidents Act 1976 can include
compensation for the value of the household services of a person killed by tortious
conduct, even if those services were performed gratuitously and no money will be
expended to replace them. Thus a husband may obtain damages for the value of
his wife™s lost services;58 and similarly, a child whose mother is killed may obtain
damages for the value of her services.59 So even though they may not have been in
employment, the death of a person who runs a house and cares for children can give
rise to quite large claims for damages. It has been held that it is the carer™s services
which must be valued, excluding any element of emotional or loving support; but

57 H. Street, Principles of Damages (London, 1962), 152“3.
58 Regan v. Williamson [1976] 1 WLR 305.
59 Hay v. Hughes [1975] 1 All ER 257.
148 Chapter 6

on the other hand, account must be taken of the fact that the services of a family
member who looks after other family members may be available 24 hours a day to
the family.60
There has been a certain amount of discussion of how to value such household
services.61 Two possible measures suggest themselves: replacement cost (that is, the
cost of hiring someone to perform the services) and opportunity cost (that is, the
amount the carer could have earned in paid employment). Neither measure is
entirely satisfactory. In some respects a paid domestic helper can never replace a
family member. If the person who performed the services has been out of the work-
force for a long time, or is unskilled, the opportunity cost measure might not be
very useful and will, at all events, only establish a minimum value for the services.
The courts appear not to have adopted either measure as an invariable rule, but
rather seek to assess the ˜reasonable value of the services™. In the absence of any
objective price-¬xing mechanism, this approach is essentially arbitrary.
If the life expectancy of a tort victim is reduced by the injuries, damages may be
recovered for loss of earnings not only up to the date of expected death but also in
respect of the years when, but for the injuries, the claimant would have been alive and
earning (the ˜lost years™). Since, by de¬nition, the claimant will have no personal
living expenses in the lost years, these are deducted from the award.62 The theoretical
justi¬cation for ˜lost years damages™ is the full compensation principle, but the main
function of the award is to provide support for dependants of the injured person after
death. For this reason, the estate of a deceased person cannot recover lost years
damages in an action under the 1934 Act “ the dependants can recover for loss of
support under the Fatal Accidents Act 1976, and any non-dependent bene¬ciaries of
the estate do not need the award (this is sometimes put by saying that such an award
would provide a ˜windfall™ to non-dependent bene¬ciaries). So if the deceased had
no dependants, the damages payable to the estate will be limited to the losses su¬ered
by the deceased between the date of the accident and the death. Here the law recog-
nizes that commitment to the full compensation principle does potentially over-
compensate. But this recognition has so far only a¬ected the rules governing fatal
cases. The living claimant can still recover substantial lost years damages even if there
are no dependants (although in such a case the amount deducted for living expenses
will be considerably higher than in the case where there are dependants); and a
claimant who has been severely and permanently disabled or even reduced to a per-
sistent vegetative state63 can recover full damages for loss of earnings for the rest of
their life even where there are no dependants, despite the fact that the injured person

60 Regan v. Williamson [1976] 1 WLR 305.
61 K.A. Clarke and A.I. Ogus, ˜What is a Wife Worth?™ (1978) 5 British J. of Law and Society 1;
N.K. Komesar, ˜Towards a General Theory of Personal Injury Loss™ (1974) 3 J. of Legal Studies 457;
F.J. Pottick, ˜Tort Damages for the Injured Homemaker: Opportunity Cost or Replacement Cost?™
(1978“9) 50 U. of Colorado LR 59.
62 Pickett v. British Rail Engineering Ltd [1980] AC 136.
63 See further 6.5.3.
Damages for personal injury and death 149

can make no use of the award personally (the cost of caring for the victim will con-
stitute a separate head of damages) and the award will eventually accrue as a wind-
fall to the non-dependent bene¬ciaries of the claimant™s estate.64
The full compensation principle is seen as a corollary of the basis of tort liabil-
ity “ that the defendant is a ˜wrongdoer™. It is not based on any notion of the pur-
poses for which damages are being awarded. By contrast, the idea that an award can
constitute a ˜windfall™ to those who bene¬t from it is based on the idea that damages
serve the purpose of meeting ¬nancial needs. Both of these lines of reasoning are
present in the law, but the relationship between them is yet to be worked out con-
sistently. It seems undeniable that the purposive approach is much more in line with
modern ideas about the role of tort law in a mixed economy. As it is, tort law pro-
vides very generous ¬nancial bene¬ts to a very few injured and disabled persons who
can prove fault. There is no justi¬cation for extending those bene¬ts to persons who
have su¬ered no physical or ¬nancial loss as a result of the fault.
Finally, another important application of the full compensation principle de-
serves to be noted here. Only a minority of the population are earners, but many non-
earners engage in productive activity (most notably, housework and childcare) and
many earners (especially women) combine signi¬cant amounts of unpaid work with
paid work. As we have seen, compensation for the value of unpaid work may be avail-
able under the Fatal Accidents Act 1976 where the unpaid worker is killed. Where a
person™s ability to engage in unpaid work is impaired by non-fatal tortious injuries,
the law™s original answer was to give male spouses an action (called the ˜actio per quod
servitium amisit™) against the negligent person; but female spouses could not sue in
respect of the unpaid work of a male spouse. This inequality of treatment was based
on outmoded ideas of the relationship between men and women, and the action for
services was abolished in 1982.65 Even before this happened the courts had allowed
unpaid domestic workers to recover damages for loss of the ability to perform
domestic tasks.66 It does not have to be shown that anything will actually be spent on
hiring someone to perform the tasks, probably because it is thought that damages
ought to be awarded even if the family decides to cope with the situation by doing
more around the house themselves. After all, such damages are in the nature of an
award for loss of income, not an award for expenses (to be) incurred. The Australian
High Court has recently held, however, that while account can be taken of loss of
capacity to perform domestic tasks in assessing ˜general damages™ (e¬ectively,
damages for non-pecuniary loss), such loss of capacity is not a separate head of
damages in its own right.67 The practical e¬ect of this decision will be to reduce
signi¬cantly the amount likely to be awarded on account of such loss.

64 Lim Poh Choo v. Camden AHA [1980] AC 174.
65 Administration of Justice Act 1982, s. 2. This section also abolished the employer™s action for loss
of the services of an employee.
66 Daly v. General Steam Navigation Ltd [1980] 3 All ER 696. Rea¬rmed in Lowe v. Guise [2002] QB
67 CSR Ltd v. Eddy [2005] HCA 64.
150 Chapter 6

6.2.3 Medical and other expenses
Since accident and emergency services are only available through the NHS,
most victims of personal injury obtain their initial medical treatment free of
charge;68 and no damages can be recovered for medical expenses if no expenses
have in fact been incurred. Damages would be recoverable in respect of NHS ser-
vices (such as dentistry) for which charges may be made. People are, of course,
entitled to seek private medical treatment if they wish, and if they incur expense
in doing so, the expense is recoverable as an item of damages in a tort action if it
was reasonably incurred “ and it is provided by statute that it is not unreason-
able to ˜go private™ just because precisely the same treatment is available free in
an NHS institution.69 Before the trial or a settlement is concluded, a claimant
cannot normally be sure that the defendant will be liable. If the claimant knew
that the defendant™s insurers would accept liability, they might be tempted to go
private. Otherwise, claimants might be wary of incurring the expense of private
treatment unless they had medical insurance. A survey of more than 600 suc-
cessful tort claimants found that a signi¬cant minority had opted for some
private medical treatment, but that very few received only private treatment. In
a majority of cases, only some of the cost of this treatment was covered by insur-
ance or damages received. Various reasons were given for going private: because
the service was quicker, or better than, or not available, through the NHS; or (in
more than a third of cases), in order to have an examination necessary for the
In some cases “ relatively few in number “ a claimant may be so severely disabled
or incapacitated that medical and nursing treatment may be required inde¬nitely,
or indeed for the rest of the injured person™s life. In these cases the claimant is entit-
led to damages for private treatment, for example, to employ a private nurse at
home, or to enable them to reside permanently in a private nursing home or insti-
tution. Even if such facilities are made available by the NHS, the claimant is enti-
tled to go private.71 Yet once the damages are paid over, there is no obligation to
spend them on private care; it is open to the claimant to ¬nd a place in a public hos-
pital or institution which levies no charges72 and use the money for some totally
di¬erent purpose. The court may reduce the damages if it feels convinced that the
claimant will spend substantial periods in a free state institution, for example,
where other suitable facilities simply do not exist.

68 All ninety of the cases studied in the Harris survey obtained free medical treatment under the
NHS: Harris 1984 Survey, 240“2.
69 Law Reform (Personal Injuries) Act 1948, s. 2(4).
70 How Much is Enough?, 46“8.
71 The position is di¬erent in relation to residential care services provided by local authorities:
Sowden v. Lodge [2005] 1 WLR 2129.
72 Although local authorities have statutory power to charge for residential care services, tort
damages are not available to meet such charges. This means that the cost of such care cannot be
recovered in a tort action: Sowden v. Lodge [2005] 1 WLR 2129.
Damages for personal injury and death 151

It is not at all obvious why, sixty years after the beginning of the NHS, we should
continue to subsidize those who seek private treatment in the way that the tort
system does. Why should persons with a tort claim enjoy private treatment at the
expense of the large proportion of the population who pay or contribute to liability
insurance premiums, when others desiring private hospital or nursing care must pay
for it themselves by taking out health insurance? It seems di¬cult to justify the
present position, and the Pearson Commission proposed that in future the expenses
of private medical treatment should only be recoverable if it was reasonable on
medical grounds for the patient to have private treatment.73 However, given the
ever-increasing pressure of demand on the NHS, such a proposal is perhaps unlikely
to be attractive to politicians. Indeed, in 1999 the Law Commission recommended
against changing the law in this respect.74 On the other hand, the Chief Medical
O¬cer has recently recommended that the rule not apply to medical negligence
claims against the NHS, and that the treatment needs of successful claimants should
be met by a ˜care package™ provided by the NHS.75
Other out-of-pocket expenses incurred as a result of an accident are recoverable
in the same way as medical expenses; for example, the cost of fares to attend an out-
patient department at a hospital, the cost of special medical appliances, or indeed
expenses that have nothing to do with medical costs, such as the cost of doing
repairs around the house which the claimant is no longer able to do personally, or
the cost of alterations to a house necessitated by a permanent disability. The sums
awarded for such losses are a signi¬cant item in some awards.
Finally, we should note that a claimant may recover damages representing the value
of nursing or domestic services provided gratuitously by friends or family members.76
Normally, tort compensation is given for losses su¬ered or expenses incurred, but in
this case the injured person incurs no expense (indeed, they receive a bene¬t) because
the services are provided for nothing. So the courts say that the compensation is given
on account of the fact that the tort has created a need for services. In reality, the loss is
su¬ered by the carer, but the law does not allow the carer to sue in their own name.
Instead, the claimant holds the damages ˜on trust™ for the carer which means, in e¬ect,
that they must be paid over to the carer. The justi¬cation for awarding such damages
is that if the injured person chose to employ a professional carer instead of relying on
a friend or relative, the cost of doing so could be recovered. Indeed, a common

73 Pearson Report, vol. 1, paras. 339“42.
74 Law Com. No. 262, Damages for Personal Injury: Medical, Nursing and Other Expenses, Collateral
Bene¬ts, 3.1“3.18.
75 Department of Health, Making Amends: A Consultation Paper Setting out Proposals for Reforming
the Approach to Clinical Negligence in the NHS (2003), 127“8. In November 2005 the NHS Redress
Bill was introduced into Parliament. The Bill enables the establishment of a special scheme for
dealing with medical negligence claims against the NHS; but it does not directly deal with the issue
of damages for private medical treatment. However, a person who accepts an o¬er of compensa-
tion under the scheme will normally not be able to bring a tort claim and so will, e¬ectively, be
unable to recover damages for private medical treatment.
76 Donnelly v. Joyce [1974] QB 454; Giambrone v. JMC Holidays Ltd (No. 2) [2004] 2 All ER 891.
152 Chapter 6

measure of damages in this context is the reasonable market cost of services equiva-
lent to those provided by the carer; but if the carer has given up paid employment to
look after the claimant, the wages foregone may set the amount of the award.
Such damages may not be awarded in cases where the carer is the defendant.77
There are two somewhat con¬‚icting arguments for this restriction. One is that if
D performs the services and then has to pay damages representing their value,
D e¬ectively bears the cost of the services twice. The other is that since the claim-
ant holds the damages on trust for D, the odd result is that the tortfeasor receives
damages in respect of the tort. The issues that arise here throw light on the rela-
tionship between the principles of personal responsibility which underlie the rules
of tort liability and the impact on those rules of widespread liability insurance.
On the one hand, it seems contrary to the very basis of tort law e¬ectively to com-
pensate a tortfeasor for loss su¬ered by the tortfeasor as a result of the tort, whether
or not D is insured against the loss. On the other hand, if D is insured against lia-
bility, D will not personally pay any damages awarded to C; and it may seem hard
not to tap into that insurance to recompense D for their generosity. Perhaps the
strongest pragmatic reason to allow recovery is so as not to discourage provision of
care by family members in preference to professional carers where this is felt to be
more appropriate. For such reasons, the Law Commission has recommended leg-
islation allowing damages to be awarded in respect of care gratuitously provided
by the defendant.78

6.3 Full compensation for lost ˜earnings™: is it justi¬ed?
Apart from the issue of ˜windfalls™, which was mentioned in the previous section,
there are two problems which deserve detailed examination. The ¬rst is this: why
should accident victims be compensated for the same type of injury on a scale that
varies according to their previous level of earnings?79 If two people su¬er identical
permanent disabilities, but one was formerly earning £20,000 a year and the other
was earning £40,000 a year, what justi¬cation is there for compensating the latter at
a higher rate than the former? Or, still more striking perhaps, if these two people are
killed in similar accidents, what justi¬cation is there for compensating their depen-
dants at di¬erent rates? This we might call the problem of the earnings-related prin-
ciple. The second question is whether it is sensible or desirable to attempt to replace
every penny of lost income rather than some proportion of it. This we might call the
problem of the hundred-per cent principle.

77 Hunt v. Severs [1994] 2 AC 350.
78 Law Com. No. 262, 3.67“3.76.
79 In the case of non-earners the question is slightly di¬erent: why should the damages awarded be
assessed according to the earnings of a person doing for gain what the injured person was doing
Damages for personal injury and death 153

6.3.1 The earnings-related principle
The main advantage of earnings-related bene¬ts over ¬‚at-rate bene¬ts is that they
enable accident victims to maintain an approximation to their former standard of
living. To people who have long-term commitments such as mortgages, hire-
purchase instalments and so forth, real hardship can be caused by a sudden and
substantial drop in income. Moreover, ¬‚at-rate bene¬ts have the disadvantage that
a single ¬gure has to be selected for all earners, and it is almost inevitable that a low
¬gure will be inadequate for many to maintain their commitments, while a higher
¬gure will result in over-compensation for lower earners. Nevertheless, there are
real problems of equity in supporting the earnings-related principle, and these
require some consideration.
The social security system, as we shall see more fully later, is largely based on a ¬‚at-
rate principle, though there are some earnings-related bene¬ts. But it is important to
observe that earnings-related bene¬ts are only payable on a contributory or insur-
ance principle. In general, bene¬ts are the same for all. Whatever obligations may rest
on the State to see that its citizens do not want for the necessities of life, or even to
see that they have a reasonable standard of living, it is not obvious that the State owes
any obligation to maintain disabled persons (or the dependants of deceased persons)
for the rest of their lives at the standard of living which they had previously enjoyed “
at any rate, it is not obvious that this is equitable regardless of how the compensation
is paid for. In a competitive and partly market-oriented society the £40,000 a year
person receives, while working, a higher salary than the £20,000 a year person, pre-
sumably because the former is thought to provide more valuable services than the
latter. Once this person has ceased to work, this justi¬cation is no longer open. It is
not easy to justify a system under which many taxpayers would have to support a
non-working disabled person, or the dependants of a deceased person, at a standard
of living higher than their own. The only way in which this could be supported would
be by arguing that the higher income taxes paid by the wealthier person while
working justify a right to greater compensation when unable to work. This argument
proved acceptable in New Zealand;80 but in Britain it has generally been thought that
the mere fact that income taxes are progressive would not justify the payment of
earnings-related bene¬ts out of general taxation. For one thing, indirect taxes (which
represent a signi¬cant proportion of total tax revenues) are not progressive. On the
contrary, they may be said to be ˜regressive™ because they tend to consume a greater
proportion of a person™s income the lower that income is.
On the other hand, there is no objection to paying some or even full compensa-
tion for lost earnings on an insurance principle; that is, in accordance with prem-
iums actually paid. If a person earning £30,000 a year chooses to spend a substantial

80 Where the Accident Compensation Act provides for compensation of 80% of lost earnings up to
a maximum set at a high level, while not exacting any earnings-related contributions except from
the self-employed.
154 Chapter 6

part of that income on life insurance so that their dependants may enjoy the same
standard of living after they die as before, the person is free to do so; and the same
is true if that person takes out insurance against the risk of disability or chronic
disease. In practice we know that insurance against the risk of serious disease or
accident is not very common even among the relatively well-to-do, and is certainly
very rare among poorer people. This partly explains the fact that statutory sick pay
(a social security bene¬t) used to be moderately earnings-related; and that there is
an earnings-related supplement to long-term incapacity bene¬t (although this is
being phased out).81 However, such earnings-related social security bene¬ts were
never ¬nanced out of general taxes, but out of earnings-related National Insurance
contributions (according to what might be called ˜the contributory principle™). The
person who earned more got larger bene¬ts only as a result of paying higher
National Insurance contributions. There was no question of the taxpayer paying for
earnings-related bene¬ts.
When we turn to the tort system, however, things are very di¬erent. Here we
¬nd the only systematic method of compensation which pays (what are in e¬ect)
earnings-related bene¬ts without earnings-related contributions. How did this
come about? The answer is that the tort system operates, in practice, in conjunc-
tion with liability insurance and not ¬rst-party insurance. A system of liability
insurance cannot adjust its premiums according to the income of those to whom
compensation will be payable, because at the time the premiums are ¬xed nobody
knows to whom compensation may become payable under the policy. If we look,
for instance, at the road accident ¬eld we ¬nd that the liability insurance premium
is adjusted according to the risk presented by the insured, the only person that the
insurance company knows anything about. If the insured is a high-risk driver (e.g.
a young male), driving a high-risk car (e.g. a sports car), living in a high-risk area
(such as London), they pay higher premiums. But the premiums will not be
adjusted according to the income of the insured because compensation will never
be payable to the insured for loss of income under the policy;82 indeed, the insured
is the one person in the world whose income is irrelevant to the risk undertaken by
the insurance company. The person whose income is relevant is the person who
may be run over and injured or killed by the insured; and that person is, of course,
not identi¬able when the insurance is taken out.
On the other hand, since the law does at present provide earnings-related com-
pensation, insurance companies have to consider the likely amounts payable under
the policy and adjust the premiums accordingly. In other words, the more compen-
sation that is paid for lost earnings, the higher insurance premiums must go, but the
incidence is spread among all insured persons and is not borne rateably according
to the incomes of those to whom compensation will eventually be paid. If all motor

81 See 12.4.3.
82 Comprehensive policies often provide for some accident payment to the insured, but this usually
takes the form of a ¬‚at-rate bene¬t unrelated to the insured™s income.
Damages for personal injury and death 155

insurance premiums (for personal injury) are thought of as an insurance pool, it can
be seen that higher-income groups draw much more out of the pool, but do not pay
correspondingly more into the pool. Our £40,000 a year person and our £20,000
a year person will pay the same premium if they present the same type of accident
risk; while if they are the victims of accidents, the former will receive much more
compensation in the form of lost earnings than the latter. In addition, of course, a
pedestrian stands to gain while not contributing anything at all to the insurance
fund. The tort system provides a stark contrast with other compensation systems in
this respect. Another consequence of the earnings-related principle is that a spouse
or civil partner who was wholly dependent on the other spouse™s or partner™s earn-
ings is entitled to be maintained for the rest of their life at a standard of living
scarcely below that which was enjoyed while the spouse was alive.83 Even if the sur-
viving spouse or partner is young, childless and well quali¬ed to work, they need not
do so; and if they do, this will probably be ignored in assessing damages.84 There
seems no reason why a young person should be maintained for the rest of their life
by an award of damages (paid by society in one way or another) simply because their
spouse or partner was killed through someone™s fault. It is surely not right that the
law should reward idleness and discourage gainful activity in this way.
In other areas of the law the position is quite otherwise. For example, a young
childless wife separated (or divorced) from her husband cannot obtain mainten-
ance from him without taking account of her own earning capacity, even where
he was the ˜guilty™ party.85 The National Insurance system generally gives no pension
to a bereaved spouse or civil partner with no dependent children unless over 45 at
the date of bereavement: if younger, the bereaved person is expected to earn their
own living.
A very di¬erent criticism of the earnings-related principle (which is, to some
extent, at odds with what has been said so far) is that the principle (even if linked with
the insurance principle) entrenches existing inequalities in our society. For example,
it creates a preference in favour of earners as against non-earners; in favour of higher
earners as against lower earners; in favour of men as against women (because on
average, women earn less than men); in favour of the ethnic majority as against ethnic
minorities (because, on average, members of ethnic minorities earn less than
members of the ethnic majority).86 It might be replied that even if one accepts that
such inequalities ought to be eliminated from society,87 it is not the job of tort law to
do this. However, this reply has force only if we assume what we are setting out to

83 It is immaterial that the surviving spouse may have a substantial personal fortune, unless it was
used to support the spouse: Shiels v. Cruickshank [1953] 1 WLR 533.
84 6.1.2.
85 See Domestic Proceedings and Magistrates™ Courts Act 1978, s. 3(2). Despite para. (2)(g), conduct
is relevant only in exceptional cases: P.M. Bromley and N.V. Lowe, Bromley™s Family Law, 9th edn
(London, 1998), 762.
86 R.L. Abel, ˜£s of Cure, Ounces of Prevention™ (1985) 73 California LR 1003.
87 It is clearly accepted, and indeed required by law, that women should be paid the same as men for
work of equal value.
156 Chapter 6

prove, namely the validity of the earnings-related principle. This principle is not an
intrinsic feature of tort law; its adoption was the result of judicial choice. It is now far
too deeply entrenched in the law for the courts to remove it, but there is no logical
reason why tort compensation has to be earnings-related. We could choose some
other principle, if we wished, which better re¬‚ected the fact that people with similar
disabilities have similar ¬nancial needs, and the judgment (if it be accepted) that the
law should seek to lessen rather than entrench certain social inequalities.

6.3.2 The hundred-per cent principle
There has never been any question but that tort damages for lost earnings are
designed to represent the full amount of the loss. Yet most other compensation
systems, especially social security systems (and in other countries, workers™ com-
pensation laws) generally reject the hundred-per cent principle. Our own social
security system generally pays bene¬ts well below the full amount of lost earnings.
Similarly, the New Zealand Accident Compensation Act provides for bene¬ts of
80% of lost earnings; and the Australian Committee of Inquiry recommended
bene¬ts equal to 85% of lost earnings.88 Moreover, in most compensation systems
there are minimum loss quali¬cations. For instance, no social security bene¬ts are
payable in this country for the ¬rst 3 days™ loss of earnings; and the smallest award
available under the Criminal Injuries Compensation Scheme is £1,000.
There are two main reasons for rejecting the hundred-per cent principle. One
is the cost involved, particularly at the lower end. Large sums can be saved by elimin-
ating entitlement to bene¬ts for the ¬rst few days of illness or by refusing compen-
sation for losses below a certain amount. There is no doubt that the hundred-per cent
principle, as applied in tort law today, is one of the principal factors leading to over-
compensation for minor injuries, and under-compensation for more serious cases.89
The Pearson Commission proposed that social security bene¬ts should be fully o¬set
against damages for lost earnings;90 but when a scheme for recovery of social secu-
rity bene¬ts from tort claimants was introduced in 1989, claims worth less than
£2,500 were exempted from its operation. The scheme was amended in 1997 to cover
all awards of compensation for personal injuries (but not death).91 In practice, this
reform may discourage tort claims in many minor cases. On the other hand, social
security bene¬ts are not set o¬ against damages for non-pecuniary loss, which rep-
resent a disproportionately large part of many small awards.
The Pearson Commission also proposed the elimination of claims for non-
pecuniary loss su¬ered in the ¬rst 3 months after the accident. If implemented, this
proposal would have a very signi¬cant e¬ect because it appears that damages for
non-pecuniary loss represent a much greater proportion of the damages paid in

88 Australian Committee Report, para. 343.
89 See further 10.6.
90 Pearson Report, vol. 1, paras. 467“76.
91 For details see 15.4.5.
Damages for personal injury and death 157

minor cases than in serious cases. Unfortunately, there seems no prospect of this
change being implemented in the foreseeable future. By contrast, statutory thres-
holds (and ceilings) on damages for non-pecuniary loss have been introduced in
most Australian jurisdictions in recent years.
The second ground for doubting the wisdom of the hundred-per cent principle
is its potentially negative e¬ect on the injured person™s incentive to resume work
(whether paid or unpaid). In general, it seems desirable that injury victims should
be encouraged to become as active as possible as soon as possible. It is true that
this problem is not as great in a lump-sum system as it would be under a regime of
periodical payments, where resumption of paid employment may lead to a reduc-
tion of the compensation payments: once paid, lump-sum damages cannot be
taken away. But so long as the claim remains unresolved (either by a court judg-
ment or settlement out of court), the injured person has an incentive to exaggerate
their incapacity for work, knowing that if their claim is successful, they will be com-
pensated for lost earnings up to the date the claim is resolved and perhaps into the
future. The law attempts to address this problem by requiring the injured person to
˜mitigate™ the loss (by e.g. returning to work as soon as possible) and by refusing
damages for any period during which the claimant ought reasonably to have
worked. But this solution depends on being able to distinguish e¬ectively between
the ˜malingerer™ and the person genuinely incapable of work.
Even if the doubtful assertion, that the hundred-per cent principle is a corollary
of the fault principle, is accepted, there are good reasons for rejecting it which are
given e¬ect to in both the main compensation systems other than tort law. It is
di¬cult to see why (e.g.) tort victims should not forego lost earnings for the ¬rst
3 days, as social security bene¬ciaries are required to do; or why there should not
be a lower limit on tort damages of £1,000, as there is on compensation under the
Criminal Injuries Compensation Scheme. It is ironical, to say the least, that in an
era when personal initiative and individual self-reliance are the common currency
of political discourse, the tort system should continue to adhere to a principle
which, in other contexts, is seen as inimical to these ideals.

6.4 Full compensation: the commitment in practice
Assuming, for the sake of argument, that the principle of full compensation is a
desirable one for tort law to pursue, the question remains of how well that princi-
ple is implemented in practice. Forty years ago it was said that ˜grave injustice
follows from the present practice of the judges in assessing future ¬nancial losses™.92
Two common, related criticisms of the practical operation of the full-compensa-
tion principle that have been made over the years deserve attention. First, it has
been said that that awards are too greatly reduced to take account of ˜contingencies™,
i.e. the possibility that even if the claimant had not been injured, the income lost

92 JUSTICE, Report on Trial of Motor Accident Cases (London, 1966), 30.
158 Chapter 6

would not have been earned because of illness, or unemployment or being involved
in another accident. Secondly, it has often been argued that too little use is made of
actuarial evidence in calculating the multiplier (the assumption being that if they
made more use of actuarial evidence, the discount for contingencies would be
smaller and, conversely, damages awards would be higher).93 Actuarial evidence is
statistical evidence about matters such as life expectancy, disease, unemployment
rates and so on, which takes account of factors such as age, sex, place of residence
and occupation. Traditionally, actuarial tables, as such, were inadmissible as evi-
dence because they were ˜hearsay™. Actuarial evidence could only be introduced by
calling an actuary as an expert witness. Courts were very unwilling to do this for
fear of unduly increasing the length and expense of trials. Another objection to the
use of actuarial evidence was that, being statistical, it does not take account of the
peculiar circumstances of the individual claimant.94 However, this objection is mis-
placed. On the one hand, the use of actuarial tables to deal with certain contingen-
cies (such as life expectancy) would not prevent the court also taking account of
particular aspects of the claimant™s situation in calculating the multiplier. On the
other hand, the contrast between statistical evidence and the individual™s personal
circumstances is misleading because it assumes that when a court speculates on
what the future will hold for a particular claimant, it can in a meaningful sense
predict that person™s future. In reality, the courts™ predictions are based on a sort of
non-statistical averaging based on the judge™s knowledge and experience of what in
fact happens to people in general and to persons like the claimant in particular. The
di¬erence between speculation based on actuarial evidence and speculation about
the claimant in particular is that the former is based on sound scienti¬c method-
ology whereas the latter is not.
In 1973 the Law Commission recommended the publication of actuarial mor-
tality tables suitable for use in personal injury and fatal accident actions,95 and these
(often referred to as the ˜Ogden tables™)96 ¬rst appeared in 1984.97 However, it was
not until the enactment of s. 10 of the Civil Evidence Act 1995 that these tables
could be used without having to call an actuary as an expert witness to ˜prove™ them.
In 1994 revised Ogden tables were published, dealing with matters such as illness
and unemployment as well as mortality. The third edition, published in 1998,
incorporated certain other variables not previously taken into account, such as the
injured person™s place of residence.
Use of the Ogden tables is not mandatory, and even the revised Ogden tables do
not cover all of the contingencies that may be relevant in personal injury actions.

93 This was on ground on which liability insurers resisted increased use of actuarial evidence. But
the claim that awards tend to be higher when such evidence is admitted has not been rigorously
94 E.g. Hunt v. Severs [1994] AC 350, 365 per Lord Bridge.
95 Law Com. No. 56, Personal Injury Litigation “ Assessment of Damages (1973), para. 230.
96 After Sir Michael Ogden, the chair of the working party that developed the tables.
97 The tables are now in their 5th edition. See P. Barrie, Personal Injury Law: Liability, Compensation
and Procedure, 2nd edn (Oxford, 2005), ch. 23.
Damages for personal injury and death 159

But while it is still open to a court to depart from the tables to take account of
matters they do not deal with, failure to apply them in relation to matters they do
cover is likely to be overturned on appeal.98
Apart from the adjustment for contingencies, another factor critical to the prac-
tical operation of the full-compensation principle is the so-called ˜discount rate™. The
assumption on which lump-sum damages are calculated is that the recipient will
invest the damages and meet future losses out of a combination of capital and inter-
est. The discount rate is the rate of interest which, it is assumed, the recipient will
be able to earn, after tax, by investing the lump sum. Commercial interest rates have
two components: an allowance for expected future in¬‚ation and a rate of return on
the investment. The discount rate relates to the latter component, called the ˜real rate
of return™ on the investment.99 So, for instance, if the in¬‚ation rate is expected to be
3%, an interest rate of 6% would yield a real return, before tax, of 3%. For many years,
the discount rate used in calculating damages was 4“5%. This came to be considered
unrealistically high; and in 1998 the House of Lords held that 3% was appropriate.100
In 2002, the Lord Chancellor exercised a statutory power to set the rate, and reduced
it to 2.5%. The reduction re¬‚ects a desire to enable recipients of damages to favour
security over high income by investing in Index-Linked Government Stock (ILGS),
thus protecting the capital from the e¬ect of in¬‚ation. It was estimated that this 0.5%
reduction of the rate would increase the cost of the personal injury compensation
system by around £169 million annually.101 The discount rate ¬xed by the Lord
Chancellor is not binding on the courts. But it has been held that a court would be
justi¬ed in adopting a di¬erent rate only in rare circumstances not contemplated by
the published reasons102 for the chosen rate.103
Despite the increasingly actuarial approach to assessment of damages and
reduction of the discount rate, there is some reason to think that judicial practices
generally, and use of the Ogden tables in particular, may result in awards of
signi¬cantly less than full compensation, especially in more serious cases involv-
ing long-term future loss of income. Recent research compared assessment
methods adopted by English courts with those followed by US courts.104 It was

98 Wells v. Wells [1999] 1 AC 345, 378“9 per Lord Lloyd.
99 Mallett v. McMonagle [1970] AC 168; Mitchell v. Mulholland [1971] AC 666; Cookson v. Knowles
[1979] AC 556. For this reason, the common criticism that courts ignored the ravages of in¬‚ation
in assessing damages was unfounded.
100 Wells v. Wells [1998] 3 WLR 329.
101 Department for Constitutional A¬airs, ˜Damages Act 1996: Analysis of the Impact of the
Prescribed Discount Rate of 2.5%™ (March 2002).


. 6
( 18)