. 9
( 18)


national nature of the pharmaceutical and medical devices industries, group claim-
ing has become a world-scale phenomenon; and the largest claims involve
hundreds of thousands of claimants. Perhaps the largest group action so far any-
where in the world has been that against government-owned British Coal on behalf
of former employees who su¬ered respiratory disease and vibration-related condi-
tions as a result of working in the defendant™s mines. Liability was established in
1996, and compensation schemes were established by the government. Appli-
cations to these schemes have now closed. More than 746,000 claimants have reg-
istered, and it is estimated that the total compensation bill will top £7.5 billion.
Although group claims are atypical, because of their size, value and high public
pro¬le, they have become a sort of lightning rod for fears and accusations that the
tort system is out of control. At the same time, others see group claims as a symbol
of the tort system at its best, empowering the injured and ill victims of corporate
negligence, greed and deceit. Either way, group claims are the shop window of the
tort system.

110 P. Toynbee, ˜Legal leeches are bleeding the NHS™ Independent, 28 February 1996, 13.
111 For accounts of some of the leading claims see Hodges, Multi-Party Actions, Part V.

Tortfeasors and insurers

9.1 Defendants
In legal theory, the victim of personal injury who wishes to make a tort claim can
sue either the person whose negligence actually caused the accident; or, where that
person was acting in the course of employment at the time the tort was commit-
ted, the victim may sue the employer who is vicariously liable for the employee™s
tort; or both may be sued. As a matter of law, the tort victim (except in limited
circumstances: 9.3) cannot sue the insurance company that has agreed to indem-
nify the tortfeasor or the employer against the tort liability. The insurer has com-
mitted no tort, and the only person with legal rights against the insurer is the
insured. But if we look at the matter from a more practical and realistic viewpoint,
we can see certain similarities between employers who are vicariously liable and lia-
bility insurers. Both may be legally liable for tort damages in the ultimate result;
neither of them is (usually) in any way personally to blame for the victim™s loss;
both of them can act as ˜loss distributors™ in the sense that they can pass the cost of
paying damages on to others, namely premium payers (in the case of insurers) and
customers, employees and shareholders (in the case of employers). From this per-
spective, the fact that the employer can be sued by the tort victim while the other
cannot is a technicality. But even technicalities can have practical consequences,
and there are some circumstances, as we shall see, in which there is an important
distinction between the liability of an employer and that of an insurance company.
Moreover, although the liability of employers is regarded as a part of the law of
torts, the liability of insurance companies is regarded as something standing
outside the tort system.

9.2 Individuals as tort defendants
Most tort claims for death or personal injury are made against insured individuals
or against corporations or bodies which, if they do not carry liability insurance,
have su¬cient resources to pay a substantial award of damages. Most individuals
could not a¬ord to pay a substantial damages award out of their own resources, and
in 7.2 we examined the argument that the fault principle is unjust to the extent that

Tortfeasors and insurers 223

it takes no account of the means of the defendant. But there is no reason in theory
why a claim should not be made against an individual who is backed neither by
insurance nor a ¬nancially substantial employer. For example, a driver injured in
an accident through the negligence of an uninsured pedestrian or cyclist may sue
the latter for damages. What would actually happen if the victim tried to bring such
an action and enforce recovery against the negligent defendant personally?
The ¬rst matter the claimant must consider is how to ¬nance the claim. The ¬rst
thing to say here is that since the sixth edition of this book was published, the
¬nancing of personal injury claims has been transformed by the abolition of publicly
funded legal aid for most such claims and the introduction of conditional fee agree-
ments (CFAs), more colloquially (but inaccurately) known as ˜no-win, no-fee™
arrangements. Under a CFA a quali¬ed lawyer can agree to handle a claim for a client
on the basis that if the claim fails, the lawyer will receive no fee for services; but also
that if it succeeds the lawyer will be entitled to a fee ˜uplift™ (i.e. an additional
payment) consisting of a certain percentage of the fee that could be charged for the
services rendered if the claim were not being handled under a CFA. Because the basic
rule of English law (as opposed to US law) is that a claimant whose claim fails is
required to pay reasonable legal costs incurred by the defendant in defending the
claim, CFAs can be supplemented by ˜after-the-event™ (ATE) insurance against poten-
tial liability for the other side™s costs if the claim fails. If the claim succeeds, the
defendant will normally be liable to pay not only compensation to the claimant but
also the claimant™s legal costs, plus the premium for the ATE insurance and the uplift
provided for in the CFA between the claimant and their lawyer. Another method for
¬nancing personal injury claims is ˜legal expenses insurance™, also known as ˜before-
the-event™ (BTE) insurance. This can be understood as a private form of legal aid: the
insurance policy covers the reasonable costs of making the claim, and the defendant™s
reasonable costs of defending the claim if the claim fails. Legal aid was not abolished
for medical negligence claims because they are considered to be so complex, di¬cult
and risky that lawyers are likely to be very unwilling to handle them on a conditional
fee basis.1 In the present context, we can ignore medical negligence claims, because
such a claim will always be made against in insured defendant or the NHS.
The ¬rst thing to note is that if the injured person has BTE insurance, it will ulti-
mately be the BTE insurer, not a lawyer, who decides whether to pursue a claim
against an uninsured defendant.2 Unless the insurer is reasonably con¬dent that the
defendant will be able to pay the claimant™s costs if the claim succeeds, it is unlikely
to be willing to back the claim. Assuming that the injured person does not have BTE

1 See the discussion in H. Genn, ˜Access to Just Settlements: The Case of Medical Negligence™ in
A.A.S. Zuckerman and R. Cranston eds., Reform of Civil Procedure: Essays on ˜Access to Justice™
(Oxford, 1995), 399“406.
2 Trades unions also support personal injury claims ¬nancially. Typically, these will be work-injury
claims against insured defendants. But although employers™ liability insurance is compulsory, it is
not universal. A trades union may be asked to fund a claim against an uninsured defendant and
might be more willing to do so than a BTE insurer.
224 Chapter 9

insurance, the question that must be addressed is whether the fact that the defen-
dant is an uninsured individual is likely to a¬ect the willingness of a lawyer to
handle the claim. In practice, unless the lawyer is willing to act on a conditional-fee
basis, the typical personal injury claimant will probably not hire a lawyer to pursue
the claim. Most tort claims for personal injuries are settled out of court, and in the
typical case where the defendant is insured against liability, the insurer will nor-
mally pay the claimant™s solicitor™s costs (including the uplift) and the ATE insur-
ance premium as part of the settlement. If the claim fails, the ATE insurer will pay
the defendant™s costs, and the claimant will only have to ¬nd any amounts not
covered by the CFA plus (perhaps) the ATE insurance premium.3 By contrast, if the
defendant is an uninsured individual, there is a serious risk that if the claim suc-
ceeds, the claimant™s lawyer will not get paid for their services. For this reason, it is
unlikely that a lawyer would be prepared to take on a personal injury claim against
an uninsured defendant on a conditional-fee basis unless, perhaps, the defendant
were a large, self-insuring corporation or a very wealthy individual.
Does the claimant have any other options? Besides CFAs, there are two other
avenues for ¬nancing personal injury claims: through a claims management
company (CMC), (sometimes disparagingly called a ˜claims farmer™) and through
a claims assessor.4 CMCs are essentially legal services intermediaries who organize
for an injured client the range of services and service-providers (legal, medical and
so on) needed to make a personal injury claim. They typically operate on a no-win,
no-fee basis. But because they are not quali¬ed lawyers, they cannot make CFAs
with their clients or charge an uplift. They may, however, require the client to buy
ATE insurance. There is no obvious reason why a CMC would be any more likely
than a lawyer to take on a claim against an uninsured defendant. Claims assessors,
like CMCs, are not quali¬ed lawyers, and so cannot represent clients in legal pro-
ceedings. Claims assessors, it seems, typically operate on a no-win, no-fee basis and
contract for a share of any compensation recovered.5 The service they o¬er is the
negotiation of settlements. But because they cannot initiate and conduct court pro-
ceedings on behalf of clients, they are unable to use a direct ˜threat™ of legal action
to give the defendant an incentive to settle.6 For the same reason, they might not be
in a good position to negotiate a settlement that made allowance for the assessor™s
entitlement, as against the client, to a share of the compensation. Once again, there
is no obvious reason why a claims assessor would be prepared to handle a claim
against an uninsured defendant unless the defendant were a substantial self-

3 ATE insurance may be expensive, and poorer claimants may have to borrow to pay for it in the
¬rst instance.
4 See generally Department for Constitutional A¬airs, The Report of the Lord Chancellor™s
Committee to Investigate the Activities of Non-Legally Quali¬ed Claims Assessors and Employment
Advisers (2000).
5 In US parlance, this would be called a ˜contingent-fee™ arrangement as opposed to a conditional-
fee arrangement. In the UK it is illegal for practising lawyers to enter contingent-fee arrangements.
6 All they can do is point out that the claimant may end up going to a lawyer if the defendant does
not make a reasonable o¬er.
Tortfeasors and insurers 225

insured corporation or a very wealthy individual. Finally, an injured person might
attempt to negotiate directly with the person thought responsible; but without
being able to wield the threat of legal action, this strategy is unlikely to be success-
ful, especially if the other person hires a lawyer.
Even if a claimant manages to overcome the funding di¬culty and recovers judg-
ment against, or negotiates a settlement with, an uninsured defendant, the next
problem will be to enforce the judgment. For a defendant who has no substantial
assets at all and only a modest income, the most e¬ective and simplest (although
perhaps not the most attractive) way of escaping liabilities is to ¬le a petition for
their own bankruptcy. The cost of doing this is relatively low, and once the petition
is granted, the claimant will get no money and will have incurred much trouble and
expense. If the defendant owns a house or has other assets such as a life insurance
policy with a cash surrender value, or a car with some secondhand value, the
claimant may be in a slightly better position. Bankruptcy of the defendant will
enable C to lay hands on some of the bankrupt™s property. But even so C will be faced
with di¬culties if, as is likely, the house is mortgaged, for in this event the mortgagee
has ¬rst claim on the proceeds. Moreover, C will have to get D out of the house
before it can be sold; this will take time, and if D is recalcitrant, further legal pro-
ceedings will be necessary to obtain an order for possession. When all this has been
done the mortgagee may insist on making a quick sale (at a price su¬cient to cover
the mortgage) even though C may think a better price could be obtained by waiting.
A bankrupt can be made to pay damages in weekly or monthly instalments out
of income rather than in the form of a lump sum. But in practice this procedure is
very hard to use satisfactorily. An undischarged bankrupt is entitled to retain
enough of their income to maintain self and family, and most people likely to ¬nd
themselves in this position (i.e. those with no capital assets) are unlikely to have
anything much to spare out of their income once they have taken what they need
for their maintenance. In any case, even if this procedure worked smoothly, the
bankrupt would be entitled to ask for discharge in due course “ long before the debt
would be paid o¬. If the bankrupt could show the indebtedness arose from some
momentary piece of negligence, and if they had co-operated satisfactorily with the
o¬cial receiver, the court would probably grant a discharge within a year or 18
months. Thus, if the judgment were for a substantial sum, the bankrupt would
probably escape paying more than a relatively small part of it.
If the defendant does not choose the bankruptcy option, there are several routes
by which, in theory anyway, the claimant might secure payment. Court orders for
the sale of goods or real property may be obtained and ˜executed™. If the defendant
has a bank or building society account, it may be possible to obtain an order, called
a ˜third party debt order™, requiring the bank or building society to freeze the
defendant™s account(s) and to pay a speci¬ed sum to the claimant out of the defen-
dant™s account.7 There are provisions to deal with cases in which the freezing of the

7 Civil Procedure Rules (CPR), Part 72.
226 Chapter 9

account causes hardship to the defendant or their family. An ˜attachment of
earnings order™ requires the debtor™s employer to deduct a speci¬ed sum from
wages each week and remit the money to the court o¬ce. These orders ¬rst became
generally available for the enforcement of judgment debts under the Admin-
istration of Justice Act 1970, which abolished imprisonment for debt.8 They had
not been an unquali¬ed success in matrimonial cases, where they had been avail-
able since 1958; but some members of the Payne Committee (on whose recom-
mendations the 1971 legislation was based) thought that they would be much more
e¬ective for short-term liabilities of a commercial character.9 There are many prob-
lems in enforcing attachment of earnings orders against a really recalcitrant debtor.
Frequent moves and changes of employment may be made. The debtor may simply
disappear, and the creditor may have no means of tracing them. The police will not
help because this is a civil matter, though it would be di¬erent if the debtor were
¬rst made bankrupt.
But even if all these di¬culties are overcome and the defendant complies with
the order, the procedure may not be a very e¬ective way of compensating the
claimant. In one case10 the claimant™s husband had been killed while a passenger
in the defendant™s car. The claimant recovered judgment for £7,100 against the
defendant, who was uninsured, and the court ordered payment at the rate of £10
per month. If regularly paid, this sum would doubtless be better than nothing.
But the payment of £10 per month never actually gave the claimant what she was
held entitled to, since it represented a payment of interest alone of under 2%,
while a judgment debt at that date carried statutory interest at 4%.11 Thus the
defendant™s indebtedness would have been increasing all the time, despite the pay-
ments. In this light, it is not in the least surprising that few claims are brought
against uninsured individuals. The di¬culties in the way of enforcing a judgment
by periodical payments are in practice enormous. On the whole this seems no bad
thing. It is questionable whether justice is served by requiring a defendant to
shoulder the burden of paying a certain amount a week out of income for the
inde¬nite future.
It may be said that we have underestimated the chances of enforcing a judg-
ment for a capital sum against an individual defendant. No doubt many people
have some capital “ for instance, a life insurance policy with a surrender value, or
savings in a building society or bank. And a tiny percentage of people own really

8 See now the Attachment of Earnings Act 1971. In 2004, more than 40,000 attachment orders were
made to secure payment of a judgment debt. Also in 2004, more than 300,000 ˜warrants of exe-
cution against goods™ were issued; and about 6,400 third party debt orders. But we do not know
how many, if any, of these various orders were issued in personal-injury cases.
9 Report of the Committee on the Enforcement of Judgment Debts (Cmnd 3909, 1969), para. 602.
10 Jones v. Lloyd, The Times, 22 March 1967. Insurance for this type of case was not compulsory until
11 There is some doubt as to whether statutory interest is due where the judgment is made payable
by instalments, but this is usually an academic question. Surprisingly, it does not appear to have
been settled by the Attachment of Earnings Act 1971.
Tortfeasors and insurers 227

signi¬cant capital investments. Furthermore it would be a mistake to think that
all or most tort claims are for huge sums. As we shall see later, a very large
proportion of tort claims are in practice settled for modest amounts up to a few
thousand pounds, and there must be many people who could raise such a sum
without bankruptcy or even serious ¬nancial strain.12 Yet despite this, it seems
that claims are rarely made against or paid by individual defendants out of their
own resources, except in one type of case. Where a modest amount of damage is
done in a road accident and the claim is no more than a few hundred pounds “
which usually means that the claim is for property damage only “ the defendant
may prefer to pay personally rather than ask the insurance company to pay. If the
insured is required to pay an excess of £100 or more under the policy and has a
valuable no-claims bonus, then it may seem preferable to pay a claim of such an
amount rather than make a claim. But where personal injury has been su¬ered,
there seem to be few cases in which individual defendants are asked to pay tort
claims out of their own resources, even where these are of amounts that many
could a¬ord to pay.
Several factors may contribute to this perhaps surprising result. First, people
who do have capital (and so are able to pay) are probably more likely to be insured
against less obvious risks of personal injury “ for example, to have a comprehen-
sive householder™s policy covering risks to visitors, or a comprehensive liability
policy covering liability for damage done by a dog or a child. Secondly, the most
serious injuries tend to be caused by machines; and the most common kind of
machine is the motor vehicle, which is required to be insured. Other kinds of
machines, and also injury-causing processes, are likely to be owned or operated by
employers or corporations. There are, of course, many household accidents, but
claims are rarely made in such cases even if the accident is covered by a house-
holder™s comprehensive policy. Thirdly, if the defendant is neither insured nor
backed by an employer, the claimant™s solicitor (if one is consulted) will very prob-
ably advise that it is pointless to proceed.
Whatever the reason, the fact remains that very few individual tortfeasors pay
tort damages out of their own resources. One survey found that fewer than 3% of
the claimants received any tort payment from individual tortfeasors.13 The Pearson
Commission was unable to provide any estimate of the number or proportion of
tort claims which are made against individuals,14 but it must be very small indeed,
probably only a fraction of 1%. The fact that the vast majority of tort payments are
made by insurers or corporations is of fundamental importance to a proper under-
standing of the tort system and of the way in which it could best be reformed.

12 The Criminal Injuries Compensation Board found that only a minute fraction of o¬enders would
normally be worth suing (CICB Third Report (Cmnd 3427, 1967), para. 21; CICB Seventh Report
(Cmnd 4812, 1971), para. 17), but there must be a certain proportion of negligent tortfeasors with
some assets.
13 A.F. Conard and others, Authomobile Accident Costs and Payment (Ann Arbor, 1964), 221.
14 Pearson Report, vol. 2, para. 61.
228 Chapter 9

9.3 Employers and corporations as tort defendants
Unless a private individual carries liability insurance, it will usually be a waste of
time and money to sue them. Not so if the tortfeasor is a ¬nancially substantial
body such as a corporation or other business enterprise, a local authority, an NHS
health authority or a government department. Anyway, even when not required to
do so by law,15 many such bodies protect themselves from the risk of tort liability
by purchasing liability insurance from a commercial insurer (that is, an insurer who
aims to make a pro¬t). Some, particularly professionals, may belong to what is
called an ˜insurance mutual™, that is an organization which, because it does not aim
to make a pro¬t, can provide insurance cover more cheaply than a commercial
insurer. Also, mutuals are often formed in the hope that they will be able to settle
claims more cheaply than a commercial insurer in terms of administrative costs.
A very large body may act as a ˜self-insurer™, which means that it does not buy insur-
ance from a commercial insurance company nor does it belong to a mutual insurer.
Such a body calculates that over a period of years the cost of dealing with any tort
claims made against it, plus the cost of paying compensation, will be less than the
cost of buying appropriate liability insurance. Self-insurance only makes sense if
the likely amount of any tort claim is small relative to the total resources (either
actual or potential) of the entity that will have to pay it, so that paying it will not
produce undue ¬nancial dislocation or lead to liquidation.16 An entity which can
borrow signi¬cant amounts of money easily (such as a large multi-national corpo-
ration) or which has access to public funds (such as a government department) may
¬nd it attractive not to buy insurance against legal liabilities, or indeed against other
losses (such as losses to their property). So far as personal injuries are concerned,
the Pearson Commission estimated that defendants who self-insured met about
12% of all tort claims, and paid about 6% of total tort payments in 1973.17
An alternative to self-insurance for a large organization that is a prime target for
litigation is to set up a ˜captive insurer™, that is, an insurance company owned by the
insured itself. London Transport did this in 1995 because it was dissatis¬ed with the

15 Since 1969 all employers (subject to a few exceptions) who carry on a business or who are
incorporated or unincorporated bodies, have been required to insure against tort liability to their
own employees (but not to others): Employers™ Liability (Compulsory Insurance) Act 1969. The
insurance must cover claims up to a total of £5 million arising out of any one occurrence. It is
not compulsory to insure against illness and injury su¬ered by employees while they are working
abroad; and the employer is under no common law (or statutory) obligation to arrange or to
advise the employee of the desirability of procuring such insurance: Reid v. Rush & Tompkins
[1990] 1 WLR 212. It is a criminal o¬ence not to comply with the compulsory road-accident
liability insurance legislation, but failure to comply with the compulsory employers™ liability
insurance legislation is not.
16 The principles underlying self-insurance can be illustrated by a choice between a lower and a
higher ˜excess™ on an ordinary motor insurance policy. An excess provision, under which the
insured bears the ¬rst £X of any claim, is essentially a form of self-insurance. People are prepared
to accept such provisions if they calculate that they will save money by agreeing to pay the excess
in the event of a claim in return for a reduced premium; and because the amount of the excess is
usually small relative to the wealth of the insured.
17 Pearson Report, vol. 2, table 119.
Tortfeasors and insurers 229

high level of premiums being charged by its commercial insurers. Unlike a mutual
insurer, a captive aims to make a pro¬t, but the insured is able to take the pro¬t.
So far as the law of torts is concerned, the liability of an employer (whether a
corporation or an individual, and whether to employees or others) is said to be of
two distinct kinds. First, it may be vicarious liability, that is to say, liability imposed
on the employer simply because the damage was the result of a tort committed by
an employee acting in the course of employment.18 Secondly, the liability may be
˜personal™ in the sense that the employer or corporation is itself responsible, regard-
less of whether any identi¬able employee was responsible. The distinction,
however, appears to have little practical importance; and it may well be that cases
tend to be classi¬ed as vicarious or personal according to whether the damage was
done by misfeasance or nonfeasance. In the former case, there is usually no
di¬culty in identifying a particular employee as responsible; in the case of nonfea-
sance, however, it may be that no one person is responsible. Indeed, the gist of the
complaint may be that the corporation has not nominated anybody as the person
responsible for taking the precautions that were neglected on the occasion in ques-
tion. We may therefore commit the solecism of speaking of ˜vicarious liability™ as
covering both these species of legal liability.
Ever since the doctrine of vicarious liability came into the law in around 1700 it
has been steadily expanded by the courts in two principal directions: the categories
of persons for whose torts such liability is recognized has been enlarged;19 and the
kinds of act for which vicarious liability can be imposed have been extended.20
There can be no doubt that in doing this the courts have been profoundly in¬‚u-
enced by the fact that imposing vicarious liability was a satisfactory way of secur-
ing the payment of compensation to injured claimants, without imposing crushing
liabilities on a negligent tortfeasors. Thus, whereas the doctrine of vicarious liabil-
ity was originally only used to render an employer liable for the acts of menial
servants under the employer™s direct ˜control™, it came in course of time to cover all
skilled and professional employees, however attenuated the control possessed by
the employer. Moreover, there has been a steady increase in the number of situ-
ations in which an employer can be held liable for the torts of an ˜independent
Originally, there was no vicarious liability for the torts of independent contract-
ors. This was because the main reason why the courts imposed vicarious liability “
the fact that the employer of labour was usually an organization better able to bear
the risk of loss or damage than the injured person “ did not necessarily apply

18 Suppose an employee intentionally in¬‚icts personal injury on another. In certain circumstances,
such conduct may be in the course of employment, and the employer may be vicariously liable
for it. The victim may also be able to claim under the Criminal Injuries Compensation Scheme.
The relationship between this scheme and the tort system is considered in ch. 15.
19 See P.S. Atiyah, Vicarious Liability in the Law of Torts (London, 1967), Parts II, III and VII. For more
recent developments see e.g. N.J. McBride and R. Bagshaw, Tort Law, 2nd edn (Harlow, 2005), ch.
20 Atiyah, Vicarious Liability in the Law of Torts, Part V.
230 Chapter 9

to the employer of an independent contractor. In one sense, the ˜employer™ of an
independent contractor included every individual who engaged a company to
perform some service: the individual client who engaged a builder to build a house;
the individual who hired a vehicle and driver; even perhaps the traveller who pur-
chased a railway ticket and travelled by train. On the other hand, the person who
engages the services of an independent contractor is often a substantial corpora-
tion. Instead of being a large building contractor employed by an individual client,
the independent contractor might be a small business employed by a local author-
ity; instead of being a large concern renting vehicles with drivers to individuals, the
independent contractor might be a single individual lorry owner who hires the
lorry with their services as driver to a large company; instead of being a railway
company which carries the public, the independent contractor might be a one-
person, car-hire ¬rm renting cars to a substantial company. In cases of this kind,
the independent contractor might well be an inadequate risk bearer, at least in cases
where the contractor has no liability insurance beyond what the law requires. For
this reason, the courts have been willing over the years to extend vicarious liability
to cover the torts of independent contractors in certain circumstances.
It may be objected that there is a danger that in our anxiety to compensate the
injured we will create a rule which works well where the contractor is a small busi-
ness and the employer a large enterprise, but badly when the contractor is a large
enterprise and the employer a small business or private individual. In fact this
danger has not been realized, and the explanation is probably that claimants o¬ered
two potential defendants, one of whom can easily bear the liability and one who
cannot, will choose the former. In other words, if both employer and contractor are
liable, and one is a large organization and the other is not, the injured person will
sue the former whether that party is the employer or the contractor. If both are
small organizations and neither is able to bear the liability, the chances are that the
injured person will go uncompensated. If both are large organizations, then the law
may have to decide which is the more appropriate body to bear the loss; but in this
case there will at least be no di¬culty in the injured person obtaining compensa-
tion from someone.
From the claimant™s point of view, multiplication of possible defendants is obvi-
ously a good thing because it increases the chance that someone will be able to pay
compensation or will be insured. But there are disadvantages; in particular, if, as will
often be the case, it is unclear which of a number of possible defendants will ulti-
mately bear a loss in the end, they may all take out insurance against the same poten-
tial liability, which is economically wasteful. Furthermore, any arguments about
who should ultimately bear the claimant™s loss may well take place between insur-
ance companies at great expense but without any real gain to society as a whole.
Employers and corporations are more likely than individual defendants to have
assets out of which to meet a tort claim and are more likely to be adequately
insured. But a business might be forced into bankruptcy by a tort claim of
signi¬cant size; and even large corporations can be plunged into serious ¬nancial
Tortfeasors and insurers 231

di¬culties by large or multi-party tort claims. Perhaps the most signi¬cant example
is provided by asbestos litigation which, in the USA in particular, has led many cor-
porations, large and small, into liquidation. The social impact of the liquidation of
large corporations can be enormous, and in both the USA and the UK it is possi-
ble for corporations to implement reorganization plans as an alternative to liqui-
dation in cases where the corporation™s liabilities exceed its assets (where, in other
words, the corporation is insolvent). The basic aim of such plans is to secure com-
pensation for tort victims while at the same time enabling the tortfeasor to con-
tinue its business.21 However, the reality is much less attractive than the theory. In
the USA, corporations have been able to enter ˜Chapter 11 bankruptcy™ (as the pro-
cedure for corporate reorganization is colloquially called) even though not insol-
vent, and in this way to limit the funds available for compensation of tort claimants
to an amount set aside for the purpose.22 The result may be that the fund is
exhausted before all present and future claimants are compensated. In Australia, the
corporation which had been the largest manufacturer of asbestos products over
many years attempted to protect itself by channelling its tort liabilities into a newly
created and custom-designed o¬shore subsidiary, and by creating a fund to meet
all its existing and future asbestos liabilities. The object was to provide ¬nancial
protection for the operations of the main trading corporation in Australia. The
fund proved inadequate after only a couple of years of operation. Negative public-
ity, government and union pressure, and a public inquiry forced the corporation to
pump large additional funds into the trust.
Such techniques for controlling corporate exposure to tort liabilities are particu-
larly objectionable in cases where, as with asbestos, it seems quite clear that corpo-
rations deliberately concealed known risks of the products they were producing
and marketing. The story is similar in the case of tobacco products, the crucial
di¬erence being that cigarette manufacturers, unlike asbestos manufacturers and
users, have been largely successful in defending themselves against tort liability.
Such glaring examples of successful concealment of health risks raise serious ques-
tions about the idea that the risk of tort liability provides incentives to take pre-
cautions against such risks.
Quite apart from the fact that vicarious liability may help to provide the injured
person with a defendant who can a¬ord to pay compensation, it also has certain
other advantages. The most important of these is that it may enable the claimant
to get compensation even where the identity of the person responsible for the
damage is unknown. A person is run over by a van bearing the name ˜Bloggs
Bakery™; the claimant can make a claim against Bloggs Bakery even though they
cannot identify the driver or the registration number of the van. A person who is
injured while unconscious on the operating table of a hospital may have no idea

21 See J.G. Fleming, The American Tort Process (Oxford, 1988), 250“1.
22 P. Spender, ˜Blue Asbestos and Golden Eggs: Evaluating Bankruptcy and Class Actions as Just
Responses to mass Tort Liability™ (2003) 25 Sydney LR 223.
232 Chapter 9

who was responsible for the injury; but if all those who might have been respon-
sible are employees of the same hospital authority, the patient can sue the author-
ity, and it is immaterial that they cannot identify any particular tortfeasor.23 A
person is injured in a factory by the negligence of a workmate who cannot be
identi¬ed; again vicarious liability can be imposed. This advantage of vicarious lia-
bility may be particularly valuable where the claimant™s argument is that some
(unidenti¬ed) person in the defendant™s organization ought to have taken pre-
cautions against the harm su¬ered, but no-one did. It is, of course, a requirement
of the imposition of vicarious liability that some employee of the defendant com-
mitted a tort against the claimant. This means that a claimant who seeks to obtain
compensation from an employer must establish two things: ¬rst, that a tort was
committed by an employee of the defendant; and, secondly, that the defendant is
vicariously liable for that person. The process is thus a double-barrelled one,
similar to that involved in liability insurance (is the insured liable and does the
insurance policy cover the liability?), and di¬erent from that involved, for
example, in the case of ¬rst-party insurance (where the only question is whether
the insurance policy covers the victim™s loss).
There are important respects in which vicarious liability di¬ers from liability
insurance. First, the obligations of a liability insurer under the policy are owed to
the insured, not to the claimant. As a general rule, this means that the claimant
cannot sue the insurer; but if the insured goes bankrupt or into liquidation, the
Third Parties (Rights Against Insurers) Act 1930 entitles the claimant to claim
directly against the insurer instead of having to claim as a creditor of the defend-
ant.24 The fact that the insurer™s obligation is normally to the insured also explains
s. 153 of the Road Tra¬c Act 1988, which is designed to prevent the proceeds of a
compulsory liability insurance policy being intercepted by creditors of a bankrupt
insured rather than being paid in full to the claimant. By contrast, a person who is
vicariously liable is under a legal obligation owed directly to the injured person.
Secondly, a liability insurer cannot recover the amounts paid out under the
policy from the insured. By contrast, an employer held vicariously liable for the tort
of another is legally entitled to seek to recover the amount of the liability from the
tortfeasor, whether employee or independent contractor. Indeed the tortfeasor is
legally liable to ˜indemnify™ the employer.25 In practice, however, an employee (as
opposed to an independent contractor) is not likely to be called on to indemnify an
employer.26 This point is pursued later when we look at the issue of subrogation
and its attendant problems.27

23 Cassidy v. Ministry of Health [1951] 2 KB 343.
24 The third party™s claim is subject to any defences the insurer may have against the insured: Lefevre
v. White [1990] 1 Lloyd™s Rep 569, 577.
25 Lister v. Romford Ice & Cold Storage Co. Ltd [1957] AC 555 (employee; the same rule applies to
independent contractors).
26 Atiyah, Vicarious Liability in the Law of Torts, 426“7.
27 See 15.3.
Tortfeasors and insurers 233

9.4 Insurers
The vast majority of personal injury claims arise out of circumstances in which
liability insurance is compulsory by law, and the Pearson Commission estimated
that in personal injury cases, 88% of claims were handled, and 94% of compensa-
tion payments were made, by liability insurers.28 Moreover, most other personal-
injury cases involve as defendants large corporations or public authorities who act
e¬ectively as self-insurers29 and can, for most practical purposes, be treated as
though they were insurers as well as tortfeasors. The e¬ect of this is that in most
situations only an insurer has a real stake in a tort claim, and most tort claims are
handled throughout by an insurance company rather than by the tortfeasor.
In practice, the great majority of tort claims are settled by agreement between the
claimant (with or without legal advice) and the insurance company concerned, and
only a tiny handful ever see the inside of a court. Of this tiny handful, only a very
small fraction ever get to an appellate court where issues of ˜law™, as opposed to issues
of ˜fact™, are likely to be discussed. This does not mean that the decisions of appellate
courts concerning liability for personal injuries are irrelevant to the settlement of
cases out of court, any more than statutory rules about liability for personal injuries
are irrelevant to that process. Cases are settled ˜in the shadow of the law™ as it were;
settlements are more or less in¬‚uenced by what the parties think would happen if
their case was tried by a court. On the other hand, the dynamics of the settlement
process are very di¬erent from the dynamics of the court room, and it is wrong to
assume that the result reached by an out-of-court settlement between a claimant
and an insurer will accurately re¬‚ect what a court would have decided in that case.
We will examine the settlement process in more detail in chapter 10.
Sometimes liability for personal injuries will be met by an insurer which has not
provided insurance against the risk of liability for personal injuries. This happens
most commonly where an injured person sues a ¬rm of solicitors as a result of
whose negligence (commonly in neglecting to start proceedings in time)30 they
have failed to recover damages for personal injury. A claim of this kind will be
handled by the solicitor™s liability insurers,31 and if the negligence is clear enough,
the claim will come to be treated very much as though it were an ordinary personal
injury action.32 Similar claims may occasionally be made, for example, against an
insurance broker through whose negligence a policy has expired; here too, the
broker™s insurer may take over the claim.33

28 Pearson Report, vol. 2, para. 509.
29 Self-insurers in respect of road accidents are required to lodge a security deposit of £500,000 with
the Accountant General of the High Court: Road Tra¬c Act 1988, s. 144.
30 The courts have a discretion to extend the period within which a personal injury claim must be
made, and the fact that the claimant can sue their (insured) solicitor is a factor the court can take
into account in deciding whether to exercise the discretion in the claimant™s favour: Hartley v.
Birmingham CC [1992] 1 WLR 968.
31 Solicitors are required to insure against liability up to a speci¬ed amount.
32 Paterson v. Chadwick [1974] 2 All ER 772.
33 See McNealy v. Pennine Insurance Co. [1978] RTR 285.
234 Chapter 9

9.5 The nature of liability insurance
Looked at from the point of view of the insured and the insurer, liability insur-
ance does not di¬er greatly from most other forms of insurance. Like other forms
of insurance, it is an agreement whereby, in return for a premium, the insurer agrees
to indemnify the insured against a loss “ in this case, certain types of legal liability.
On the face of it, the purpose of this type of insurance “ as of all others “ is to
protect the insured against some contingency. It is the insured who pays the cost of
the insurance because it is the insured who gets the bene¬t of it; the injured party
has no direct claim against the insurer. Nor is there any question of the insurance
company paying damages under the policy for the bene¬t of a claimant who cannot
establish a claim in law against the insured. Since the policy is purchased by the
insured and is an ordinary contract between the insured and the insurer, it is gov-
erned by the ordinary principles of the law of contract. So, for instance, if the
insured has obtained the policy by fraud or misrepresentation or has failed to
comply with the conditions of the policy, the insurance company should not be
liable even though the real su¬erer in such circumstances may be the claimant, not
the insured.
From the insured™s point of view, then, liability insurance is a protective
device. From the claimant™s point of view, it is a vital means of ensuring that
people injured by others™ torts obtain compensation; and in the context of per-
sonal injuries, this is how liability insurance is now viewed.34 Indeed, insurance
against liability to third parties in respect of personal injury is compulsory under
the Road Tra¬c Act 198835 and under the Employers™ Liability (Compulsory
Insurance) Act 1969; and this is not for the protection of drivers and employers
but for the protection of injured road users and employees.36 It is misleading to
think of tort law as being the primary vehicle for ensuring payment of compensa-
tion to accident victims, with liability insurance as an ancillary device to protect
the insured. It is more accurate to view insurance as the primary medium for the
payment of compensation, and tort law as a subsidiary part of the process. As we
have previously seen, one of the chief reasons why the great mass of personal injury

34 For a thorough analysis of the relationship between tort liability insurance and the goals of tort
law see G.T. Schwartz, ˜The Ethics and Economics of Tort Liability Insurance™ (1990) 75 Cornell
LR 313.
35 Under this Act it is also compulsory to insure against third-party property damage up to a total
of ‚¬ 1 million.
36 There are signi¬cant di¬erences between the compulsory motor and compulsory employers™ lia-
bility insurance schemes. For instance, motor insurance has to be for an unlimited amount of lia-
bility, whereas employers™ liability insurance only has to provide £5 million of cover for liability
˜arising from any one occurrence™. In the case of motor, but not employers™ liability, insurance the
insurer™s right not to pay if the policy was obtained by fraud or misrepresentation is limited. In the
case of both types of insurance, insurers are prohibited from inserting certain types of clauses in
the policy; but there are more prohibited clauses in the case of motor than in the case of employ-
ers™ liability insurance. Failure to take out motor insurance is actionable in tort as breach of statu-
tory duty, but failure to take out employers™ liability insurance is not. See generally C. Parsons,
˜Employers™ Liability Insurance “ How Secure is the System?™ (1999) 28 Industrial LJ 109.
Tortfeasors and insurers 235

claims arises out of road accidents and industrial injuries is that insurance is nearly
always available in these cases.
If this is the right way to view the relationship between tort liability and liabil-
ity insurance, certain di¬culties at once become apparent. First, why should the
protection of insurance cover be restricted to those cases in which the victim™s
injuries are the result of actions of someone liable in law? Secondly, as a compen-
sation mechanism, a system of tort liability coupled with liability insurance in
which everybody is e¬ectively insuring for the protection of everybody else, is vastly
more complicated and expensive than would be a system of insurance (which in
this book has been called ˜¬rst-party™ insurance) in which everybody insured
against their own losses without the interposition of tort liability.
It may help to appreciate the signi¬cance of these considerations if we pause for
a moment to compare liability insurance with ¬rst-party (or ˜loss™) insurance. The
risks covered by tort liability insurance could also be, and to a certain extent in fact
are, covered by ¬rst-party insurance. For example, a person who has a comprehen-
sive insurance policy on a car is insuring not only against the risk of incurring legal
liability but also against the risk of any damage to the car arising from any of the
events speci¬ed in the policy. Since one of these events will be the car being damaged
in an accident caused by the negligence of another driver, this risk may be covered
twice over: the owner of the car may claim either under their own insurance policy
or may make a tort claim against the other driver which will be met by that driver™s
insurance company. If both drivers carry comprehensive insurance policies, the risk
of damage to either car by the negligence of the driver of the other is covered twice.
The protection a¬orded by the tort claim and the tortfeasor™s liability insurance
is much less extensive than the protection a¬orded by ¬rst-party insurance. Lia-
bility insurance can only protect where there is legal liability, and this, as we saw in
Part Two, generally means that the insured must have been at fault. Of course the
more ˜strict™ tort liability is, the nearer liability insurance approaches to ¬rst-party
insurance, until it eventually reaches the point at which it is, in e¬ect, loss insur-
ance purchased by one person for the bene¬t of others.37 This is the basic position
with regard to liability arising from nuclear hazards. Tort liability here is (by
statute) so strict that in e¬ect, the owner of a nuclear installation is simply required
to provide ¬rst-party insurance for everybody else.
However strict the liability, it is still the case that the insured must be found,
and must (if the insurer insists) be proved to have been liable. But a person™s prop-
erty may be damaged or destroyed in ways other than by the liability-attracting
behaviour of someone else who carries liability insurance. For example, a car may
be damaged without anyone™s fault, or solely by the fault of the owner of the prop-
erty, or by the fault of someone who carries no liability insurance, such as a pedes-
trian or a cyclist. The average motorist who has invested many thousands of

37 S. Jorgensen, ˜Towards Strict Liability in Tort™ in F. Schmidt ed., Scandinavian Studies in Law 1963
(Stockholm, 1963), 25, 33.
236 Chapter 9

pounds in a car is clearly unwilling to run these risks and is willing to buy
insurance against them. On the other hand, although the motorist™s ¬rst-party
(˜comprehensive™) insurance covers many more risks than liability insurance, a
motorist will generally be better o¬ by claiming against another person and,
through that person, against a liability insurer, than by claiming against their own
loss insurer.
In practice, there is a very important di¬erence between personal injury and
damage to a car or other property, namely that people do not generally insure
themselves against the risk of personal injury. Although a person™s most valuable
asset is typically their own earning power, very few people insure against loss of, or
damage to, this asset. In the case of people of modest means, this is understandable
because they are already (in e¬ect) insured under the social security system, which
gives them a signi¬cant degree of protection. But even people whose standard of
living is a good deal higher than could be enjoyed on social security bene¬ts rarely
insure themselves against injury; and if they are injured, tort law and liability insur-
ance may be their only method of obtaining really substantial compensation. It is
not simply a question of preferring to claim against the other party to save a no-
claims bonus: it is a choice between tort liability and liability insurance or no com-
pensation at all other than social security payments, occupational sick pay or such
bene¬ts as may be available under a superannuation scheme if the victim retires
early as a result of the injuries. If personal income-loss and disability insurance
were as common as comprehensive insurance for cars, tort claims for personal
injury would probably decline enormously in number and importance;38 more-
over, people would be compensated for personal injury even if it occurred without
anybody™s fault, just as they are at present compensated under comprehensive
insurance policies for damage to their cars even though caused without anybody™s
If ¬rst-party loss insurance became the normal method of ¬nancing com-
pensation for personal injuries, the premium would be related to the value of what
is at stake (i.e. the earning capacity of the insured). Thus, just as the person with
the more expensive car pays more for comprehensive insurance (though not for
third-party liability insurance, because a more expensive car, as such, is no more
likely to damage another than a less valuable car), so the person with higher
earning capacity would have to pay more to insure against diminution or loss of
that capacity.
Another di¬erence between liability insurance and ¬rst-party insurance con-
cerns the extent of the insurance cover. The likely monetary size of legal liability
may be very di¬cult to predict in advance, but the insured may well want (or be
required by law to have) insurance to cover the liability no matter how great it turns
out to be. On the other hand, if the potential liability is unpredictable but could be

38 Although under present law, a tort claimant who received insurance bene¬ts would also be enti-
tled to recover tort damages and so, in e¬ect, bene¬t twice over: see further 15.4.3.
Tortfeasors and insurers 237

very large, insurers may not be prepared to grant unlimited cover.39 Where a person
takes out loss insurance, however, this problem does not normally arise because the
insured is much more likely to be able to predict the size of any possible loss, and
the insurer can agree in advance with the insured a ceiling on the insurer™s liability
under the policy.
A major di¬erence between liability insurance and ¬rst-party insurance is that
a liability insurance system is a¬ected by the law relating to contributory negli-
gence. Since contributory negligence reduces the liability of the tortfeasor, and the
insurance company is only bound to indemnify the tortfeasor against the insured™s
legal liability, it follows that the money actually payable by the insurance company
under a liability policy will be reduced by the contributory negligence of the injured
party. Most forms of ¬rst-party insurance do not work in this way because people
wish to protect themselves against the risk of their own negligence as well as against
the risk of other people™s negligence; and also the risk of pure accident. When
people insure houses against the risk of ¬re or cars against ˜all risks™, they do not
expect the insurance payout to be reduced if they are themselves partly or even
wholly to blame for any resultant loss. It is true that some insurance policies may
require the insured to take ˜all reasonable care™ or all ˜reasonable precautions™
against the occurrence of the risk, but it is usually only in extreme cases that con-
ditions of this kind prevent an insured person from recovering compensation
under a ¬rst-party policy.40 On the other hand, a policy may also specify particular
precautions which the insured must take, and if these are not taken, the insurer may
refuse to meet a claim under the policy.
Where both liability and ¬rst-party insurance cover the same risks, as in the case
of a comprehensive motor policy, the e¬ect of contributory negligence can, in
theory, be extremely complicated. For instance, suppose that A and B are involved
in a collision for which both are partly to blame, and that both cars, which are com-
prehensively insured, are damaged. In law A can recover from B and through B,
from B™s insurers, for the damage to A™s car to the extent that B is to blame for the
accident; and from A™s own insurer in respect of the balance of the loss (less any
excess which A is required to pay under the policy). Similarly B can recover part of

39 Large potential liabilities are often insured by several insurers, each of whom insures a certain part
of the potential loss. Another common practice is for insurers to protect themselves by ˜reinsur-
ing™, that is by ¬nding another insurer who will insure part of the ¬rst insurer™s potential liability
to the insured. Reinsurers may do the same thing by purchasing ˜retrocession™ insurance. Certain
syndicates at Lloyds su¬ered very large losses in the early 1990s by accepting highly risky reinsur-
ance business in respect of liability for asbestos-related diseases and for environmental pollution
in the USA.
40 A motorist who damages their own car by careless driving can normally recover under a com-
prehensive policy unless e.g. the driver had crammed eight people into a small car and as a result
lost control of it: Clark v. National Insurance Corporation [1963] 3 All ER 375. See also Stephen v.
Scottish Boatowners Mutual Insurance Association [1989] 1 Lloyd™s Reports 535. The issue of rea-
sonable care often arises in connection with claims for loss of property under holiday insurance
policies; and in this context, too, the courts are indulgent to the policy-holder. In Morley v. United
Friendly Insurance Plc [1993] 1 WLR 996 the phrase ˜wilful exposure to needless peril™ in a per-
sonal accident policy was interpreted generously to the insured.
238 Chapter 9

the compensation from their own insurer and part from A™s insurer. This appears
to be a wasteful and complex procedure, and for many years insurers preferred to
bypass it, if they could, by ˜knock-for-knock™ agreements. Under such arrange-
ments, insurers agreed, in e¬ect, to convert comprehensive insurance policies, so
far as they related to damage to vehicles, from liability-cum-loss insurance into loss
insurance alone. Thus A™s insurer paid A for the damage to A™s car, and B™s insurer
paid B for the damage to B™s car; this saved much trouble and expense. In any par-
ticular case, the e¬ect of the knock-for-knock agreement may have been less advan-
tageous to one insurer than the other; for example, if A was largely to blame for the
accident and the damage to B™s car was more serious than the damage to A™s car,
then the e¬ect of the knock-for-knock agreement was that A™ s insurer got o¬ much
more lightly than if the agreement had not existed, while B™s insurer came o¬ much
worse. But insurers calculated that in the long run these advantages and disadvan-
tages were likely to iron themselves out as between large insurers, and that if they
stuck to the letter of the law they themselves (and hence their customers) would
bear the cost of adjusting the compensation according to the degree of fault of the
However, an insurer has an incentive to enter knock-for-knock agreements only
if it sells both liability-only and comprehensive policies. The e¬ect of a knock-for-
knock agreement is that holders of liability-only policies are subsidised at the
expense of holders of comprehensive policies. In recent years, insurers selling both
types of policy have come under competitive pressure from insurers selling only or
mostly comprehensive policies, who were able to o¬er lower premiums. As a result,
most knock-for-knock agreements have now collapsed. The complexity and cost
which such agreements were designed to avoid have been mitigated by the adop-
tion between insurers of agreed formulaic approaches to the apportioning of blame
between the parties involved in various types of accidents.
We can see, then, that liability insurance is a much more cumbersome form of
insurance than personal insurance. It involves three parties (the victim, the
insured and the insurer) rather than two (the insured and the insurer), and it
involves two legal issues (the insured™s liability to the victim and the validity and
extent of coverage of the policy) rather than one (validity and extent of coverage
of the policy). Moreover, in practice, the issue of legal liability of an insured is
much more likely to raise di¬cult problems of proof and adjudication than the
issue of coverage under a policy. The latter may sometimes be an issue, but in the
great majority of cases of ¬rst-party insurance, no problem at all arises about
whether the risk falls within the cover granted by the policy. On the other hand,
it is only too common that di¬cult questions arise about the liability of an insurer
under a liability policy.

41 A major factor in the failure in 1972 of the V & G Insurance Company was that it thought it could
operate more e¬ciently without a knock-for-knock agreement: Report of the Tribunal of Inquiry
(HMSO, 1972).
Tortfeasors and insurers 239

9.6 Some problems of liability insurance
These features of liability insurance make it particularly problematic for insurers.
In principle, liability insurance premiums are based on an estimate of the likely
number and size of claims, plus the administrative costs of selling insurance, col-
lecting premiums and processing claims,42 plus an allowance for the insurer™s
pro¬t; on the other side, allowance is made for income which the insurer expects
to earn on investments. The calculation of the likely cost of claims can be particu-
larly di¬cult. The cost of claims is a function of three main variables, namely the
number of successful claims against policy holders, the amount of harm su¬ered
by claimants and the levels of damages awards. In the case of road and industrial
accidents, the ¬rst two variables do not present great problems. Given previous
experience from which the insurer can judge trends in accident numbers and sever-
ity of injuries, an insurance company can make reasonably sound estimates for the
future. Of course, insurers can be caught out by sudden, unexpected changes in
trends; but large ¬‚uctuations in these variables, in the case of road and industrial
accidents, are rare. Estimating these variables “ particularly the number of claims “
can, however, present much greater problems in the case of liability for diseases.43
An employer, for example, may be held liable to pay damages to employees who
contract diseases as a result of exposure to dangerous working conditions. Many
diseases take time to develop,44 and in some cases symptoms may not be noticeable
for 20 or 30 years after exposure. Employers and insurers generally may not be
alerted to the risk of claims until after the ¬rst successful claims are made. Even if
they do become aware sooner, and insurers start charging premiums to cover such
liability, it may be extremely di¬cult for them to ¬x premiums at adequate rates
because of lack of past experience. In such circumstances, insurers may be faced
with a large number of claims over a short period of time in respect of which no,
or only inadequate, premiums have been collected.45
The third variable “ the amount of damages awards “ can also cause consider-
able di¬culties in two ways. First, the size of awards is partly a function of the rules
governing assessment of damages. If the rules are changed without warning in such
a way as to increase considerably the level of awards, this can upset insurers™ calcu-
lations signi¬cantly. This problem is exacerbated by the fact that changes in
common law rules operate retrospectively; that is, they apply to all trials and set-
tlements after the date of the change, including those relating to events that
occurred before the change, even though the relevant insurance premiums, which
fund the compensation, were calculated several years previously when the law was
di¬erent. Examples are the introduction in 1974 of damages for gratuitous nursing

42 The cost of reinsurance is also relevant.
43 J. Stapleton, Disease and the Compensation Debate (Oxford, 1986), 130“3.
44 As it is put: they have a ˜long latency period™.
45 This assumes that what triggers the insurer™s liability under the policy is the making of a claim.
If the trigger is the incident which gives rise to the liability the insurer may have even greater
240 Chapter 9

services; in 1979 of damages for loss of earnings in the ˜lost years™; in 1980 of
damages for loss of ability to provide domestic care; and in 2000 of substantially
increased awards for non-pecuniary loss. Another example was the reduction of the
discount rate for the calculation of lump sum awards. Statutory changes in the law
are not, in strict theory, retrospective; but they may be so in e¬ect. For example,
when s. 4 of the Law Reform (Miscellaneous Provisions) Act 1971 greatly increased
the entitlement to damages of widows in Fatal Accident Act 1976 claims, the Act
came into force one month after it was passed. Similarly, when the law relating to
interest on awards of damages for personal injuries was changed in 1969, the e¬ect
was a substantial increase in the aggregate amounts which insurance companies
had to pay. These increases took e¬ect long before insurers were able to allow for
them by increases in premium rates.
The general problem that insurers may be confronted with claims under liabil-
ity insurance policies based on events that occurred years before the claim is made
and years after the premium for the relevant policy was calculated, is referred to in
terms of ˜long-tail liability™. Suppose, for instance, that a claim is made in 2006 in
respect of events that occurred in 1976. The premiums charged in 1976 might have
taken no account of the risk of such liability, and would certainly not have taken
account of changes in the law since 1976. Liability for traumatic accidents is ˜short-
tail™ because typically the harm occurs and the claim is made at the time of or
relatively soon after the harm-causing events. So far as personal injuries are con-
cerned, the most important forms of long-tail liability arise out of workplace illness
and disease, certain forms of medical negligence and liability for certain products,
notably drugs.
The long-tail problem arises under liability insurance policies that provide cover
on an ˜occurrence™ basis. This means that the policy covers liability based on events
that occur during the period of the policy regardless of when liability arises and the
claim on the policy is made. For instance, an occurrence-based policy current in
1976 will cover liability arising out of events that occur in 1976, whenever the lia-
bility arises. An insurer might seek to overcome the long-tail problem by selling
insurance on a ˜claims-made™ basis, covering only liability arising out of claims
made during the period of the policy. This would mean, for instance, that a policy
for the year 1976 would cover only claims made in that year, and a policy for the
year 2006 would cover only claims made in that year. This reduces greatly the risk
of legal and other changes (such as improvements in medical technology) that
might a¬ect the number and size of claims between the date when the premium is
calculated and the date of the claim because that period will probably only be a year
or so. So claims-made cover is good for insurers but bad for potential defendants
and claimants.
Suppose a potential defendant buys a claims-made policy in 2006, then retires
or goes out of business in 2007. Suppose further that a negligence claim is made
against the insured in 2016 arising out of events that occurred in 2006. Unless the
insured has purchased ˜run-o¬ cover™ in respect of claims for events in 2006,
Tortfeasors and insurers 241

whenever made, the insured may have to meet the claim personally or the claimant
may be unable to enforce the liability in whole or in part if the defendant lacks the
necessary resources. For this reason, claims-made cover might be thought socially
undesirable. The private insurance industry is not capable of resolving these
di¬culties; and suggestions are sometimes made for tax-funded compensation
schemes to deal with long-tail personal injury liability. The social importance of
this issue cannot be overestimated. Witness the current ¬nancial rami¬cations of
liability for long-past exposure to asbestos. Asbestos-related diseases have latency
periods of up to 30 years. It is anticipated that deaths and illness from past expos-
ure will not peak until 2020; and that the total compensation bill for asbestos-
related liability will top £8 billion.
Another source of di¬culty in estimating the third variable is in¬‚ation. This
a¬ects most components in damages awards “ compensation for loss of earnings
and for costs (such as medical and hospital expenses) resulting from the injuries
are a¬ected quite directly; and damages for non-pecuniary loss are now increased
more or less in line with in¬‚ation. In¬‚ation causes most di¬culty when the rates of
increase of wages and prices vary erratically from year to year, for this makes pre-
dicting the size of future awards almost impossible. But even when, as in recent
years, the pattern of in¬‚ation has become much more regular, predicting future
rates is always a risky activity. And yet it is essential when, say in Autumn 2006, an
insurer is calculating premiums for 2007, to calculate them in the light of expected
damages in 2007. The position is even worse than this because of the length of time
it takes to settle many personal injury claims. The premiums collected in 2007 must
be adequate to pay all claims based on liabilities arising in 2007, even though a sub-
stantial proportion of these claims will not be settled for 2, 3 or 4 years, and a small
number (probably of the larger claims) will still be outstanding in 5, 6 or 7 years.
When the claims are ¬nally paid, they will be paid in the light of wages and prices
at that time, just as a court assesses damages in the light of wages and prices at the
date of judgment. So an insurer, to be safe, may have to estimate in¬‚ation rates up
to 8 years into the future. Even in times of relatively stable currency values, this is a
daunting task. And because liability insurance is a competitive business, it is not
easy for the insurer to play safe and ¬x premiums on the assumption of larger than
expected increases in earnings rates and costs over the next 3 or 4 years. If it does
this, it may be undercut by other companies which assume a lower rate of in¬‚ation.
If, on the other hand, the insurer underestimates the rate of in¬‚ation, it can expect
large losses.
Accurate prediction of the future will be particularly important in cases where
a court makes a periodical payment order (PPO) ( When damages are
assessed as a lump sum, the insurer is able to close its books on the claim once the
damages have been paid. From that point, the risk of future in¬‚ation and other
adverse ¬nancial developments shifts from the insurer to the claimant. Similarly,
when a liability insurer makes a structured settlement ( based on a lump-
sum assessment, the liability insurer will typically fund the settlement by using part
242 Chapter 9

or all of the lump sum to purchase one or more ˜annuities™ from a life insurer, thus
shifting (part of) the ¬nancial risk onto the life insurer. Any remaining risk will pass
to the claimant. Once again, the liability insurer can close its books on the claim.
By contrast, an insurer against whose policy-holder a PPO is made will remain
liable, under the policy, to meet the order for as long as, and however long, the order
remains active. Even if the liability insurer chooses to meet (some of) its liabilities
under the PPO by purchasing annuities, the risk that the arrangements it has made
will not be adequate to meet its policy-holder™s obligations under the order will
remain with the liability insurer, which cannot close its books on the claim until the
PPO expires.
The problems for liability insurers of accurately predicting future liabilities are
aggravated by the fact that accounting principles require private insurance compa-
nies to balance their books from year to year. This means that in any particular year
the company must (in theory, at least) have su¬cient resources to meet all claims
under the policies current in that year, even claims that may be made in future years
and claims for losses that will be su¬ered in future years. In other words, the pre-
miums charged in any particular year must be su¬cient to enable the insurer to
build up ˜reserves™ to meet future liabilities under policies current in that year. If an
insurer signi¬cantly underestimates its future liabilities it may ¬nd itself without
su¬cient reserves to meet all its liabilities to policy-holders. If other insurers have
made better predictions it may, in practice, be impossible for the under-resourced
insurer to build its reserves by raising its premiums because it will be undercut by
its competitors. In that case, the under-resourced insurer may be in a position
where it is unable or, at least, likely at some future date to be unable, to satisfy claims
against it and, for that reason, to be insolvent.
This simple picture of a properly functioning, competitive insurance market in
which a well-run insurance company underestimates its future liabilities is com-
plicated by a phenomenon which has become known as the ˜insurance crisis™. The
main symptoms of this phenomenon are sudden, unusually (and sometimes very)
large increases in premiums for various lines of liability insurance. In Australia,
such an event occurred in 2002 a¬ecting mainly medical indemnity and public lia-
bility46 insurance. Various measures were taken by governments to deal with the
crisis, including a raft of reforms to the rules of tort liability, most notably those
a¬ecting assessment of damages. In Britain, the early twenty-¬rst-century crisis has
primarily a¬ected employers™ liability insurance.47 The largest expense facing lia-
bility insurers is the cost of claims. But it is unlikely that increases in the cost of
claims are a major part of the explanation of the typical insurance crisis because
the cost of claims does usually not increase su¬ciently far or su¬ciently fast to
explain the speed and magnitude of the increase in premiums that characterizes

46 E¬ectively, insurance against occupiers™ liability.
47 Department for Work and Pensions, Review of Employers™ Liability Compulsory Insurance, First
Stage Report (June, 2003); Second Stage Report (December, 2003).
Tortfeasors and insurers 243

insurance crises. An important part of the explanation is what analysts call ˜the
cyclical nature of the insurance market™. The liability insurance market has been
shown to experience successively ˜hard™ and ˜soft™ phases. For various reasons, what
economists call ˜barriers to entry to and exit from™ the liability insurance market are
relatively low. This means that it is relatively easy for new insurers to set up in busi-
ness when it seems that easy pro¬ts can be made, and for existing insurers to leave
the market when conditions are not so good. In periods when there is a lot of capac-
ity in the market (that is, when there are plenty of insurers o¬ering cover), compe-
tition may become very ¬erce, and this may force down premiums. If premiums
drop to uneconomic levels (in the sense that the income of the insurer is insu¬cient
to cover its expenses, including the cost of claims), and if this goes on for any length
of time, there will sooner or later come a point at which an adjustment has to be
made to the level of premiums to restore pro¬tability, and that adjustment may
need to be sudden and large. If, as a result of excessive competition forcing premi-
ums down, a smaller or less well-managed insurance company collapses, this may
aggravate the problem because like any market, the liability insurance market is
sensitive to changes in the balance of supply and demand. If the demand for liabil-
ity insurance remains constant, but the supply suddenly drops as a result of a
company failure, insurers remaining in the market may take the opportunity pre-
sented by the change in market conditions to increase premiums to make up for
past losses and uneconomically low premiums.48
Liability insurance crises are particularly problematic for activities in relation
to which liability insurance is either actually or e¬ectively compulsory. Without
adequate insurance cover, such activities are either illegal or extremely unattrac-
tive. But even in other areas, sudden increases in the cost of, or sudden decreases
in the availability of, liability insurance may a¬ect people™s willingness to engage
in socially desirable activities. The fact appears to be that because of the
signi¬cant role played by the tort system and liability insurance, many people
have come to think of such insurance as a sort of ˜essential service™ rather like
public utilities such as gas and electricity. However, an important di¬erence
between liability insurance and these public utilities is that the price and supply
of the latter is more heavily regulated by the state those of the former. In both the
USA and Australia, ˜tort reform™ legislation, designed to limit tort claims and
levels of tort compensation, has proved a politically popular response to insur-
ance crises, despite serious uncertainty about the role of liability law in causing
such crises.
If, as a result either of bad luck, bad management or the outworking of the insur-
ance cycle, a liability insurer becomes insolvent and is unable to meet its liabilities,
the potentially disastrous e¬ects for its policy-holders will largely be forestalled by
the operation of the Financial Services Compensation Scheme (FSCS), established

48 For a more detailed discussion of the insurance cycle see P. Cane,˜Reforming Tort Law in Australia:
A Personal Perspective™ (2003) 27 Melbourne ULR 649, 658“63.
244 Chapter 9

under Part XV of the Financial Services and Markets Act 2000 and administered by
the Financial Services Authority (FSA). The FSCS is funded by levies on ¬rms
authorized by the FSA to carry on business. In the cases of compulsory (motor and
employer™s liability) insurance, the FSCS pays 100% of any claim against the failed
insurer. In the case of non-compulsory liability insurance, it pays 100% of the ¬rst
£2,000 of any claim, and 90% of the remainder. So far this has not happened in the
UK; but neither have governments been prepared to regulate the insurance market
with a view to mitigating the e¬ects of its cyclical nature.

9.7 First-party insurance for the bene¬t of others
There is one important species of insurance which is a sort of cross between ¬rst-
party and liability insurance. If a person, for instance, a warehouse owner, a carrier,
a repairer or the like “ a ˜bailee™ in legal terms “ has possession of valuable goods
belonging to other persons, it is quite possible that the parties will wish to take out
some insurance beyond that provided by property and liability insurance. Liability
insurance would not normally satisfy the bailor because it would only provide pro-
tection if the goods were lost or damaged in circumstances in which the bailee was
liable in tort; and the latter would only be liable if at fault in some way. So if the goods
are stolen despite all precautions, or if they are struck by lightning or burnt despite
all ¬re precautions, there will be no tort liability and therefore, no insurance com-
pensation. On the other hand, the bailee would not be protected by the normal form
of property insurance taken out by the bailor, because this kind of insurance does not
prevent the bailor from suing the bailee in tort (although it does prevent the bailor
from recovering both types of compensation).49 How is this situation to be dealt
with? One possibility is for the bailee to exclude the bailee™s normal tort liability by
contract, i.e. to persuade the bailor to agree that the bailee is not to be liable for loss
or damage to the goods. But although possible, this is a dangerous course because in
practice the courts frown on this kind of exclusion and try to interpret such provi-
sions favourably to the bailor. There is also the possibility of such a clause being
ine¬ective under the provisions of the Unfair Contract Terms Act 1977 designed to
prevent the use of unreasonable exclusion clauses in certain classes of contract.
The second way of meeting the problem is for the bailee to take out property
insurance for the full value of the goods, partly to protect its own interest in the
goods and partly to protect the interest of (or, in other words, in trust for) the
owner of the goods.50 Under this form of insurance, either bailor or bailee may
claim against the insurance company, and tort liability once again drops out of the
picture. Nothing corresponding to this type of insurance exists in the case of injury
to the person. In fact the law places great obstacles in the way of taking out insur-
ance against bodily injury directly for the bene¬t of another person, as opposed to

49 The insurance company is subrogated to the claim of the bailor; see 15.3 concerning subrogation.
50 E.g. Hepburn v. Tomlinson [1966] AC 451.
Tortfeasors and insurers 245

taking out much more cumbersome liability insurance. In practice, few people
would wish to insure other persons directly against such risks, but an employer may
well wish to insure its employees. However, the doctrine of privity of contract
places serious di¬culties in the way of such insurance being legally enforceable by
the employee against the insurance company.51 Thus, if an employer takes out an
insurance policy designed to protect its employees against risks of bodily injury,
including those for which legal liability would not arise, the policy may be unen-
forceable by the employee, unless the requirements of the Contracts (Rights of
Third Parties) Act 1999 are met. Moreover, if an employer does take out such a
policy and the insurance company (as may be anticipated) does not take any tech-
nical point about privity of contract but pays under the policy, the payment may
be disregarded in any subsequent action against the employer. This will certainly be
the case where the employee is killed and the employee™s dependants receive the
insurance money, for it is explicitly provided by statute that bene¬ts to the depen-
dants resulting from the deceased™s death (and this includes insurance policies on
the deceased™s life) are to be disregarded in an action under the Fatal Accidents Act
1976.52 It is perhaps not clear whether the result would be the same if the worker
were only injured and not killed, but it seems probable that it would. If the
employer attempts to meet the absurdity of this position by inviting its employees
to give up their common law rights of action in return for better rights under some
insurance scheme which would cover them even for injury or death not caused by
fault, it will be found that this course is actually prohibited by law.53
The result of all this is that in the one area in which direct insurance against
bodily injury and death by one person for the bene¬t of another might have been
expected, the law has discouraged the possibility so strongly that, in practice, no real
alternative usually exists to liability insurance, on the one hand, and ¬rst-party
insurance, on the other. However, there are some cases in which such insurance may
be found despite these di¬culties, for example, an athletic club may insure its
members against the risk of personal accident.

9.8 The impact of liability insurance on the law
The widespread availability of liability insurance has had a profound e¬ect on tort
law as a mechanism for compensating for death and personal injuries.

9.8.1 Statutory provisions
The most obvious way in which this has happened is through the enactment of
statutory provisions. The Law Reform (Miscellaneous Provisions) Act 1934, for
instance, which enabled proceedings to be brought against the estate of a deceased

51 Green v. Russell [1959] 2 QB 226.
52 Fatal Accidents Act 1976, s. 4.
53 Law Reform (Personal Injuries) Act 1948, s. 1(3); Smith v. BEA [1951] 2 KB 893.
246 Chapter 9

tortfeasor, was a direct consequence of the system of compulsory road accident lia-
bility insurance introduced in 1930. The Law Reform (Contributory Negligence)
Act 1945 also owed much to the fact that, with the spread of liability insurance, it no
longer seemed reasonable to deprive a claimant of all right to damages merely
because they had been negligent. The abolition of the doctrine of common employ-
ment by the Law Reform (Personal Injuries) Act 1948 was at least partly due to the
complete change, which the practice of employers™ liability insurance had produced,
in the legal treatment of industrial accidents. The Law Reform (Husband and Wife)
Act 1962, which enabled husbands and wives to sue each other in tort, was largely
due to the feeling that it was unfair that a spouse was the one person who could not
claim damages from an insurance company in the event of injury in a road accident
through the negligence of the other spouse. Similarly, although the Congenital
Disabilities (Civil Liability) Act 1976 provides that a child who su¬ers pre-natal
injury may not sue its mother for negligence, there is an exception where the child™s
injuries were the result of a motor accident for which the mother was responsible.
Above all, the introduction by statute of compulsory liability insurance, ¬rst
for road accidents and later for industrial accidents, has changed the whole focus of
the law of torts from penalizing tortfeasors to compensating their victims. The system
of compulsory road accident liability insurance under what is now the Road Tra¬c
Act 1988 recognizes over and over again the fact that liability insurance is a method of
compensating accident victims, and cannot be treated as though it were merely a
device for protecting a tortfeasor against legal liability. One of the principal concerns
of the Road Tra¬c Act 1988 is that matters arising under the contract of insurance
between the insured and the insurer should not be allowed to defeat the claim of the
accident victim as against the insured. Thus, for instance, if the insured obtains a
policy by fraud or misrepresentation, the insurer™s normal legal right to avoid the con-
tract is severely restricted. If the insured breaks some condition in the policy relating,
for example, to the condition of the vehicle or the driver (e.g. that the vehicle must not
be driven in an unroadworthy condition, or that the driver must not drive while
drunk) this does not a¬ect the rights of the accident victim; although the insured may
be under a liability to indemnify the insurer against the damages which the insurer has
to pay. Similarly, a failure by the insured to comply with a condition of the policy as to
anything occurring after an accident, for example, that the insured must give notice
to the insurers within a stipulated time, does not a¬ect a third-party accident victim.
In fact, as we shall see shortly, the provisions of the Road Tra¬c Act 1988 are in prac-
tice supplemented by the extra-legal liability of the Motor Insurers™ Bureau, which
renders even less e¬ective many stipulations in insurance policies so far as concerns
the accident victim. In addition, since 2002 it has been possible for the accident victim
to claim directly against the insurer without ¬rst claiming against the insured driver.54

54 European Communities (Rights Against Insurers) Regulations 2002. However, the accident victim
stands in the shoes of the insured; and so the insurer is entitled to avoid paying by pleading against
the victim defences that would have been available against the insured but which would have been
of no avail if the insured had been sued instead of the insurer.
Tortfeasors and insurers 247

Employees, however, have no analogous right under the Employers™ Liability
(Compulsory Insurance) Act 1969.
Section 153 of the Road Tra¬c Act 1988, and the Third Parties (Rights against
Insurers) Act 1930,55 deal with the possibility that the defendant might become
bankrupt. Normally it is not permissible to take legal proceedings against a person
who has been adjudicated bankrupt or against a company in process of liquida-
tion; such claims must be made to the trustee in bankruptcy or the liquidator, who
will pay a dividend out of the bankrupt™s assets. But if there is an insurance
company standing behind the bankrupt, this process can be by-passed under s.
153 of the Road Tra¬c Act 1988, and the claimant can recover from the insurer
in the normal way. However, this provision did not solve all the potential prob-
lems. If the claimant sued the tortfeasor to judgment and the tortfeasor was
unable to pay the damages personally, the claimant could originally do nothing
to prevent the tortfeasor going bankrupt after the judgment.56 As a result, the
insurance money would have been claimed by the tortfeasor™s trustee in bank-
ruptcy or the liquidator, and the accident victim (whose injuries were the sole
reason for the payment of the insurance money) would have had to submit a
proof for the claim and rest content with a share of the insurance money along
with all the other creditors. This was changed by the Act of 1930, so that if a tort-
feasor becomes bankrupt, the tortfeasor™s rights against the insurance company
pass to the accident victim, who can sue the insurers direct. This once again rec-
ognizes that the purpose of liability insurance is to protect the accident victim and
not the insured.
A couple of important gaps in the provisions of the Road Tra¬c Act 1988
deserve mention. The compulsory policy need only cover the use of a vehicle by
drivers speci¬ed in the policy; it is not necessary to cover use of the vehicle by
anyone. If an insured owner allows a person not speci¬ed in the policy to use the
car, the owner is guilty of an o¬ence and may also be sued for damages for breach
of statutory duty in respect of death, injury and property damage resulting from
negligence of the uninsured driver.57 But this right of action will not be of much
use unless the owner is wealthy enough to pay the damages. In practice, however,
the damages in such a case will usually be paid by the Motor Insurers™ Bureau, which
we will discuss later. Another gap results from the fact that the requirement to
insure relates only to the use of a vehicle on a road. If the accident occurs in a car
park, for instance, an injured person may be without redress if the driver is unin-
sured or the policy does not cover what occurred.58

55 This latter Act is not limited in its operation to road accident cases.
56 The Law Commission has proposed that under the 1930 Act, the claimant should be able to sue
the insurer in the same proceedings: Law Com. No. 272, Third Parties “ Rights Against Insurers
57 Monk v. Warbey [1935] 1 KB 75; but the principle is not applicable to pure economic loss: Bretton
v. Hancock [2005] EWCA Civ 404.
58 E.g. Charlton v. Fisher [2002] QB 578; Cutter v. Eagle Star Insurance Co. Ltd [1998] 4 All ER 417.
248 Chapter 9

Parliament again recognized the crucial role of liability insurance when it estab-
lished the Legal Aid Scheme in the 1940s, under the terms of which an insured tort-
feasor was normally unable to secure legal aid as a defendant.59 The practice of
insurers conducting the insured tortfeasor™s defence was already established when
the scheme started, and it was thought undesirable to throw these costs on to the
taxpayer and thereby relieve the insurance companies. Yet, remarkably enough,
Parliament did just this when it created the NHS only one year before the Legal Aid
Scheme. Medical costs arising from road accidents are by and large borne by the
taxpayer and not by insurance companies and their premium payers. However,
from 2006 a scheme to recoup a proportion of NHS costs from tortfeasors will be
in operation.60
Another notable example of Parliament ignoring insurance considerations is the
provision contained in the Occupiers™ Liability Act 1957 to the e¬ect that an occu-
pier of land will be liable for the negligence of an independent contractor only if
the occupier was in some way at fault, for example, in selecting an incompetent
contractor or in failing to supervise the contractor adequately.61 Recognition that
the occupier is the person best placed to insure against personal injuries occurring
on the premises, and that contractors may be impecunious or may become impos-
sible to trace or identify, might have led to a di¬erent result, at least in the case of
business as opposed to private premises. Much the same criticism could be levelled
against the Animals Act 1971, which pays insu¬cient attention to possible exten-
sions of liability insurance. On the other hand, it is interesting and encouraging to
note that the Unfair Contract Terms Act 1977 requires the courts to have regard to
the availability of insurance (among other factors) in deciding whether certain
types of exclusion clauses are reasonable.62

9.8.2 The impact of insurance on the common law
What impact, if any, has the prevalence of liability insurance and, indeed, the fact
that it is compulsory in the great majority of cases in which tort claims are made,
had on the reasoning of courts in personal injury cases? This not an easy question
to answer, and it will help if we disentangle a number of separate issues. In the ¬rst
place, there is the question whether the fact that the defendant in any particular
case is or is not insured against liability is likely to a¬ect the outcome in that case.
The second question is whether the prevalence of insurance against certain types
of legal liability has a¬ected the rules of law or procedure in cases of those types.

59 Legal aid is no longer available for most personal injury claims.
60 15.3.
61 In Gwilliam v. West Hertfordshire Hospitals NHS Trust [2003] QB 443 it was held (over a strong
dissent by Sedley LJ) that the occupier of premises may be liable to a visitor for failure to ascer-
tain whether an independent contractor engaged to do work on the premises has liability insur-
62 See s. 11(4) of the Act.
Tortfeasors and insurers 249 Effect of insurance in the particular case
With regard to the ¬rst question, it seems reasonably clear that the presence or
absence of liability insurance in any given case is not a material factor in deciding
whether the claimant or the defendant is entitled to judgment. In the days when
personal injury cases were commonly tried by juries, it was a ¬rm rule that the
jury was not to be informed that the defendant was insured. This rule probably
does not apply to trial of an action before a judge alone,63 but judges can be
assumed to be aware of the law and practice of liability insurance (e.g. that it is
compulsory under the Road Tra¬c Act 1988). It has been vigorously asserted that
the judge must disregard the possibility that the defendant is insured. In Davie v.
New Merton Board Mills, an action by a worker against his employer, Viscount
Simonds said that it was ˜not the function of a court of law to fasten on the for-
tuitous circumstances of insurance to impose a greater burden on the employer
than would otherwise lie on him™.64 It seems unlikely that decisions in particular
cases by judges, either on issues of liability or assessment of damages, would be
a¬ected by whether or not the defendant had insurance against liability, if only
because judges probably assume in the normal case that the defendant is
It has been suggested that the best reason for rejecting as irrelevant the fact
that the defendant is or is not covered by insurance in any given case is that any
other rule would make it impossible to ¬x premium rates for liability insurance.66
However, liability insurance rates are ¬xed by insurers in the light of past exp-
erience as to the number of incidents covered by the policy for which legal liabil-
ity is likely to be incurred. If liability itself turned on whether insurance existed,
the extent of such insurance would merely be one more matter relevant to the
assessment of the risk. Of course, con¬ning liability to cases in which liability
insurance existed would greatly discourage this form of insurance, for many
people would not willingly pay premiums to cover a risk for which they would not
be liable if they did not insure.
Just as whether or not the defendant has liability insurance is ignored, so also,
whether or not the claimant has ¬rst-party insurance is irrelevant.67 In a personal
injury case, the established rule is that the claimant is permitted to keep insurance
proceeds and recover damages in addition.68 In a property damage case (where

63 Harman v. Crilly [1943] 1 KB 68.
64 [1959] AC 604, 627.
65 There is some empirical evidence from the USA which suggests that a jury is likely to award higher
damages against an insured defendant than against one who is uninsured and is likely to award
less than full compensation against an uninsured defendant: H. Kalven,˜The Jury, the Law and the
Personal Injury Damage Award™ (1958) 19 Ohio State LJ 158, 171. Jury trials are extremely rare in
personal injury actions in the UK.
66 Jorgensen, ˜Towards Strict Liability in Tort™, 29.
67 ˜In determining the rights inter se of A and B, the fact that one of them is insured is to be disre-
garded™: Lister v. Romford Ice [1957] AC 555, 576“7 per Viscount Simonds.
68 15.4.3.
250 Chapter 9

¬rst-party insurance is more common), the insurance company takes over the
claimant™s tort rights by subrogation.69 Possible in¬‚uence of practice of insurance on the common law
When we turn to consider the e¬ect of the practice of insurance on the develop-
ment of common (judge-made) law as opposed to the result in individual cases, it
seems very likely that it has been in¬‚uenced by the growing availability and use of
liability (and other) insurance.70 It is not easy to prove this proposition rigorously,
but it is surely the case that the steady expansion of liability for negligence during
the past hundred years or so is partly due to the fact that insurance enables judges
to give e¬ect to their desire to compensate claimants without imposing undue
hardship on defendants. It has been said that the standard of care required in the
law of negligence has been tightened up over the years partly as a response to the
prevalence of liability insurance.71 The tendency to ˜objectivize™ the standard of
care, and to ignore the personal characteristics of the defendant may also have been
in¬‚uenced by insurance considerations.72 The fact that more subjective considera-
tions are taken into account in deciding questions of contributory negligence than
in deciding questions of negligence73 seems to support the general proposition; as
does the fact that the courts nearly always ˜bend™ the law in favour of the claimant
in cases in which a trivial act of negligence has resulted fortuitously in serious per-
sonal injury.74
Judges do occasionally discuss the relationship between liability rules and liabil-
ity (and other) insurance. For instance, in one case, in which an injured rugby
player sued an amateur referee for negligent failure to enforce the rules of the game,
both the trial judge and the Court of Appeal thought it weighed in favour of rec-
ognizing a duty of care in such circumstances that a large rugby organization,
which was vicariously liable for the negligence of the referee, was able to insure
against liability.75 On the other hand, the possibility that players might purchase
personal accident insurance was also mentioned. By contrast, in a case in which an
injured boxer sued the boxing regulatory body for failure to require adequate ring-
side medical facilities, a duty of care was recognized in express disregard of any
¬nancial or insurance di¬culties this might have caused the defendant.76 It has
been held that an occupier of land owes a duty to visitors to the land to check that

69 In this type of case, the insured cannot keep both the insurance monies and the tort damages: if
the insurer pays out and the claimant recovers tort damages, the insured must pay the damages
over to the insurer.
70 But see J. Stapleton, ˜Tort, Insurance and Ideology™ (1995) 58 Modern LR 820, 824“8.
71 See e.g. Jorgensen, ˜Towards Strict Liability in Tort™, 53 ¬.
72 Lord Denning, perhaps the greatest English judge of the twentieth century, was certainly
in¬‚uenced by insurance considerations: Nettleship v. Weston [1971] 2 QB 691, 699“700.
73 2.5.1.
74 5.3.3.
75 Vowles v. Evans [2003] 1 WLR 1607 at [12].
76 Watson v. British Boxing Board of Control [2001] QB 1134 at [89].
Tortfeasors and insurers 251

an independent contractor employed to do work on the land has public liability
insurance cover.77 However, the duty was held not to extend to checking that the
policy was current or that it provided adequate coverage. Moreover, it has been held
that a school owes no legal duty to advise parents to insure their children against
personal injury su¬ered while playing sport;78 and that an employer owes its
employees no duty to advise them to take out personal accident insurance when
working abroad for the employer.79
Of course, although it seems very likely that the prevalence of liability insur-
ance has in¬‚uenced the development of the law, it does not follow, even in cases
such as road accident cases, where liability insurance is compulsory, that the
claimant will always recover. For example, problems of proving fault are still
signi¬cant, and the tort system is very far from being one of strict liability
backed up by liability insurance.80 On the other hand, defences “ especially
contributory negligence “ the e¬ect of which is to deprive an uninsured claimant
of compensation which would be paid by an insured defendant, remain an
important aspect of negligence law. In fact, tort law is, at bottom, a system of rules
and principles of personal responsibility for conduct and its consequences; and
although tort law could not operate as e¬ectively as it does as a compensation
system, and would probably not have developed as it has, without widespread lia-
bility insurance, the basis of tort liability is personal responsibility, not the avail-
ability of insurance. The fact that liability insurance was available only tells us that
the defendant could pay any damages awarded, not that D should be held liable to
pay compensation.
At the same time, however, it is surely the case that the size of damages awards
in personal injury cases is explicable only on the basis that judges are in¬‚uenced by
the widespread presence of insurance. It can hardly be supposed that judges would
habitually award thousands or tens of thousands of pounds in damages without a
thought for the e¬ect of such awards on the defendant, if they did not appreciate
that the damages would not be paid by the defendants themselves.81 The fact that
tort damages are usually paid by insurance companies does not necessarily make
the size of damages awards a matter of no concern. First, and at the most general
level, it can be argued that whoever pays tort damages, compensating people uses

77 Gwilliam v. West Hertfordshire NHS Trust [2003] QB 443. But contrast Naylor v Payling [2004]
EWCA 560, discussed in R. Lewis, ˜Insurance and the Tort System™ (2005) 25 LS 85, 108.


. 9
( 18)