. 132
( 155 .)


recommended that a remedy for wrongful administrative action should be
introduced by legislation, ˜which might take some such form as the following™
(para 11.83):

Subject to such exceptions and immunities as may be specifically provided, compensation in
accordance with the provisions of this Act shall be recoverable by any person who sustains
loss as a result of either:

(a) any act, decision, determination, instrument or order of a public body which materially
affects him and which is for any reason wrongful or contrary to law; or
(b) unreasonable or excessive delay on the part of any public body in taking any action,
reaching any decision or determination, making any order or carrying out any duty.

It is added that ˜wrongful™ and ˜public body™ would need to be carefully de¬ned.
(Note, however, the reservations as to this far-reaching proposal expressed by
Lord Woolf, Protection of the Public: A New Challenge (1990), pp 57“8.) To date,
no such remedy has been enacted into law.
One area of di¬culty in considering the liability in tort of public authorities
is the relationship in determining the limits of such liability between public law
concepts (ultra vires, irrationality, etc) and private law concepts (duty of care,
breach of duty, etc). Formerly, it appeared that a negligence action could
succeed against a public authority only if the authority had acted Wednesbury
unreasonably (see eg, Home O¬ce v Dorset Yacht [1970] AC 1004, per Lord
Diplock). Latterly, the courts appear to have relaxed this rule (see, especially,
Barrett v En¬eld London Borough Council [2001] 2 AC 550, considered below).
However, public law concepts have not been rendered wholly irrelevant when
considering questions of liability in tort and, in particular, in negligence. Where
the decision of the public authority is characterised as being ˜non-justiciable™,
for example, the authority will not be liable in negligence. Matters of justicia-
bility will be determined with reference to public law concepts.
Where it is argued that a public authority is liable in negligence (or, indeed,
in other torts) the detail of the statutory scheme under which the authority was
acting will be central to the determination of liability. A critical question will be
whether the authority was exercising a statutory duty or a statutory power.
Di¬erent lines of authority now apply in each of these categories. The leading
cases with regard to statutory duties are X v Bedfordshire County Council [1995]
2 AC 633 and Barrett v En¬eld London Borough Council [2001] 2 AC 550 (but
713 The courts: judicial review and liability

see also Phelps v Hillingdon London Borough Council [2001] 2 AC 619, which is
particularly important on vicarious liability). We shall turn to X and Barrett in
a moment. The leading authorities with regard to statutory powers are Stovin
v Wise [1996] AC 923 and Gorringe v Calderdale Metropolitan Borough Council
[2004] 1 WLR 1057. These latter cases make it clear that it continues to be the
case that litigants are very unlikely to be able to show that a public authority has
acted negligently in the exercise of a statutory power unless they can show that
the authority has acted Wednesbury unreasonably.
We turn now to the case law concerning negligence and statutory duties. In
X v Bedfordshire County Council (above) several claimants argued that their
local authorities had acted negligently inter alia in not investigating serious alle-
gations of parental abuse and neglect and in failing to commence appropriate
measures of child protection. The local authorities applied to have the claims
struck out. The House of Lords ruled in favour of the local authorities.
Negligence is composed of three elements, all of which need to be proved before
liability will be imposed: ¬rst, there needs to be a duty of care, secondly there
needs to be a breach of duty and thirdly the breach of duty needs to have caused
recoverable, non-remote damage. (This is the case in all actions for negligence,
whether against public authorities or not.) In X v Bedfordshire Country Council
the House of Lords focused on the ¬rst of these elements. Their Lordships held
that it would not be ˜fair, just and reasonable™ to impose a duty of care on local
authorities in respect of their responsibilities under child protection legislation
(for the ˜fair, just and reasonable™ test, see Caparo Industries v Dickman [1990]
2 AC 605, 617“18). A variety of overlapping reasons was o¬ered in support of
this conclusion: (1) a duty of care was a blunt instrument that would cut across
the whole statutory scheme; (2) the statutory scheme was inter-disciplinary,
involving multi-party, collective decision-making (potentially including
parents, teachers, social workers, educational psychologists, the police and
others), giving rise to a problem of who, in particular, should owe any duty of
care; (3) imposing liability would lead to problems of apportionment of
responsibility; (4) alternative remedies were available, such as complaints to the
local government Ombudsmen; (5) imposing a duty of care would risk encour-
aging defensive administration (whereby decision-makers make decisions prin-
cipally in order to escape liability, rather than making decisions which are
necessarily in the best interests of the parties); and (6) a ¬nding of liability
would impose a burden on scarce public resources, both ¬nancial and human.
Barrett v En¬eld London Borough Council (above) was a negligence case about
a child who was already in local authority care. Relying on decisions such as X
v Bedfordshire County Council the local authority applied to have Mr Barrett™s
claim struck out. The House of Lords ruled that the claim should not be struck
out. X v Bedfordshire County Council was distinguished. Their Lordships ruled
that the public policy considerations on which the House had relied in X in
deciding that it would not be fair, just and reasonable to impose a duty of care
in the circumstances of that case ˜did not have the same force in respect of
714 British Government and the Constitution

decisions taken once the child was in care™. Their Lordships ruled that Barrett
should be allowed to proceed to full trial. The key issue at that trial, their
Lordships thought, would be whether the local authority had breached its duty
of care to Mr Barrett.
Now, it may be that the di¬erences between X and Barrett can be explained
simply by the di¬erences in the facts of the two cases: the children in X were not
in care whereas Mr Barrett was. This, however, seems implausible. There is
a bigger shift taking place here than this explanation gives credit for. For one
thing, Barrett marks a more substantial step away from the equation in Home
O¬ce v Dorset Yacht between public law and private law concepts. To the extent
that X v Bedfordshire can be read as authority for the proposition that, in matters
a¬ecting discretionary policy, public authorities will not owe a duty of care (and
it may be that, in any event, this was always too broad a summary of their
Lordships™ decision in X), this proposition clearly now needs to be quali¬ed.
As Lord Hutton expressed it in his opinion in Barrett (at 583):

the fact that the decision which is challenged was made within the ambit of a statutory
discretion and is capable of being described as a policy decision is not in itself a reason
why it should be held that no claim for negligence can be brought in respect of it . . . It is
only where the decision involves the weighing of competing public interests or is dictated
by considerations which the courts are not fitted to assess that the courts will hold the issue
is non-justiciable.

Additionally, there is some signi¬cance to be attached to the fact that the focus
in Barrett was on the breach of duty issue rather than the duty of care issue. This
shift was at least partly attributable to certain rulings of the European Court of
Human Rights. In Osman v United Kingdom (1998) 29 EHRR 245, that court
had ruled that a decision to strike out a negligence action against the police
violated the right to a fair trial under Article 6. (The decision to strike the
case out was taken on the basis, established in Hill v Chief Constable of West
Yorkshire [1989] AC 53, that the police enjoy an immunity in such actions as
they owe no duty of care to the victims of crime.) The decision in Osman was
roundly condemned, the European Court of Human Rights being (rightly)
accused of misunderstanding the nature of the striking out action (see eg,
Gearty, ˜Unravelling Osman™ (2001) 64 MLR 159). Osman was decided after X v
Bedfordshire but before Barrett v En¬eld. Like Osman, both X and Barrett were
striking out applications. It is clear that, in Barrett, their Lordships were
concerned that a similar result to that in Osman could ensue in these cases. This
may explain something of the turn from the focus on whether there is a duty of
care to questions instead of breach of duty: whereas the former may be
addressed at the preliminary stage of an application to strike out, the latter will
ordinarily require a full trial.
As it turned out, it may be that their Lordships™ concern in Barrett was unnec-
essary. In the light of the extensive criticism of its decision in Osman, the
715 The courts: judicial review and liability

European Court of Human Rights changed its approach to striking out cases
and Article 6. Having lost in the House of Lords the claimants in X v
Bedfordshire took their case to Strasbourg. The Court of Human Rights ruled in
its judgment in the case (Z v United Kingdom (2002) 34 EHRR 3) that in Osman
it had misunderstood the nature of the striking out action and that, in the
present case, there had been no violation of Article 6. The Court of Human
Rights went on to rule that there had been a violation of Article 13 (the right to
an e¬ective remedy). (See Gearty, ˜Osman unravels™ (2002) 65 MLR 87.) The
impact of Barrett is considerable. It is not a one-o¬, but has been regularly fol-
lowed in subsequent case law. Its e¬ect is that claimants in negligence actions
against public authorities are now much less likely to lose at the preliminary,
striking-out stage, as the courts are much more reluctant to rule that public
authorities owe no duty of care. This does not necessarily mean that claimants
are winning vastly more cases than previously, of course, as the imposition of
a duty of care is not the same as “ and is but the ¬rst step towards “ a ¬nding
of liability. Breach of duty and causation of non-remote damage still need
to be proven on the facts. (For a full analysis of X and Barrett, see Craig and
Fairgreave, ˜Barrett, negligence and discretionary powers™ [1999] PL 626.)
A public authority may have the defence to an action in tort that the act done
was authorised by statute. For instance, an authority is not liable in tort for a
nuisance resulting from its performance, without negligence, of a statutory
duty. If, on the other hand, a nuisance is caused by the authority in exercising a
public power, it will ordinarily be liable if the power could have been exercised
without causing the nuisance. (See eg, Marcic v Thames Water Utilities Ltd
[2003] UKHL 66, [2004] 2 AC 42.)
Damages for a tort committed by a public authority or o¬cer are assessed on
ordinary principles, but exemplary damages may be awarded if the authority
was guilty of ˜oppressive, arbitrary or unconstitutional action™ in performing
public functions. (Rookes v Barnard [1964] AC 1129, 1225“6. See also Holden
v Chief Constable of Lancashire [1987] QB 380; Kuddus v Chief Constable of
Leicestershire [2001] UKHL 29, [2002] 2 AC 122.)
In addition to negligence, there is a separate tort of breach of statutory duty.
Even though ministers and other public authorities are under a large number
and range of statutory duties, it is rare for an action for breach of statutory duty
to succeed (apart from in some cases concerning industrial accidents). The
courts will generally not allow an action for breach of statutory duty to proceed
unless two conditions are met: ¬rst, that Parliament evinced an intention that
the statutory duty in question should be actionable in this way and secondly
that the duty was intended to confer a bene¬t only on a particular group and
not on the public at large (see eg, Phelps v Hillingdon London Borough Council
[2001] 2 AC 619).
A further tort “ misfeasance in public o¬ce “ provides a remedy for abuse of
power by a public o¬cer. The elements of the tort, as identi¬ed by the House
of Lords in Three Rivers District Council v Bank of England (No 3) [2003] 2 AC 1,
716 British Government and the Constitution

are that a public o¬cer caused injury or loss to the claimant by an unlawful
act, either (˜targeted malice™) with the intention of causing such injury or (˜untar-
geted malice™) knowing that the act was unlawful and that it would probably
injure the claimant, or being reckless, in deliberate disregard of a serious risk
that injury to the claimant would result from the conduct known to be unlawful.
(See also Watkins v Home O¬ce [2006] UKHL 17, [2006] 2 AC 395.)
On the considerable impact of European Community law on questions of
liability against the state and against public authorities, see chapter 5 (above,
pp 312“15).
See generally on this area of law D Fairgrieve, State Liability in Tort: A
Comparative Law Study (2003) and C Harlow, State Liability: Tort Law and
Beyond (2004).

7 Tribunals
In R v Secretary of State for the Home Department, ex p Saleem [2001] 1 WLR
443, 457“8, Hale LJ observed:

There are now a large number of tribunals operating in a large number of specialist fields.
Their subject matter is often just as important to the citizen as that determined in the ordi-
nary courts. Their determinations are no less binding than those of the ordinary courts: the
only difference is that tribunals have no direct powers of enforcement and, in the rare cases
where this is needed, their decisions are enforced in the ordinary courts. . . . In this day and
age a right of access to a tribunal or other adjudicative mechanism established by the state
is just as important and fundamental as a right of access to the ordinary courts.

Tribunals are properly to be regarded, said the Franks Committee, ˜as machin-
ery provided by Parliament for adjudication rather than as part of the machinery
of administration™ (Administrative Tribunals and Enquiries, Cmnd 218/1957, para
40). Three points are indicated by this statement: almost all tribunals are created
by statute; they are intended to be independent of the administration; their func-
tion is to adjudicate on matters in dispute. A few tribunals (eg employment
tribunals and rent assessment committees) are set up to determine disputes
between one citizen and another; the great majority adjudicate on disputes
between the individual and a public authority concerning the rights or obliga-
tions of the individual under a statutory scheme. In this chapter our business is
with tribunals of the latter kind.
In 2005“06 there were over seventy di¬erent administrative tribunals in
Britain (excluding some that were rarely convened or moribund) and in that
year they decided about 700,000 cases (see the Annual Report of the Council on
Tribunals, HC 1210 of 2005“06, Appendix J). Many of these were multiple
bodies sitting in di¬erent parts of the country (eg there were ¬fty-six Valuation
Tribunals in England in 2006). The two oldest tribunals are the General and
717 The courts: judicial review and liability

Special Commissioners of Income Tax, dating from the beginning of the nine-
teenth century; among the newest are the Information Tribunal and the
Competition Appeal Tribunal. The busiest tribunals are the uni¬ed Social
Security and Child Support Appeals Service Tribunals, which cleared 262,816
cases in 2005“06. Some tribunals, on the other hand, rarely sit.
Most tribunals “ and those of present concern to us “ are appellate bodies,
set up to hear appeals from decisions of ministers, o¬cials, regulatory bodies or
lower tribunals. We should notice, however, that some tribunals have a ˜¬rst-
instance™ jurisdiction to deal with applications for licences or other bene¬ts (eg
the Tra¬c Commissioners and the Civil Aviation Authority). Broadly speaking,
it is an appellate tribunal™s function to decide whether the administrative
authority came to the right conclusion on the facts, correctly applied the rele-
vant statutory provisions and also, in many cases, whether it properly exercised
a discretion entrusted to it.
Appeals may be taken to specialised tribunals on a great variety of matters.
As noted in the Leggatt Report (Tribunals for Users: One System, One Service
(2001), para 1.16), the subjects they deal with ˜cover the whole range of politi-
cal and social life, including social security bene¬ts, health, education, tax,
agriculture, criminal injuries compensation, immigration and asylum, rents,
and parking™. In many of these cases it might have been decided to entrust the
courts with jurisdiction, but as the Franks Committee remarked (para 38):

tribunals have certain characteristics which often give them advantages over the courts.
These are cheapness, accessibility, freedom from technicality, expedition and expert knowl-
edge of their particular subject. It is no doubt because of these advantages that Parliament,
once it has decided that certain decisions ought not to be made by normal executive or
departmental processes, often entrusts them to tribunals rather than to the ordinary courts.

This is not to say that the government will necessarily have chosen to set up a
tribunal as a more appropriate forum than a court; the choice may rather have
been, as Prosser notes ((1977) 4 British Journal of Law and Society 39, 44)
˜between appeal to tribunals and no appeal™. Alan Boyle remarks that the use of
tribunals has been ˜essentially selective™ and that it is ˜left to governments to
make the selection on whatever criteria™ they choose: ˜only certain areas of
governmental decision-making have been surrendered into the hands of
independent tribunals™ (in G Richardson and H Genn (eds), Administrative Law
and Government Action (1994), p 84).
Since tribunals deal with such varied subject matter it is not surprising
that they have di¬ered greatly in their constitution, membership and working
arrangements. A few tribunals are one-person bodies (eg parking adjudicators
and tra¬c commissioners) but the majority are composed of a chairperson,


. 132
( 155 .)