The impact of liability on public bodies: lessons from the literature
Alex Marsh
The Law Commission &
Social for Policy Studies
University of Bristol
alex.marsh@bristol.ac.uk
Paper to the Socio-Legal Studies Association Conference 2008, Manchester, 18th-20th
March
Abstract
The negligence liability of public bodies is a topic of recurring debate in public law.
Views on the impact of changing liability differ sharply. Will increased liability promote
administrative efficiency or encourage a retreat to defensive administration? Evidence
with which to adjudicate between the competing perspectives is slight. The courts have
therefore had relatively unfettered discretion to adopt differing positions on the issue at
different times. As a consequence, in English law the boundaries of the negligence
liability of public bodies have proved to be highly unstable over the last three and a half
decades.
This paper draws upon work for a Law Commission project entitled Remedies against public
bodies. It draws upon a range of literatures – socio-legal, law and economics, public
administration and policy analysis – to explore what existing theory and evidence allows
us to say about the likely impact of changing liability. The paper argues that while direct
evidence regarding impacts is limited the picture we can construct from more indirect
sources suggest that we should expect changing liability to have a differential impact in
differing contexts, and in many instances the impact can be expected to be modest.
Introduction
The negligence liability of public bodies in situations involving the exercise of
discretionary power is a topic of recurring, indeed almost continual, debate in public law.
The issue is particularly vexed when significant public policy considerations are
identified. Since Dorset Yacht1 the courts have regularly returned to the question of liability
and resolved it in different ways. As a consequence, in English law the boundaries of the
negligence liability of public bodies have proved to be highly unstable. Over this period
the central concern has changed from duty to breach. A variety of devices rooted in
‘policy’ concerns have been proposed with which to draw relevant distinctions and deny
liability.2 The instability and uncertainty of the law has attracted the attention of the Law
Commission, initially in a scoping paper focusing upon monetary remedies3 and more
recently in a project entitled Remedies against public bodies that broadens the terrain over
which reform is to be considered.
At the centre of the debate over the appropriate extent for the negligence liability of
public bodies is the question of the impact that changing and extending liability would
1
2
Home Office v Dorset Yacht Co Ltd [1970] AC1004, HL
Reviewed in detail by C Booth and D Squires, The Negligence Liability of Public Authorities
(2006).
3
Law Commission, Remedies against public bodies: a scoping report (2006)
1
have upon bureaucratic behaviour. Views on the impact of changing liability differ
sharply. Will increased liability promote administrative efficiency or encourage a retreat
into defensive administration? Evidence with which to adjudicate between the competing
perspectives is slight. The courts have therefore had relatively unfettered discretion to
adopt differing positions on the issue at different times.
This paper draws upon a range of literatures – socio-legal, law and economics, public
administration and policy analysis – to explore what existing theory and evidence allows
us to say about the likely impact of changing liability. The paper argues that while direct
evidence regarding impacts is limited the picture we can construct from more indirect
sources suggest that we should expect changing liability to have a differential impact in
differing contexts, and in many instances the impact can be expected to be modest.
Negligence liability and policy concerns
While the courts acknowledge that the principal obligation in dealing with a claim that
discretionary power has been exercised negligently is to provide a remedy for wrongs, set
against this has been an array of arguments described as “policy” concerns. These policy
concerns have been deployed to deny or circumscribe the state’s duty of care, and hence
liability. These arguments can be summarized as follows4:
1. The efficient use of public resources;
2. Whether liability will encourage public authority employees to act in an overly
defensive or otherwise undesirable manner;
3. Whether liability would disrupt delicate relationships and would be impossible to
apportion fairly given the multi-disciplinary nature of a public authority’s activities;
4. The risk that recognition of a duty of care will flood the court with vexatious and
complex litigation;
5. The availability of alternative remedies;
6. Whether imposition of duty of care would be inconsistent with the purpose behind a
statutory regime.
The majority of these arguments concern the impact that finding a duty of care would
have upon the behaviour of relevant parties. In respect of (1), (2), and (3) it is the
response of public bodies that is in question. In (4) it is the response of potential
claimants and their representatives that is the concern.
Policy arguments advanced by the courts weigh heavily on the side of denying a duty of
care. Latterly the courts have been more willing to acknowledge a duty of care, but have
then used policy concerns to set a demanding standard for demonstrating that a public
body has acted negligently.
A variety of “miscellaneous positive consequences of imposing a duty of care”5 have
been identified, but this tends to be less common. Most relevant for our purposes is the
argument that imposing a duty of care will result in a higher standard of performance by
public authorities.6
4
C Booth and D Squires, The Negligence Liability of Public Authorities (2006), para 4.03.
Above.
6
See Barrett v Enfield London Borough Council [2001] 2 AC 550, 568; Phelps v Hillingdon
London Borough Council [2001] 2 AC 619, 672.
5
2
So two broad perspectives on the impact of liability can be identified. First, there is a
concern that it will lead to undesirable ‘defensive administration’, or otherwise inhibit
effective performance by public authorities. Second, there is the view that placing a duty
of care upon a public authority will improve the standard of administrative performance.
We could view these two perspectives as differing regarding whether the imposition of
liability is seen as deterring the careless execution of a task to an appropriate degree, and
results in effective administration, or whether the consequence of imposing liability is
“over-deterrence” and a retreat into defensive practice. However, in a public law context
the concern is often not careless performance of a simple task with a clearly defined
objective, but rather activities in which a balance must be struck between conflicting
priorities:
Officials are commissioned to act in precisely those situations in
which precious but conflicting values – order, public morality,
individual liberty, care of the dependent and vulnerable,
personal security, and … the social virtues of autonomous
individual activity – must be delicately balanced.7
In such circumstances, if the imposition of liability leads a public body to defensive
practices, such as overly thorough assembly of evidence or documentation of process,
then its priorities have been distorted. Other equally valued features of the administrative
process, such as expeditious decision making, may suffer as public bodies seek to avoid
exposure to claims of negligence. This is a particular concern in fields such as children’s
services, where timely action is of considerable significance. There are risks attached to
both unnecessarily delayed and unnecessarily precipitous action.
Embedded in both sides of the debate over the impact of liability is the assumption that
changing liability will have a discernible impact upon administrative behaviour. A third
position is that changing the liability regime will have little or no impact upon the
practice of public bodies. A qualified version of this position is that liability is only likely
to play a significant role in shaping administrative behaviour under certain conditions.
Clearly, whether the imposition of liability promotes more effective administration,
encourages defensive practices, or has no discernible effect is at heart an empirical
question. In large part it is a question of calibrating the impact of liability on
administrative behaviour. It is, however, well-established that there is little or no
empirical evidence available to shed light on the question of impact. In one sense this is
understandable: precisely because there has been reluctance under English law to impose
liability upon public bodies the scope for examining the consequences of its imposition
are rather limited. We therefore turned to consider indirect evidence.
The nature of liability
It would be a plausible hypothesis that the impact of liability would depend upon the
detail of how and upon whom it falls. For example, direct liability upon front-line
workers is likely to have a more immediate impact upon bureaucratic behaviour than
increasing the vicarious liability of employers because in the latter case indirect
mechanisms embodying weaker incentives are required to transmit increasing liability at
7
P Shuck, Suing government (1983) at p64.
3
organisational or senior officer level to the reshaping of front-line behaviour. 8 9
A further hypothesis would be that the impact of liability depends upon the wrong one is
seeking to deter. This point can be brought out with reference to the four sources of
government illegality proposed by Shuck:
1. Comprehension-based illegality: a failure to understand what is expected may result in
non-compliance. The transmission of messages through an organisation can lead to
significant distortion by the time they reach those whose responsibility it is to
implement.
2. Capacity-based illegality: an official may not be able to do what is expected of him or her
because resources are inadequate and/or the task is practically impossible.
3. Motivation-based illegality: a range of motivations, including fear of criticism, concerns
for personal safety, conflicting professional values, indolence, and malevolence, can
militate against undertaking a mandated act.
4. Negligence-based illegality: a failure to act may be the result of simple neglect.10
While for some of types of illegality it is plausible to hypothesise that monetary liability
could act directly to improve administrative performance, for others monetary remedies
would act indirectly at best. The existence of a monetary penalty could in theory provide
public bodies with the incentive to deal with negligence-based aspects of
maladministration, such as delay or inattention, that could be addressed through
improvements in organisational processes. In contrast, for problems arising from
insufficiency of resources, for example, monetary liability can have no direct impact on
performance, unless having to pay damages from the same budget results in the initial
problem becoming more acute. For a monetary remedy to have an effect upon this type
of problem it would have to work back to the policy level through the reallocation of
resources in favour of the service in question. If the task front-line officials have been set
is practically impossible then making it subject to a monetary remedy may have no
positive consequence for its performance.11
This is not to claim that the availability of a monetary remedy should be expected to have
a differential deterrence effect upon different wrongs. Rather it is simply to recognise
that public administration can fall short of the ideal in a number of different ways: it
should not be assumed that they are all equally (un)responsive to a specific mode of
deterrence.12
8
Recognising that liability insurance can blunt the incentives in the former case.
It has be suggested, in the US context, that these mechanisms are extremely indirect: ‘As the
Supreme Court explained … most constitutional torts are committed by relatively low-level officials
subject to limited control by elected officials or the voters … the responsibility of elected officials for
most governmental torts that give rise to damages liability is highly attenuated’: L Rosenthal, ‘A theory
of governmental damages liability: torts, constitutional torts, and takings’ (2006) Journal of
Constitutional Law, vol 9, pp1-73.
10
P Shuck, Suing government (1983), pp4-13.
11
For a discussion of “the tendency for implementers to be left to try to cope with the dilemmas
built into the policy itself” see G Fimister and M Hill, “Delegating implementation problems: social
security, housing and community care in Britain”, in M Hill (ed) New agendas in the study of the policy
process (1993), 128
12
Shuck approaches the topic from the opposite direction by exploring the question of the
appropriate remedy for different types of governmental illegality. This perspective advances the
stronger claim that different types of deterrence require different types of remedy, and that monetary
remedies are only effective in achieving certain remedial goals: ‘Because the sources of official
misconduct are diverse and complex, …, the strategies for deterring it can be no less so’. P Shuck,
9
4
Contextualising the impact of liability on public bodies
The impact of increasing liability upon the behaviour of public bodies is a product of
several factors. If one takes an instrumental perspective then one can argue that changes
in liability must affect the behaviour of claimants, their representatives, and the judiciary
before it will impact upon public bodies.13
The stance taken by the judiciary
If the judiciary do not embrace the principle of increased state liability and show a
reluctance to determine that public bodies are in breach of a duty of care, or seek to
minimise the financial consequences of being found to be in breach, then this will
mitigate any statutory impulse to increase liability. It will correspondingly decrease the
potential impact of liability upon the behaviour of public bodies.14
Increasing claims
Policy concern (4) above is often referred to as ‘opening the floodgates’: the implication
is that a change in liability will elicit a substantial increase in litigation. Public bodies
would consequently have increased incentive to attend to the question of liability. The
possibility of the floodgates being opened by increased liability was identified by, among
others, Lord Browne-Wilkinson in X v Bedfordshire County Council15, but has also been
contested.16
Changing liability could lead to an increased volume of litigation as a result of at least
three distinct processes. First, a change in the liability regime might broaden the range of
grounds upon which claims could be brought. This could increase the volume of claims
against a public body, without necessarily increasing the volume of claims made with
respect to a particular ground. Second, a change in the liability regime may lead to an
increase in the propensity to claim with respect to a particular ground for complaint.
That is, a larger proportion of a given pool of potential claimants will lodge a claim under
the reformed system than would be the case under the current system. Third, changing
the liability regime may lead to an increase in vexatious and/or unmeritorious litigation.
Whether these fears are well-founded is again primarily an empirical question. There is
limited direct empirical evidence available upon which to ground estimates of the likely
impact of increased public liability. It is, however, well-known in the socio-legal literature
that there are many reasons other than the absence of a monetary remedy why potential
plaintiffs do not commence proceedings.17 Conversely, there was a significant rise in
medical negligence claims in the 1980s and 1990s even though the liability regime did not
change.18 Hence, the relationship between the nature of the liability regime and the
Suing government (1983), at p103.
13
From a non-instrumental perspective, this need not be the case.
14
The extent to which the judiciary can be seen as departing from the will of Parliament, or
arrive at judgements that apparently lack due deference to the doctrine of the separation of powers, is a
substantial and contested research topic in its own right. See, for example, M. Cohn ‘Judicial activism
in the House of Lords: A composite constitutional approach’ (2007) Public Law, pp95-115, for a recent
discussion and a multidimensional model.
15
[1995] AC 633, 750-751. See also eg. S. Bailey, ‘Public authority liability in negligence: the
continued search for coherence’ (2006) Legal Studies, vol 26, pp155-184 at p175.
16
For example, Stuart-Smith LJ in Capital & Counties Plc v Hampshire County Council [1997]
QB 1004, 1043-1044.
17
eg. H Genn, Paths to Justice: What People Do and Think about Going to Law (1999).
18
As noted by P Cane ‘Consequences in judicial reasoning’ in J. Horder (ed), Oxford essays in
5
propensity to litigate is by no means straightforward.19
Settlement
More subtly, a change in the liability regime may alter the way in which claimants engage
with the process of litigation. At present monetary damages are not typically available
under judicial review. Under the current regime a proportion of cases are settled.20 If,
however, the prospect of monetary damages were introduced then this may lead more
plaintiffs to decline to settle and to continue to court in the hope of receiving a higher
award.
One can analyse this question theoretically from a law and economics perspective by
considering the relative costs and benefits of going to court or settling the case before it
reaches that stage.21 If the claimant’s assessment of benefits of litigating, net of costs,
exceeds the settlement the defendant is willing to offer then it is likely that the parties will
be unable to settle and the case will go to trial. The defendant’s settlement offer will, in
turn, be shaped by their perception of the costs they are likely to incur if the case
proceeds to trial.
A failure to settle, under standard assumptions of rationality, can arise as a result of
information imperfections leading to inaccurate assessments of the expected return to
litigating.22 However, from a behavioural economics perspective, it has been
demonstrated experimentally that even where claimant and defendant possess identical
information about the facts of the case they can still frequently fail to settle. This is
ascribed in part to an over-optimism bias, whereby the parties systematically read the
evidence as supporting their own position. Shared information does not encourage
convergence of views.23
Attaching a monetary remedy to a public law action would tend to increase the claimant’s
incentive to seek a trial, other things being equal, but would at the same time increase the
amount of money a defendant is willing to offer in pre-trial settlement. The net effect of
these opposing forces cannot be predicted a priori. It will depend on the parties’ attitudes
to risk. It will also depend upon institutional factors such as the extent of the defendant’s
jurisprudence (2000).
19
An alternative route to gaining an appreciation of this issue is to consider evidence from
countries where state liability is more extensive. Markesinis and Fedtke have recently argued that
evidence from Germany indicates that fears that extending liability will open the floodgates could well
be unfounded. In Germany state liability is more extensive and, in general, the culture more litigious,
yet state resources devoted to processing and defending negligence cases remain relatively modest. See
B Markesinis and J Fedtke, ‘Authority or reason? The economic consequences of liability for breach of
statutory duty in a comparative perspective’ (2007) European Business Law Review, vol 18, pp5-76.
See alternatively B Markesinis and J Fedtke, ‘Damages for the negligence of statutory bodies: The
empirical and comparative dimension to an unending debate’ (2007) Public Law, pp299-330.
20
Sunkin notes that ‘by the mid-1990s only approximately one-sixth of all judicial review
claims in England and Wales reached a substantive hearing’. M Sunkin, ‘Conceptual issues in
researching the impact of judicial review on government bureaucracies’ in M Hertogh and S Halliday
(eds) Judicial review and bureaucratic impact: international and interdisciplinary perspectives (2004)
at p50. Not all claims that failed to reach a hearing should be taken to represent a settlement.
21
See, for example, R Cooter and D Rubinfeld, ‘Economic analysis of legal disputes and their
resolution’ (1989) Journal of Economic Literature, vol 27, pp.1067-1097.
22
G Priest and B Klein, ‘The selection of legal disputes for litigation’ (1984) Journal of Legal
Studies, vol 13, pp1-55.
23
G Loewenstein, S Issacharoff, C Camerer and L Babcock, ‘Self-serving assessments of
fairness and pre-trial bargaining’ (1993) Journal of Legal Studies, vol 22, pp135-159.
6
discretion in offering settlements and whether the costs of settlement and litigation fall in
the same place.
It is not possible to shed much light on this question by drawing on direct empirical
evidence. However, there is mixed evidence from other jurisdictions that moves in the
opposite direction – for example, the imposition of caps upon the damages available in
tort cases, including medical malpractice, in several US states – can have effects such as
increasing the probability of pre-trial settlement.24 The US literature also explores the
‘shadow effect’ of highly uncertain punitive damage awards in increasing the incentive for
pre-trial settlement: this debate is very similar in structure to our current concern.25
Liability in the private sector
The deterrence approach to tort law has its roots in the law and economics approach. 26
From this perspective the function of tort law is to provide economic agents with
incentives to take precautions against causing harm to others. Rational economic agents
will weigh up the costs and benefits of taking harm-reducing precautions and invest to
the point at which the marginal cost of taking precautions equals the marginal benefit.
The tort system forces economic agents to take into account costs and benefits that
would go unacknowledged in a system based upon the private costs of production.
Much of the discussion of state liability starts by taking this model as a reasonable
account of the way in which the tort system affects the private sector. It then proceeds to
examine whether it has similar relevance to the way in which public bodies would
respond to the imposition of liability. This model is, however, at best a partial
representation of how private bodies respond to tortious liability. In practice the picture
is much richer, with liability being an effective way to encourage private bodies to take
certain types of socially desirable action but not others.27 Changing liability regimes can
also trigger responses other than increasing the care taken by firms. One such response
would be to manage risk by restructuring production so that the activities most exposed
to litigation are performed by smaller producers who have less to lose should they be
sued.28
The simple model of private sector liability focuses upon providers’ expectations of
monetary liability. This assumes that the direct monetary outlay associated with litigation
24
Evidence on the varied experiences across US states is briefly reviewed in L Babcock and G
Pogarsky, ‘Damage caps and settlement: a behavioural approach’ (1999) Journal of Legal Studies, vol
28, pp.341-370.
25
T A Eaton, D B Mustard and S M Talarico, ‘The effects of seeking punitive damages on the
processing of tort claims’ (2005) Journal of Legal Studies, vol 34, pp.343-369.
26
eg. R. Posner, ‘A theory of negligence’ (1972) Journal of Legal Studies, vol 1. For an
overview of this approach see L Kaplow and S Shavell, ‘Economic analysis of law’ in A Auerbach and
M Feldstein (eds) Handbook of public economics (1999) or S Shavell, Economic analysis of accident
law (2003) NBER Working Paper 9694.
27
See, generally, D Dewees, D Duff and M Trebilock, Exploring the domain of accident law:
taking the facts seriously (1996). See also L Liang, F Sloan and E Stout, ‘Precaution, compensation and
threats of sanctions: the case of alcohol servers’ (2004) International Review of Law and Economics,
vol 24, pp49-70.
28
These issues are discussed, in relation to the impact of regimes for dealing with environmental
pollution based upon negligence and strict liability, by A Alberini and D Austin, ‘Accidents waiting to
happen: Liability policy and toxic pollution releases’ (2002) Review of Economics and Statistics, vol
84, no 4, pp729-741.
7
is the providers’ main concern. However, a broader conception of the impact of litigation
might also to acknowledge that bad publicity associated with being found liable in
negligence may have far-reaching consequences for reputation and future demand for
products. Producers may rationally take greater account of a liability regime than the size
and frequency of successful negligence actions might suggest. This is one explanation
that has been offered for what has been described, from the perspective of strictly
rational action, as the ‘puzzle’ of extensive voluntary compliance with regulation.
Liability in the public sector
The law and economics perspective and risk reduction
Both sides of the debate over the impact of liability can be interpreted using a law and
economics approach. Those who see liability as promoting effective administration are
implicitly assuming that liability will give public bodies the incentive to invest sufficiently
in the ‘harm-reducing’ precautions of good administrative practice up to the point at
which the costs of doing so outweigh the benefits. In contrast, those who fear defensive
administration as a consequence of increased liability see public bodies over-investing in
such harm-reducing precautions.
This line of argument highlights an underlying difficulty with the defensive
administration argument. Why should rational actors invest in harm-reducing precautions
to the point where the result is defensive administration? The law typically requires no
more than that a public body should achieve a reasonable standard of performance in
order it not to be held to be negligent, and in a public law context the criteria for
demonstrating that behaviour is unreasonable are demanding, such over-investment of
resources would not seem rational.
One possible explanation is that there is little preventing it from happening. Public
bodies do not face the countervailing pressures of product market competition. If
increased expenditure on defensive administration causes cuts in the level or quality of
service, customers cannot transfer their business to an alternative supplier. Hence riskaverse suppliers will devote more than the optimal level of resources to reducing their
exposure to negligence claims.
This argument could hold for acts of negligence or maladministration where greater
attention could effectively reduce exposure. It is less certain that it holds for types of
tortious liability for which the source of the problem lies elsewhere. Indeed, some public
bodies may be exposed to the risk of a claim for negligence from more than one of the
parties involved, and minimising the risk of exposure from one quarter may increase the
risk from another. This being the case, the absence of competition would not in itself
significantly reduce a public body’s incentive to consider the issue of liability.
Another possible reason for a retreat into defensive administration relates to
information. The basic economic argument assumes that the costs and benefits of the
various courses of action are known.29 It is therefore possible for economic agents to
calibrate their response to the risk of litigation.
If, in contrast, the standard of reasonableness or, more generally, the limits of liability are
29
At least probabilistically, if modelled in terms of expected values attached to the various
options and prospects.
8
highly uncertain then it is not possible to identify the appropriate level of investment in
harm-reduction measures: this could lead those who are risk-averse to invest heavily in
‘good’ administrative practice to maximise their chances of meeting whatever standard
will be applied, to the extent that de facto defensive administration is the result.30
The way in which English law on the negligence liability of public bodies has evolved
over the last two decades suggests that it would not be imprudent for public bodies to
assume that the boundaries of negligence are rather uncertain.
Attitudes to risk are also relevant to the more fundamental question: precisely which
practices constitute ‘defensive administration’? The term lacks any widely-accepted
definition. A practice that would be deemed ‘defensive’ by a public official with a high
tolerance for risk might be seen as entirely appropriate and prudent – and not ‘defensive’
in a negative sense - by a second who is risk averse.
In the field of medical negligence there is evidence from both the UK and elsewhere to
suggest that changing liability or increasing risk of litigation leads to changes in clinical
practice that could be deemed to be ‘defensive’.31 Yet, others argue that some of these
same practices are entirely appropriate and hence greater exposure to litigation has
improved practice.32 In this respect distinguishing risk-reducing from risk-avoiding
strategies can be valuable. If exposure to liability encouraged risk-reducing strategies then
that may well be socially desirable, whereas investment in risk-avoiding strategies which
do not directly benefit the recipient of services are largely redundant.
The responsiveness of public bodies to monetary remedies
As noted above, an underlying premise of much of the debate is that public bodies, by
analogy with private bodies, are responsive to monetary remedies. This has been
forcefully contested in the US literature on constitutional torts.
Levinson33 argues that the view that monetary remedies will have a significant impact
upon public bodies is misconceived. Public bodies are more responsive to the political
costs and benefits of administrative action than the financial costs, which can ultimately
be passed on to others. Hence, changing the behaviour of public bodies requires
regulatory mechanisms that are politically salient.
Moreover, Levinson argues that one’s understanding of the impact of liability will depend
upon which of the several available models of government one uses. Levinson considers
This point has been made in relation to ‘overkill’ arguments in judicial reasoning, of which
defensive administration is one example, by P Cane ‘Consequences in judicial reasoning’ in J. Horder
(ed), Oxford essays in jurisprudence (2000) at p45, although it is not cast in quite the same terms.
31
D Dewees, D Duff and M Trebilock, Exploring the domain of accident law: taking the facts
seriously (1996) at p417: ‘Empirical evidence suggests that the civil liability system has had a
significant impact on medical practice, such as increased record keeping, increased discussion with
patients about treatment risks, referrals to other health care professional, and increased diagnostic
testing’. See also J Elmore et al, ‘Does litigation influence medical malpractice?’, Radiology, vol 236,
iss 11, pp37-36.
32
See eg. A Symon, ‘Litigation and changes in professional behaviour: a qualitative appraisal’
(2000) Midwifery, vol 16, iss 1, pp.15-21. It is also debated whether defensive practice makes a
discernible impact upon the risk of medical injury, and hence represents an effective defence against
the perceived health risks. Given that it is not the risks to patient health that are being minimised but
the risk of litigation, this is perhaps not surprising.
33
D Levinson, ‘Making governments pay: Markets, politics and the allocation of constitutional
costs’ (2000) University of Chicago Law Review, vol 67, pp345-420.
30
9
that ‘without a sufficiently complex model, any predictions about the incentive effects of
constitutional cost remedies on government behaviour are highly suspect’.34
Rosenthal has observed that ‘after Levinson’s attack, the justification for any form of
governmental damages liability law is in shambles’.35 While Rosenthal accepts Levinson’s
argument that government responds primarily to political incentives his conclusion is
disputed. Rosenthal argues that governmental liability reduces the resources available to
politicians to pursue their favoured political agenda and that in turn affects their chances
of re-election. Hence there is an incentive to minimize exposure to tort claims.
Rosenthal deploys the further indirect argument that in the US the widespread
implementation of immunity legislation should be taken as an indicator that liability
represents a constraint upon the discretion of public bodies. By implication, the costs of
liability must be perceived by legislators to be greater than the political costs associated
with passing immunity legislation.36
While the detail of this argument is shaped to some degree by the political structures of
the US, the broad points that it raises are important. It is essential to recognise that
administrative decision-making is typically embedded in a political process, and that this
can inject other priorities into the decision-making calculus.
The opportunities to study the impact upon public bodies of specific changes in the
contours of the liability system in the UK are rare. Hartshorne and colleagues report on a
study of the impact of Capital and Counties plc v Hampshire County Council37 upon the Fire
Service.38 The research demonstrates that there are many variables that influence the
impact of liability. The authors note that:
In Hampshire … the discussion of whether the imposition of liability may lead to
defensive fire fighting rested on the premise that fire fighters would be aware
both of any legal decision imposing liability, and its full legal implications. The
research demonstrates that instead, this information is taking time to filter
through to all ranks, and in some instances is becoming distorted in the process.
Moreover, there is a suggestion that in some cases information supplied to lower
ranks may be, albeit for what are conceived as good reasons, to a certain extent
diluted. The research even demonstrates that as regards the highest echelons of
an organisation, the supply of legal information to it may be patchy, and the
understanding of it variable.39
Very similar variables have been identified in relation to the impact of judicial review,
discussed below.
D Levinson, ‘Making governments pay: Markets, politics and the allocation of constitutional
costs’ (2000) University of Chicago Law Review, vol 67, at pp386-87.
35
L Rosenthal, ‘A theory of governmental damages liability: torts, constitutional torts, and
takings’ (2006) Journal of Constitutional Law, vol 9, pp1-73, at p30.
36
L Rosenthal, ‘A theory of governmental damages liability: torts, constitutional torts, and
takings’ (2006) Journal of Constitutional Law, vol 9, pp1-73, at pp44-45.
37
[1997] QB 1004.
38
J Hartshorne, N Smith and R Everton, ‘Caparo under fire: a study of the effects upon the Fire
Service of liability in negligence’ (2000) Modern Law Review, vol 63, no 4, pp502-522.
39
J Hartshorne, N Smith and R Everton, ‘Caparo under fire: a study of the effects upon the Fire
Service of liability in negligence’ (2000) Modern Law Review, vol 63, no 4, pp502-522. at p518.
34
10
A small US study of the effect of a negligent release lawsuit had upon the practices of a
psychiatric hospital found clearer evidence of deterrence.40 While suits of this type are
claimed to carry benefits in terms not only of deterrence but also compensation,
education and improvement of subsequent practice, only deterrence – leading to more
cautious release – was identified. Some commonly theorised negative effects upon front
line workers – particularly stress and litigaphobia – were found to be present, alongside
lower morale.
A key point here is that over-caution on the part of decision-makers regarding the release
of patients may have reduced the scope for negligent release lawsuits, but it impinged
upon the freedom rights of non-released patients, which not only led to harm to such
patients but also opened the possibly of legal challenges from a different quarter. This is
a good illustration of the complexity associated with decision making in public sector
contexts.
Public bodies and decision making
The discussion so far has drawn primarily upon the law and economics model of the
rational decision maker to underpin arguments about the impact of liability. This is,
however, only one way that decision making can be conceptualised. In this section we
briefly examine four alternative strands in the academic literature on the decision-making
process.
All policy embodies assumptions about the behaviour of those affected by it.41 The same
can be said for institutional designs. Changing our understanding of decision making
processes can potentially change our appreciation of the likely impact of imposing
liability, and of the value of alternative regulatory mechanisms for delivering effective
public administration.
Behavioural law and economics
Embedded in the law and economics approach to the impact of liability is the model of
rationality that underpins mainstream economics. The core of this version of rationality
is typically taken to be the clear-eyed pursuit of self-interest. It is undoubtedly possible to
detect elements of this type of cost-benefit thinking in the response of public bodies to
the liability placed upon them.42
The plausibility of this perspective as a general approach has been challenged by
behavioural economists, drawing on psychological evidence. The behavioural economics
perspective argues that adopting alternative starting assumptions about rationality can
deliver a richer understanding of legal phenomena.43
N Poythress and S Brodsky, ‘In the wake of a negligent release lawsuit: An investigation of
professional consequences and institutional impact on a state psychiatric hospital’ (1992), Law and
Human Behavior, vol 16, no 2, pp.155-173.
41
As most prominently discussed by J Le Grand, Motivation, agency and public policy: Of
knights & knaves, pawns & queens (2003).
42
The way in which German local authorities respond to changing requirements regarding
highway maintenance would appear to be an example. Rather than incur the costs of additional
maintenance it is cheaper to pay out the occasionally claim for damages. See B Markesinis and J
Fedtke, ‘Damages for the negligence of statutory bodies: The empirical and comparative dimension to
an unending debate’ (2007) Public Law, pp299-330.
43
See eg. C Jolls, C Sunstein and R Thaler, ‘A behavioural approach to law and economics’
(1998) Stanford Law Review, vol 50, pp.1471-1550; L Babcock and G Pogarsky, ‘Damage caps and
40
11
The central propositions of behavioural law and economics are, first, that decision
makers often use rules of thumb (‘heuristics’) rather than full cost-benefit calculation
and, second, decision-making is subject to a range of biases. These biases include
egocentrism, over-confidence, over-optimism, confirmation bias, and the endowment
effect.44 Heuristics include the availability heuristic.45 46
Several, though not all, departures from rationality can be curbed by placing
accountability requirements upon decision-makers. This can force decision-makers to
approach their decision in a more structured and evidence-based way. In this respect,
judicial review possesses many of the characteristics seen as desirable to ‘debias’ decision
making when the bias arises from ‘cognitive loafing’ – the failure to expend sufficient
care and mental effort in deliberation.47
What are the practical implications of adopting this view of decision making? In terms of
the impact of liability, behavioural law and economics can be used to consider the way in
which the availability heuristic affects public bodies’ perceptions of the risks associated
with particular courses of action. We have already noted above the relevance of one
finding: the impact of over-optimism on the decision to settle a claim or go to trial.
Over-confidence bias can equally lead to misapprehension of the robustness of
procedures and success in judicial review.48 These are analytical tools that have yet to be
systematically applied in the field of administrative justice, at least in the UK.
Decision research
Behavioural law and economics draws upon some of the key insights from decision
theory, a sub-field of psychology. It does not, however, exhaust the contribution that
theories of decision making can make to our understanding of the impact of liability. It is
not possible to provide a full overview of the field here.
The content and significance of a decision is likely to affect the process by which it is
made and the decision criteria applied. Whether a decision entails irreversible
commitments can shape how it is made. Decision characteristics might also be expected
to influence the amount of information acquired in advance of the decision and the
number of people involved in the decision, although the latter does not appear to differ
systematically across decision types.49
settlement: a behavioural approach’ (1999) Journal of Legal Studies, vol 28, pp.341-370.
44
The endowment effect refers to a situation in which the price at which an individual is willing
to sell something that they own exceeds the price at which they would be willing to buy an identical
item from someone else.
45
The availability heuristic means, for example, that individuals rely upon easy to recall or
recent experience of an event as the basis for assessing its likelihood. The consequence is overestimation of the probability of easily recalled or recently experienced events.
46
This alternative model of decision-making has itself been challenged on the basis that it
overstates the prevalence of departures from rationality: G Mitchell, ‘Why law and economics’ perfect
rationality should not be traded for behavioral law and economics’ equal incompetence’ (2002)
Georgetown Law Journal, vol 19, pp. 67-167.
47
See, for example, M Seidenfeld, ‘Cognitive loafing, social conformity, and judicial review of
agency rulemaking’ (2002) Cornell Law Review, vol 87, pp487-548; See also C Jolls and C Sunstein,
‘Debiasing through law’ (2006) Journal of Legal Studies, vol 35, pp.199-241.
48
This would appear to have been the case in the local authorities studied by Halliday. See S
Halliday, ‘The influence of judicial review on bureaucratic decision-making’ (2000) Public Law,
pp.110-122.
49
B Bozeman and S K Pandey, ‘Public management decision making: effects of decision
12
A well-established alternative to rationality as understood by mainstream economics is
Herbert Simon’s concept of bounded rationality.50 From the bounded rationality
perspective decision makers have aspirations with respect to a decision outcome. They
will search for a solution that reaches their aspiration level and then stop. This process is
referred to as ‘satisficing’ rather than ‘maximising’. It has been suggested that similar
processes operate with respect to the acquisition of the information upon which to base
a decision.51
Key questions then become how aspiration levels are set and what constitutes ‘good
enough’ information; these parameters can shape the ‘quality’ of the resulting decision.
The type of accountability mechanisms that exist under public law could be seen as
providing guidance on what would constitute a ‘good enough’ (for example, nonnegligent) decision. While this is the case, the guidance offered is not entirely
unproblematic because some of the requirements it identifies are rather open textured
and uncertain.52
Decision quality can be affected by a range of other factors such as decision makers’
ability to cope with information flows, in particular performance in the face of
information overload; the use of structured methods and expert systems to support
decision making; and the interaction between decision outcomes and emotions. More
fundamentally, psychological research has identified a range of factors contributing to
decision avoidance, and it has been suggested that because most people experience
omission bias – they are more likely to experience or anticipate regret from acts of
commission than acts of omission – they can be an in-built tendency towards inertia.53
Group decision making
So far the discussion has been framed in terms of a single, unified ‘decision-maker’ who
is intendedly, if not perfectly, rational. But in many policy contexts this is not an accurate
reflection of how decisions are made. Many decisions that could be open to legal
challenge will be made by groups of people with diverse interests following processes
that are fundamentally political rather than demonstrably rational.
The issue of multi-disciplinary working has entered debates on remedies against public
bodies through concerns over the difficulties in attributing liability. There is, however,
potential to apply insights from social psychology to the rationality of such group
processes in order to further our understanding of the role that legal principles and legal
constraints will play in such deliberations.54
content’ (2004) Public Administration Review, vol 64, no 5, pp.553-565.
50
H A Simon, Models of bounded rationality vol 2: Behavioural economics and Business
Organization (1982).
51
See eg. J Berryman, “What defines ‘enough’ information? How policy workers make
judgements and decisions during information seeking: preliminary results from an exploratory study’
(2006) Information Research, vol 11, no 4 (http://informationr.net/ir/11-4/paper266.html#zsa; Last
visited 04 Sept 2007).
52
See K Pick and M Sunkin, ‘The changing impact of judicial review: the independent review
service of the Social Fund’ (2001) Public Law, at p. 754.
53
C J Anderson, ‘The Psychology of Doing Nothing: Forms of decision avoidance result from
reason and emotion’ (1998) Psychological Bulletin, vol 129, no 1, pp.139-167.
54
See eg. R Prentice, ‘Chicago man, K-T man, and the future of behavioural law and
economics’ (2003) Vanderbilt Law Review, vol 56, pp1664-1777.
13
Multi-agency working brings together representatives of organisations with different
professional values and experiences. Some may be employed in the voluntary or private
sector. It is not self-evident that all participants will place the requirements of robust
procedure, viewed from a public law perspective, at the centre of their concern. Whether
public law concerns feature prominently in deliberations will depend on the complex
intersection of priorities and power relationships that the organisations bring to a
decision. This can be seen, in part, as a more general version of the problem of
competing normative frameworks we discuss further below.
In terms of decision processes, group decision making can result in ‘groupthink’: the
tendency for groups to seek a quick consensus without adequate scrutiny of underlying
assumptions and alternatives.55 Group think is more likely where a group has a cohesive
culture, but this is not inevitable. A shared commitment to openness, critical reflection
and flexibility in thought can lead to more effective group functioning. In this sense,
group norms are important to understanding decision outcomes, but only some norms
promote more effective outcomes.56
Yet, much of the literature examines the role of conflict in group functioning. The
relationship between conflict and effective group decision making is complex. The
literature can be seen as embodying the idea of an efficient level of conflict: too little
conflict leads to homogeneity of thought and, potentially, groupthink. Too much conflict
prevents a group from functioning effectively. Somewhere between the two is a range of
conflict that enhances group functioning. Conflict means that alternative views are
presented and the case for particularly courses of action has to be made and defended,
hence increasing the chance of an effective course of action being identified. An
alternative approach is to differentiate between conflict that is task-oriented, which can
be positive, and conflict that is interpersonal, which is dysfunctional.57 Task-oriented
conflict can be the result of bring together people with different skills and/or
information. Whether this type of heterogeneity inevitably generates better decision
outcomes is debated.58
Socio-legal approaches
Halliday has drawn on research in local housing authorities to offer an analysis of the
nature of administrative decision-making framed in more sociological terms.59 This
account resonates with many of the insights derived from the approaches we have
already considered, but grounds the analysis in issues of more direct concern to this
project.
Halliday proposes that an important element in understanding the impact of judicial
review upon administrative decision making is the nature of the decision-maker. In
particular, he highlights their level of legal knowledge; their legal conscientiousness; and
55
Folowing I L Janis, Victims of groupthink (1972).
T Postmes, R Spears, and S Cihangir, ‘Quality of decision making and group norms’ (2001)
Journal of Personality and Social Psychology, vol 80, no 6, pp.918-930.
57
Eg. V D Wall, G J Galanes and S B Love, ‘Small, task-oriented groups: conflict, conflict
management, satisfaction and decision quality’ (1987) Small Group Research, vol 18, pp.31-55.
58
Eg. D J Devine, ‘ Effects of cognitive ability, task knowledge, information sharing, and
conflict decision-making effectiveness’ (1999) Small Group Research, vol 30, pp.608-634; S SchulzHardt, A Mojzisch, F C Brodbeck, R Kerschreiter and D Frey, ‘Group decision making in hidden
profile situations: Dissent as a facilitator for decision quality’ (2006) Journal of Personality and Social
Psychology, vol 91, no 6, pp.1080-1093.
59
S Halliday, Judicial review and compliance with administrative law (2004).
56
14
their legal competence.
Low levels of legal knowledge within an organisation will reduce the likelihood that
administrative practice will be in accord with the law, or that changes in the law will be
adequately reflected in changes in practice. If decision-makers are legally conscientious
then they will seek to comply with the law, whereas if they exhibit low levels of legal
conscientiousness then they will either comply creatively or fail to comply. Finally, legal
competence refers to the ability to appreciate the significance and implications of judicial
pronouncements. Some judgments may be framed in rather vague terms or it may be
necessary to extract from the facts of the case the underlying principle to be applied
more widely. In these cases a high degree of legal competence will allow the
administrative decision-maker to apply the judgment effectively, whereas a low level of
legal competence will lead to narrow application.60
Policy-oriented approaches
Underlying Halliday’s framework, as with much socio-legal work in this field, is the
concept of ‘street-level bureaucracy’, which has had an equally significant impact upon
the literature examining the nature of the policy process.61 Street-level bureaucrats are
those “who interact directly with citizens in the course of their jobs, and who have
substantial discretion in the execution of their work”.62 They work in spheres where
direct supervisory oversight is difficult, thus increasing the scope for the exercise of
discretion. Much of the literature has focused upon frontline police, social and welfare,
and educational services. Activities characteristic of street-level bureaucrats have been
identified in a range of empirical settings in a number of countries.
Street-level bureaucrats are seen as using their discretion to develop coping strategies to
manage the overload they face as a result of effectively unlimited demands but limited
resources. Strategies can take various forms, such as strict adherence to rules when the
exercise of discretion would be expected, in order to process cases quickly.63
If such strategies become embedded in the standard operating procedures of an
organization or engrained in the organization’s culture they can generate organizational
inertia. The incentives needed to deter inappropriate practice may need to be
correspondingly stronger.64 The difficulties in effecting permanent changes to
organisational practices should not be underestimated.65
60
From the survey responses received by Hartshorne et al it appears that in some areas of the
Fire Service legal competence was relatively low: in some areas the Hampshire judgment had been
interpreted as the need to revise policy with respect to sprinkler systems (the facts of the case) rather
than negligence more generally. See J Hartshorne, N Smith and R Everton, ‘Caparo under fire: a study
of the effects upon the Fire Service of liability in negligence’ (2000) Modern Law Review, vol 63, no 4,
pp.502-522.
61
M Lipsky, Street-level bureaucracy: Dilemmas of the individual in public service (1980).
62
M Lipsky, Street-level bureaucracy: Dilemmas of the individual in public service (1980), at
p3.
63
See I Loveland, ‘Housing benefit: administrative law and administrative practice’ (1988)
Public Administration, vol 66, no 1, pp57-75, for illustrative examples.
64
This point is also made, in a slightly different context, by B C Canon, ‘Studying bureaucratic
implementation of judicial policies in the United States: Conceptual and methodological approaches’ in
M Hertogh and S Halliday (eds) Judicial review and bureaucratic impact: international and
interdisciplinary perspectives (2004).
65
See eg. K M Sutcliffe and G McNamara, ‘Controlling decision-making practice in
organizations’ (2001) Organization Science, vol 12, no 4, pp.484-501.
15
Front-line workers can engage in triage in an attempt to manage their workload and
maintain a throughput of cases. A range of triaging principles has been identified in the
literature including: the complexity of the case; how deserving the bureaucrat considers
the client to be; and how interesting the case is to the bureaucrat. Some such coping
strategies are therefore precisely the type of procedural wrongs that administrative law is
designed to address.
The exercise of discretion does not always act to the disadvantage of service users.
Workers can bend the rules or subvert procedures as a means of helping clients they feel
deserve greater assistance than the rules would normally allow.66
Lipsky identified two characteristics that mean that discretion is integral to the work of
street-level bureaucrats. First, they may be dealing with complex situations and cases that
may present unique characteristics requiring a tailored response. Second, they operate in
contexts that are heavily bound by organisational rules or practices which can embody
conflicting or contradictory goals. It is left to the bureaucrat to draw from this menu of
rules the rule or combination of rules to apply in specific cases.67
While it may be possible to frame rules and procedures that reduce the scope for frontline workers to exercise their discretion,68 it is not possible to eradicate discretion entirely:
“like putty, discretion can be squeezed by oversight and rules but never eliminated; it will
shift and reemerge in some other form in some other place”. In this respect, street-level
work can be seen as: “rule saturated but not rule bound”.69 Even as the technology
associated with the administration of public services changes and computerised
management systems come to embody policy rules and procedural requirements, in a way
that would at first sight appear to limit discretion, it is possible for those on the front-line
to use their creativity to maintain scope for discretion.70
Conceiving of service delivery in terms of the activities of street-level bureaucrats opens
up the space for legal concerns and requirements to receive less weight than other
imperatives. Under what conditions will a system offer significant scope for the exercise
of front-line discretion and when will discretion be minimal? Equally importantly, under
what conditions will legal concerns receive less weight than other imperatives? These are
issues that we consider below.
Decision making in public bodies
In this section we have touched upon a broad spectrum of arguments. We are able to
See S Maynard-Moody and M Musheno, ‘State agent or citizen agent: Two narratives of
discretion’ (2000) Journal of Public Administration Research and Theory, vol 10, no 2, pp329-358.
See also M Blackmore, ‘Mind the gap: exploring the implementation deficit in the administration of the
stricter benefits regime’ (2001) Social Policy and Administration, vol 35, no 2, pp145-162.
67
Street-level bureaucrats also face the well-known hermeneutic problems associated with
seeking to relate the facts of the case in front of them to the rules which they are seeking to apply,
which are typically framed in more general terms. See J Black, Rules and regulators (1997).
68
The experiences of the teachers and employment agency workers who participated in the
small US qualitative study are compared in M Kelly, ‘Theories of justice and street-level discretion’
(1994) Journal of Public Administration Research and Theory, vol 4, no 2, pp119-140.
69
S Maynard-Moody and M Musheno, ‘State agent or citizen agent: Two narratives of
discretion’ (2000) Journal of Public Administration Research and Theory, vol 10, no 2, pp329-358 at
p339 and p334.
70
See eg. K Ellis, A Davis and K Rummery, Needs assessment, street-level bureaucracy and the
new community care (1999) Social Policy and Administration, vol 33, no 3, pp.262-280.
66
16
offer no more than a flavour of some of the debates and insights on offer. However, for
our purposes this is sufficient. Law and economics works with very simple normative
models of rationality, which in turn have inspired much of the thinking about the way in
which monetary remedies – and tort law in particular – will impact upon bureaucratic
behaviour. The aim of this section is to highlight the fact that once one moves beyond
these simple models it becomes clear that a wide range of factors, operating at the level
of the individual and the organisation, act to make the picture much more complex. In
additional, there are topics, such as the partial ways in which organisations gather
information from their environment or the unavoidable imperfections in intraorganisational communication, which we have not addressed but which would add a
further layer of complexity to the picture.
Much regarding the nature of decision-making is still to be debated. In some instances it
is competing accounts, rather than clear conclusions, that emerge. This in itself means
that simple assertions about the way decision makers will factor the risk of legal challenge
and monetary penalties into their deliberations should be treated with caution. The forces
weighing against giving priority to the requirements of public law can be substantial. We
return to this topic below.
Other regulatory mechanisms for achieving administrative justice
A range of internal and external mechanisms are available to those seeking to regulate the
activities of public bodies in order to achieve or ensure administrative justice. The
imposition of liability would be an external mechanism that would sit alongside the
existing judicial review mechanism, as well as a range of tribunals and ombudsmen. The
precise configuration of mechanisms will depend upon the policy sector. While there is
scope for conflict and competition between the various mechanisms and their
underpinning rationales, it can also be valuable to view them as potentially
complementary.71
There is a more substantial body of research examining the operation of judicial review
as a mechanism for regulating administrative decisions which can act as indirect evidence
in building our understanding of the likely impact of changing liability. We will focus
here primarily upon judicial review.
Impact of judicial review
While judicial review has the potential to promote components of administrative justice
such as the reasonableness of decisions and sufficient investigation of facts, it is not well
suited to fostering good administration more broadly:
A wider concept of good administration might, for instance, include values such
as effectiveness and efficiency in implementing public programmes and
delivering public services. The grounds of judicial review do not promote these
values as such and may, indeed, conflict with them. For instance, observing the
requirements of procedural fairness and respect for human rights may reduce the
ability of public functionaries to achieve their policy objectives quickly, cheaply
and comprehensively.72
See M Adler, ‘A socio-legal approach to administrative justice’, Law and Policy (2003), vol
25, vol 4, pp.323-352.
72
P Cane, Administrative Law (2004, 4th ed).
71
17
The notion of the impact of judicial review is complex73, and evidently it is
circumscribed. It is also the case that the mechanisms by which judicial review can
impacts upon administrative behaviour are several:74



The process of being subject to judicial review, including threats of litigation that
proceed no further, may lead to alteration in the behaviour of the public body
concerned;
The judgments emerging from judicial review can impact not only upon the
public body being reviewed but other bodies engaged in similar activities or using
similar organisational processes;
Judicial review can promote a particular set of legal norms, values and principles.
If the thinking of public bodies is influenced by these values then judicial review
may have a more indirect and diffuse impact upon administrative behaviour
through shaping the broad culture.
It can be argued that an analogue to each of these mechanisms exists in relation to
monetary remedies.
The first two mechanisms are broadly instrumental, working by means of deterrence, the
third mechanism could be thought of as a non-instrumental means of shaping behaviour:
‘juridical norms may be adopted for reasons that have little to do with the direct or
indirect threat of legal challenge.’75
The third indirect mechanism indicates the complexity of the task of assessing the full
range of potential impacts of judicial review. This is reinforced by reference to the
question of who is influenced by the judicial process and how. Sossin argues that the
principal means by which judicial requirements are communicated to front-line decision
makers is ‘soft law’ such as non-legislative guidance or internal rules and administrative
policies.76 The behaviour of front-line decision makers may therefore be strongly
influenced by judicial review but this influence may be heavily obscured.
Recognising that assessing the impact of judicial review is by no means straightforward,
do the existing UK empirical studies allow any conclusions to be drawn? Perhaps the
clearest conclusion is that judicial review appears to have limited ability to influence
administrative behaviour.
Early studies suggested a strong positive effect,77 but the accumulation of subsequent
evidence indicates that impact is limited in both administrative bureaucracies and
adjudicatory bodies.78 A study of the impact of judicial review on Mental Health Review
See G Richardson, ‘Impact studies in the United Kingdom’ in M Hertogh and S Halliday (eds)
Judicial review and bureaucratic impact: international and interdisciplinary perspectives (2004).
74
M Sunkin, ‘Conceptual issues in researching the impact of judicial review on government
bureaucracies’ in M Hertogh and S Halliday (eds) Judicial review and bureaucratic impact:
international and interdisciplinary perspectives (2004).
75
K Pick and M Sunkin, ‘The changing impact of judicial review: the independent review
service of the Social Fund’ (2001) Public Law, at p741.
76
L Sossin, ‘The politics of soft law: How judicial decisions influence bureaucratic discretion in
Canada’ in M Hertogh and S Halliday (eds) Judicial review and bureaucratic impact: international
and interdisciplinary perspectives (2004).
77
G Dalley and R Berthoud, Challenging discretion: the social fund review procedure (1992).
78
See eg. S Halliday, Judicial review and compliance with administrative law (2004); G.
Richardson and D Machin, ‘Judicial review and tribunal decision-making: A study of the Mental
73
18
Tribunals found it to be: ‘patchy at best, even with regard to procedural fairness’.79
In contrast, while it is important to acknowledge that the role of judicial review within
administrative justice systems differs between countries,80 there is empirical evidence to
suggest that elsewhere judicial review can rather more frequently have a discernible
impact upon bureaucratic outcomes, to the benefit of the claimant.81
A second conclusion is that the influence of judicial review can vary over time. Initial
exposure to review can have a significant impact upon an organisation, but repeated
exposure can lead to a less conscientious approach as the risk of judicial review becomes
seen as an integral element of the environment. Newly established organisations can be
more inclined to alter their practice in response to an adverse judicial review decision.
This inclination can lessen over time. Pick and Sunkin highlight the importance of senior
management in shaping the ‘reverence’ accorded to judicial review: a change of Social
Fund Commissioner led to a substantial change in emphasis at senior management level
and then throughout the organisation - from a focus upon legal process, and making
decisions that were ‘judicial review proof’, to a greater consumer focus.82
Third, there has been the suggestion that public bodies can effectively use the existence
of judicial review as an opportunity to delegate scrutiny of their own processes with
respect to their adherence to the requirements of administrative justice. Fourth, a level of
defensive practice - practices designed primarily to protect the organisation from legal
challenge - has been detected.83
One of the strongest messages to emerge from this literature is that the impact of judicial
review upon administrative decision-making needs to be set in the context of a range of
other influences. As Halliday argues, drawing on his case study of three local authority
homeless persons units:
How judicial review ‘impacted’ on the bureaucratic justice of the local authorities’
decision-making processes depended … upon the non-legal influences and
priorities which co-existed with concerns of legality. In different ways,
professional intuition, systemic suspicion, bureaucratic expediency, judgements
about the moral desert of applicants, inter-officer relations, financial constraints
Health Review Tribunal’ (2000) Public Law, pp494-; T. Buck ‘Judicial review and the Discretionary
Social Fund: The impact on a respondent organisation’ in T. Buck (ed) Judicial Review and Social
Welfare (1998); K Pick and M Sunkin, ‘The changing impact of judicial review: the independent
review service of the Social Fund’ (2001) Public Law, pp736-762; I Loveland, Housing Homeless
Persons (1995). This assessment can be contrasted with some of the views reported in A. Barker, ‘The
impact of judicial review: perspectives from Whitehall and the Courts’ (1996) Public Law, pp612-621.
79
G. Richardson and D Machin, ‘Judicial review and tribunal decision-making: A study of the
Mental Health Review Tribunal’ (2000) Public Law, pp494- at p514.
80
See P Kane, ‘Understanding judicial review and its impact’ in M Hertogh and S Halliday (eds)
Judicial review and bureaucratic impact: international and interdisciplinary perspectives (2004).
81
See R Creyke and J McMillan, ‘The operation of judicial review in Australia’ in M Hertogh
and S Halliday (eds) Judicial review and bureaucratic impact: international and interdisciplinary
perspectives (2004). For a US case study see M R Mandlawitz, ‘The impact of the legal system on
educational programming for young children with autistic spectrum disorder’ (2002) Journal of Autism
and Developmental Disorders, vol 32, pp495-508.
82
K Pick and M Sunkin, ‘The changing impact of judicial review: the independent review
service of the Social Fund’ (2001) Public Law, pp736-762.
83
On these points see S Halliday, Judicial review and compliance with administrative law
(2004).
19
and other values and pressures all played a part in how judicial review impacted
upon decision-making84
Impact of judicial rulings upon public bodies
Within political science, particularly in the US, there is an extensive literature seeking to
examine the impact of judicial rulings upon public bodies.85 One of the most systematic
studies of the impact of judicial activities upon a public body is O’Leary’s study of the
impact of judicial decisions in all cases involving the US Environmental Protection
Agency (EPA) over an extended study period.86 In this instance the research found that
court rulings tended to dictate priorities within the EPA, sometimes at the expense of
implementing statutory requirements, as the agency chose between competing priorities:
as court rulings were made work was reprogrammed and funds and personnel
redeployed.
Other impacts included greater difficulty in planning because discretion over
programmes of work was undermined. Some judgments handed down were vague or
broad ranging. The relative power of legal staff within the organisation increased, while
the power of scientific staff was eroded. Some departments within EPA perceived
positive effects such as resources being devoted to favoured projects or through
increasing staff motivation as they pulled together to meet the judicial agenda. O’Leary
concluded that the impact of judicial ruling upon EPA was neither entirely positive nor
negative.
Johnson examined five public bodies affected by five judicial rulings in Pennsylvania.87
He starts from the position that agencies react to judicial decisions in a series of related
stages that either use or risk resources. Te interpretation of the judgement and the
decision about how extensively to search for alternative responses are important in
shaping an agency’s response. Two agencies made only a limited interpretation of the
court rulings and, as a result, took the view that the ruling did not carry implications for
their subsequent practice. Johnson then found that when taking decisions regarding how
to respond agencies tended to weigh up perceived risks of enforcement against the costs
and possibilities of change. This suggests a relatively rational, in the law and economics
sense, approach to determining agency response. Generally the agencies examined sought
to minimise the extent of their response.
A similar stages-based model for understanding agency response to judicial rulings in the
US, which draws upon Johnson’s work,88 is presented by Canon.89 Canon argues that an
S Halliday, ‘The influence of judicial review on bureaucratic decision-making’ (2000) Public
Law, pp.110-122, at p116. See also M Sunkin and K Pick, ‘The changing impact of judicial review’
(2001) Public Law, p.736.
85
The US literature on judicial impact is reviewed by B C Canon, ‘Studying bureaucratic
implementation of judicial policies in the United States: Conceptual and methodological approaches’ in
M Hertogh and S Halliday (eds) Judicial review and bureaucratic impact: international and
interdisciplinary perspectives (2004), drawing in part upon B C Canon and CA Johnson, Judicial
policies: Implementation and impact (1999, 2nd ed).
86
R O’Leary, ‘The impact of federal court decisions on the policies and administration of the
US Environmental Protection Agency’ (1989) Administrative Law Review, vol 41, pp. 549-574.
87
C Johnson, ‘Judicial decisions and organizational change: some theoretical and empirical
notes on State court decisions and State administrative agencies’ (1979) Law & Society Review, vol 14,
no 1, pp.27-56.
88
See also C A Johnson, ‘Judicial decisions and organizational change’ (1979) Administration
and Society, vol 11, pp27-51.
89
B C Canon, ‘Studying bureaucratic implementation of judicial policies in the United States:
84
20
agency’s ‘behavioural adjustment’ needs to be understood in the light of an agency’s prior
‘acceptance decision’: does the agency welcome the judicial ruling as supporting its
mission; treat it as being of little importance; or take the view that the ruling will make it
more difficult for the agency to fulfil its goals, that the ruling conflicts with core agency
values, or that it is simply bad policy. Hence, the acceptance decision can be positive,
indifferent or negative.
An agency’s behavioural adjustment – the extent to which an agency changes its
subsequent behaviour in light of a judicial ruling – will not correlate directly with the
nature of the acceptance decision. For example, external pressures on the agency may
mean that even though its acceptance decision is negative it nonetheless perceives it
necessary to make a substantial behavioural adjustment. Yet, the behavioural adjustment
is mediated by the acceptance decision: an indifferent or negative acceptance decision
will tend to reduce the extent of an agencies response to judicial rulings, other things
being equal.
Importantly, Canon notes that the nature of agency responses to new legal requirements
need not simply be dichotomous: compliance or non-compliance. There is a continuum
of policy responses that represent a more or less compliant response to a legal ruling.
Overt non-compliance would represent direct challenge to the rule of law and, Canon
suggests, is likely to be relatively rare.
When is judicial review likely to be an important regulatory mechanism?
Halliday has offered the most comprehensive framework for contextualising the impact
of judicial review.90 We noted above that this framework identifies legal knowledge, legal
conscientiousness and legal competence as key characteristics of the decision-maker that
shape the impact of judicial review. In addition, Halliday argues that judicial review is
likely to be more effective in regulating administrative behaviour when, first, the law is
clear and consistent so there is little doubt regarding what it requires and, second, a
public body operates in a context where law has a strong presence or it does not face a
high level of competition from other normative frameworks.
What normative frameworks might be in competition with law? Halliday draws on the
framework proposed by Adler, who in turn is elaborating upon the work of Mashaw. 91
Adler identifies six normative models that can underpin administrative justice (see Table
1).
Alder argues that underlying contemporary public administration is a competition
between these different models. The first three models, drawn from Mashaw’s original
work, have, it is argued, been in competition as the dominant principle underpinning
administrative justice for an extended period. The latter three models – managerial,
consumerist and market – have emerged as key concerns for the public sector since the
reforms associated with the new public management agenda took hold in the 1980s. It is
possible to pursue each one independently, even though in practice policy development
Conceptual and methodological approaches’ in M Hertogh and S Halliday (eds) Judicial review and
bureaucratic impact: international and interdisciplinary perspectives (2004).
90
S Halliday, Judicial review and compliance with administrative law (2004).
91
M Adler, ‘A socio-legal approach to administrative justice’, Law & Policy (2003), vol 25, vol
4, pp.323-352. J Mashaw, Bureaucratic Justice: Managing social security disability claims (1983).
21
under the Thatcher-Major governments often pursued them simultaneously.
The 1980s saw a sustained challenge to the established bureaucratic and professional
powers within the public sector. The role of managers was seen as increasingly
significant. Managers need autonomy in order to act to deliver efficiency, which is judged
‘objectively’ through the use of performance measurement.
At the same time, the public sector was criticised for being insensitive to consumer
preferences: increasing emphasis was placed upon the need to give consumers a greater
say in the running of services. Indeed the idea of treating service users as ‘consumers’
represented a major reorientation in thinking. This fostered an increasingly
compensation-oriented culture.
Thirdly, greater emphasis has been placed upon the idea that public bodies can be viewed
as operating in a market environment. The relationship between an organisation
operating in a market and its consumers is at arms-length. Dissatisfied consumers exit to
another supplier. Poor performance is penalised through the ‘bottom line’. Serious
consumer dissatisfaction may lead to court action, analogous to suing for breach of
contract.
Not all of these six models are relevant to all public bodies, but in few fields will the legal
model exist unopposed.92 When other models come to dominate legal concerns may be
sacrificed in order to satisfy other imperatives. The contemporary emphasis upon
managerialism, for example, with the publication of targets and the naming and shaming
of those deemed to fail to reach an acceptable standard means that public officials will
need to be concerned with delivering satisfactory performance, even at greater risk of
legal challenge.93
The normative orientation of a public body will be the outcome of competition between
normative positions which, in turn, will be championed by different interests, with
different bargaining strength. It should be anticipated that:
These trade-offs vary between organizations and, within a given organization,
between the different policies delivered by that organization and between the
different stages of policy implementation. They also vary over time and between
countries.94
There are indications that the factors Halliday identifies as important in shaping the
influence of judicial review carry over to other types of regulatory mechanism. His recent
study of the impact of internal reviews on decision making within two Homeless Persons
Units found that internal reviews had greater influence in the circumstances predicted by
Richardson and Machin, for example, note the tensions between professional – in this case,
medical - and legal norms in relation to tribunal decision-making. See G. Richardson and D Machin,
‘Judicial review and tribunal decision-making: A study of the Mental Health Review Tribunal’ (2000)
Public Law, pp49493
The distorting effects of targets have recently been explored in the context of the health
service by G Bevand and C Hood, ‘What’s measured is what matters: Targets and gaming in the
English public health care system’ (2006) Public Administration, vol 84, no 3, pp.517-538. For a more
general discussion see C Propper and D Wilson, ‘The use and usefulness of performance measures in
the public sector’ (2003) Oxford Review of Economic Policy, vol 19, pp250-267.
94
M Adler, ‘A socio-legal approach to administrative justice’, Law & Policy (2003), vol 25, vol
4, pp.323-352 at p331.
92
22
his qualitative model. The research also noted that differences in the internal structure of
the organisations could make a difference to the extent to which decision-making was
informed by an up to date interpretation of the law: this was more apparent in the
organisation with a dedicated review officer than in the other organisation.95
Table 1: Adler’s six normative models of administrative justice
Model
Mode of decision- Legitimating goal
Mode
making
accountability
Bureaucratic
Applying rules
Accuracy
Hierarchical
Professional
Applying
knowledge
Expertise
Interpersonal
Legal
Asserting rights
Legality
Independent
Managerial
Managerial
autonomy
Consumer
participation
Efficiency gains
Performance
indicators
Consumer
charters
Consumerist
Consumer
satisfaction
of Characteristic
remedy
Administrative
review
Second opinion or
complaints
to
professional body
Appeal to a court
or tribunal (public
law)
Publicity
‘Voice’
and/or
compensation
through charters
Market
Matching supply Profit making
To owners and ‘Exit’
and/or
and demand
shareholders
private law court
action.
(Source: M Adler, ‘A socio-legal approach to administrative justice’ (2003) Law and
Policy).
The framework proposed by Halliday offers a powerful tool for thinking about the role
of law in public administration. If we unpack aspects of these models further, it is
possible to highlight additional dimensions to the issue. For example, the pursuit of
efficiency gains, under the managerial model, can lead to changes in the workforce’s
terms and conditions of employment. Employment conditions have been identified as a
factor in the quality of decision-making: the greater the insecurity and casualisation the
poorer the quality of decision-making.96
Adler’s framework usefully highlights the competing normative frameworks that can bear
upon public bodies. Yet, it treats each of the normative frameworks as relatively
unproblematic in itself. In contrast, the normative framework associated with
‘professionalism’, for example, can be contested. Different actors may have different
conceptions of what professionalism means. The framework requires the development of
a fuller account of the possibility of value conflicts within and between normative models
and the way these influence action, either at the level of policy or the individual decision.
The values of individual public officials, and how they relate to those of the organisation
or political system, are also significant in understanding how public bodies function on
the ground.97
D Cowan, S Halliday and C Hunter, ‘Adjudicating the implementation of homelessness law:
the promise of socio-legal studies’ (2006) Housing Studies, vol 21, no 3, pp. 381-400.
96
Q Alam, M Robinson and J Pacher, ‘Public sector reform: employment security and quality of
decision making in the Australian taxation office’ (2006) International Journal of Public
Administration, vol 29, pp743-758.
97
See for example M Kelly, ‘Theories of justice and street-level discretion’ (1994) Journal of
95
23
Many public bodies operate in an increasingly crowded normative landscape. Emphasis
upon managerialism, consumerism and the market, alongside traditional professional and
bureaucratic concerns, could lead to the view that bureaucratic and professional
discretion have largely been curtailed. This might suggest that as long as policies and
procedures are robust the scope for infringing administrative law has been reduced.
However, while front-line discretion may be more tightly constrained, to suggest that it
has been removed entirely would be mistaken.98
The relevance of indirect evidence
How relevant is evidence from alternative regulatory mechanisms, such as judicial review,
or other jurisdictions to understanding the likely impact of monetary remedies? We must
be conscious of the limits on the scope for learning. The impact of judicial review, for
example, may be moderated by the fact that in the UK the court does not substitute its
own decision for that of the public body, rather it instructs the public body to reconsider
its original decision. As was noted by Harlow in a seminal paper99, this opens up the
possibility of the public body making the same decision again but this time in accordance
with the requirements laid down by the court. Hence, a successful judicial review may
not materially affect the public body’s ability to achieve its desired administrative
outcome. In contrast, the award of damages will impinge upon budgets100 and will send a
more concrete message that there had been a wrong that needed remedying. In this
respect it may elicit a stronger response from public bodies. Even here, however, the
incentives can be attenuated as when, because the division of labour within an
organization, budgets are the responsibility of a separate department.101
This can be contrasted with judicial rulings either that substitute the court’s decision for
that of the public official, as in appeals, or present the public body with what is taken to
be a mandate to act in a way other than it would have chosen, as appears to have been
the experience of the Environmental Protection Agency discussed above. This would not
be the situation should the liability regime be changed and hence the observed response
is arguably stronger than that which might be expected from extending liability.
Public Administration Research and Theory, vol 4, no 2, pp119-140; D Thatcher and M Rein,
‘Managing value conflict in public policy’ (2004) Governance, vol 17, no 4, pp457-486.
98
This argument is developed further by T Evans and J Harris, ‘Street-level bureaucracy, social
work and the (exaggerated) death of discretion’ (2004) British Journal of Social Work, vol 34, no 6,
pp871-895.
99
C Harlow, ‘Administrative reaction to judicial review’ (1976), Public Law, pp116-.
100
Either directly or through increased premiums for liability insurance.
101
See S Halliday, Judicial review and compliance with administrative law (2004).
24
Conclusion
Evidence on the impact of changing liability upon public bodies is scarce. The evidence
that exists is somewhat indirect or fragmentary. When taken in the round it suggests, on
balance, that changing liability in itself would have limited impact upon administrative
behaviour. On the occasions where a change in behaviour is detected then judicial
intervention is more likely to encourage defensive administration than promote more
effective administration.
This conclusion, though, depends upon the precise structure of any reform. If the scope
of liability were to be extended substantially, the requirements to demonstrate fault
relaxed significantly, and the quantum of damages set high, then this is clearly more likely
to elicit a response from both claimants and public bodies than less radical reform.
In situations where public bodies have discretion regarding how to respond to changing
legal requirements - through internal reviews, judicial reviews, or changing liability, for
example - it is necessary to contextualise their behaviour and understand the competing
influences upon decision making. It is plausible to argue that in some circumstances and
for some organisations the range of variables acting against, or dominating, legal
concerns will mean that regulatory mechanisms rooted in law will have little purchase
upon behaviour.102 It is also evident that the prominence of legal concerns in
organisational life differs sharply, even among those bodies that in principle have the
same concerns. This underlines the importance of understanding the context in which
public bodies operate.103
It is possible to offer hypotheses regarding the circumstances in which legal concerns will
be relatively marginal – low levels of legal knowledge, legal competence and legal
conscientiousness, an environment in which other (policy) priorities dominate, subject to
a field of law that is relatively vague in its prescriptions and implications – but systematic
empirical evidence does not as yet allow us to state with confidence how these
circumstances are distributed within and between policy sectors.
102
The marginality of law as a regulatory mechanism is a similarly strong theme in the literature
on regulation eg. K Hawkins, Law as last resort (2002).
103
A recent empirical study that starts to shed light upon the way in which public law impinges
differently upon the life of public bodies is M Sunkin, K Calvo, L Platt and T Landman, ‘Mapping the
use of judicial review to challenge local authorities in England and Wales’, (2007) Public Law, pp545567.
25