JOURNAL OF LAW AND SOCIETY

VOLUME 33, NUMBER 3, SEPTEMBER 2006

ISSN: 0263-323X, pp. 364±87

Regulation and Social Solidarity

Tony Prosser*

Justifications for regulation are commonly based on the identification

of market failures. This is however inadequate to account for much
regulation, and sees regulation as inherently second best to market
allocations. This article argues that, although some regulation will be
based on market failure, other justifications can be found in the pro-
tection of rights and in the maintenance of social solidarity.
Theoretical support for this last rationale can be found in the work
of Durkheim and Duguit and the concept of public service. To accept
this rationale for regulation has important implications both for
regulatory policies and regulatory instruments.

INTRODUCTION

Regulation is pervasive in all modern economies, yet it has had an

increasingly bad press in recent years, for example, in much of the debate
around `cutting red tape', `lifting the regulatory burden', and `better
regulation', a debate which achieved a surprising degree of prominence
during the 2005 United Kingdom general election. This combination of
pervasiveness and apparently questionable legitimacy makes it particularly
important to analyse why we regulate: what are our underlying regulatory
rationales? I shall attempt to ask these questions in this article, which will
proceed through a three-stage argument. First, I shall suggest that the most
commonly-cited regulatory rationale, that of market failure, is inadequate
either to explain or to justify normatively the range of regulatory tasks
currently undertaken. Secondly, I shall suggest that one implication of the

* School of Law, The University of Bristol, Wills Memorial Building, Bristol

BS8 1RJ, England
[Link]@[Link]

I would like to thank Roger Brownsword, Harry McVea, Bronwen Morgan, Colin Scott,
and the anonymous referees of this journal for their extremely helpful comments on drafts
of this paper.

364
ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School. Published by Blackwell Publishing Ltd,
9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA
market failure approach is to see regulation as a second best to market
allocation, but that this also does not adequately explain or justify current
practice. My final step will be to suggest a further rationale for regulation,
that of social solidarity, drawing on the work of Emile Durkheim, hitherto
somewhat neglected in regulatory studies, which provides a more appro-
priate description and justification for much regulatory activity. I shall then
conclude by discussing some implications of my argument for the design of
regulatory institutions and for regulatory practice.
It is important to underline that I am not suggesting that social solidarity
provides the sole justification for regulation. In many cases market failure
will be a reason, or more accurately a prerequisite, for regulation, and
normative justification for regulation can be found in economic principle.
However, I shall argue with Hollis and Nell that this is not enough:
[p]olitical economy is the science of applying sound economic theories to men
in a social setting. It will have to combine a sound economic theory with a
sound social and political theory.1
It is in fact possible to make a threefold distinction between different rationales
for regulation; those based on economic principles, those based on individual
rights, and those based on social solidarity.2 The first is well covered in many
existing accounts of regulation.3 For the second, recent work has stressed the
regulatory role in recognizing human rights and human dignity, especially in
the context of the regulation of biotechnology.4 The third rationale, that of
social solidarity, is relatively neglected, but has found expression in continental
legal concepts of public service and in some aspects of the Community law
approach to services of general interest.5 The three rationales set out here are
similar to Roger Brownsword's `bioethical triangle' distinguishing between
utilitarianism, human rights, and dignitarianism in the regulation of bioethics.6
He points to the `essentially oppositional' implications of these approaches,
largely because of differing views of the role of informed consent. Approaches

1 M. Hollis and E. Nell, Rational Economic Man: A Philosophical Critique of Neo-

Classical Economics (1975) 265.
2 See, also, T. Prosser, The Limits of Competition Law: Markets and Public Services
(2005) ch. 2.
3 For overviews, see A. Ogus, Regulation: Legal Form and Economic Theory (1994);
R. Baldwin and M. Cave, Understanding Regulation: Theory, Strategy and Practice
(1999).
4 See R. Brownsword, `What the World Needs Now: Techno-Regulation, Human
Rights and Human Dignity' in Global Governance and the Quest for Justice, Vol. 4
Human Rights, ed. R. Brownsword (2004) 203; R. Brownsword, `Code, Control and
Choice: Why East is East and West is West' (2005) 25 Legal Studies 1.
5 See, for example, C. Barnard, `EU Citizenship and the Principle of Solidarity' in
Social Welfare and EU Law, eds. M. Dougan and E. Spaventa (2005) 157 and the
Commission's Green Paper on Services of General Interest COM(2003)270 final.
6 R. Brownsword, `Stem Cells and Cloning: Where the Regulatory Consensus Fails'
(2004±5) 39 New England Law Rev. 535.

365

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
to regulation based on economic principles and human rights may indeed result
in substantial conflicts. However, as in this article I am not concerned with
issues of rights-based consent but rather with social solidarity as linked to
social and economic rights; thus the conflict between the two latter approaches
is less pronounced here. Indeed, it may well be that social solidarity finds its
ultimate justification in arguments drawn from human rights.7 I shall return
later to the question of how the conflict between the different implications of
each rationale may be resolved, but first shall clear the ground by a critique of
the market failure approach to regulation.

MARKET FAILURE AND REGULATORY PLURALISM8

There are some differences between various accounts of what market failure
constitutes, but the core of the concept refers to `public goods' in the sense of
goods which are non-rivalrous and not-excludable, externalities, information
problems, and the exercise of market power through monopoly or abuse of
dominance.9 A further distinction is that sometimes market failure is used as
a rationale in itself for regulation; in others, more appropriately, it is used to
set a threshold which, if passed, justifies the adoption of regulation on
economic or social grounds.
In one sense, it may seem obvious that market failure cannot offer a
universal justification for regulation; the world is full of examples of state
intervention on social grounds. However, to deny the role of such inter-
vention is not the point of the market failure approach. Instead, it claims that
if there is no market failure, this in itself provides grounds for preferring a
market-based allocation. Regulation is always a second best. The language

7 See R. Dworkin, `Constitutionalism and Democracy' (1995) 3 European J. of

Philosophy 2; compare R. Brownsword, `Genetic Engineering, Free Trade, and
Human Rights: Global Standards and Local Ethics' in Genetic Engineering:
Challenges Posed by a New Technology to the World's Trading Systems, ed. T.
Cottier (2006, forthcoming).
8 For earlier discussion of the market failure approach and its limits see, for example,
H. McVea, `Financial Services Regulation Under the Financial Services Authority:
A Reassertion of the Market Failure Thesis?' (2005) 64 Cambridge Law J. 413,
especially at 414, 416; J. Black, `Critical Reflections on Regulation', LSE Centre for
the Analysis of Risk and Regulation Discussion Paper 4 (2002) especially at 7; C.
Hood, H. Rothstein, and R. Baldwin, The Government of Risk: Understanding Risk
Regulation Regimes (2001) especially ch. 5.
9 From many sources, examples include (from a standard economics text) R. Lipsey
and K. Chrystal, Principles of Economics (1999, 9th edn.) 288±380; Ogus, op. cit.,
n. 3, pp. 29±46; McVea, id., pp. 417±23, 436, and Commission of the European
Communities, State Aid Action Plan ± Less and Better Targeted State Aid: A
Roadmap for State Aid Reform 2005±2009 (2005) 7. The combination of academic
and official sources here is in itself symptomatic of the use of the market-failure
approach.

366

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
of market failure is an expression of an approach to policy making in which
economic reasoning has dominated, and has been linked to globalization:
[t]he central perception of Globalization is that civilization should be seen
through economics and economics alone. If you add disease prevention or
urbanization or preservation of identity (the characteristic of belonging
somewhere) to a commercially driven view of human existence, you merely
compound the confusion about how the world works.10
The market failure approach to regulation has met with a number of
criticisms, as we shall see, but it seems to have gained ground in policy-
making recently.11 This approach can be found in general statements about
regulatory purposes, both at European and at United Kingdom level. For
example, in the context of state aids, the Commission has recently empha-
sized that it will strengthen its economic approach to state-aid analysis as a
contribution to the relaunched Lisbon strategy, a key element in which is the
analysis of market failures such as externalities, imperfect information or
coordination problems.12
A striking United Kingdom example of a market-failure based approach is
in the perhaps surprising context of broadcasting, with the establishment of
Ofcom as a new unified communications broadcaster by the Communica-
tions Act 2003. The question of regulatory rationales is particularly
important here as the Act gives Ofcom a dual principal duty:
(i) to further the interests of citizens in relation to communications matters;
and
(ii) to further the interests of consumers in relevant markets, where appropriate
by promoting competition.13
This reflected serious controversy as to the extent to which a regulator, much
of whose work was concerned with telecommunications, would be able
successfully to regulate broadcasting where much more varied, qualitative
considerations relating to the quality of content and balance are in play.
Market failure became particularly prominent in the multi-stage review of
public service broadcasting conducted by Ofcom in the first two years of its
operation. Thus this had to be based on `a rigorous understanding of what the
market, left on its own, would deliver ± and of the nature and scale of the
market failures that policy may need to address'.14 The review then exa-
mines different types of market failure which may be potentially relevant;
these are the fact that programmes are `public goods'; the shortcomings of
advertiser-funded TV in satisfying the preferences of different types of

10 J. Ralston Saul, The Collapse of Globalism (2005) 35.

11 For examples of such criticisms, see P. Brown, `The Failure of Market Failures' (1992)
21 J. of Socio-Economics 1; Black, op. cit., n. 8, p. 7; McVea, op. cit., n. 8, p. 436.
12 EC Commission, op. cit., n. 9, paras. 21, 23.
13 Communications Act 2003 s. 2(1).
14 Ofcom, Ofcom Review of Public Service Broadcasting: Phase 1 ± Is Television
Special? (2004) para. 143.

367

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
viewers; a tendency towards monopoly/oligopoly; lack of consumer informa-
tion; the presence of externalities; and the provision of merit goods.15
So far we are on familiar ground, but the various types of market failure
which have provided justification for public service broadcasting are then
classified by Ofcom into two categories; those which help the market work
to deliver the sort of programmes consumers want, and those which provide
the programmes which as citizens we want to be widely available for as
many people as possible to watch.16 As a result of the increased choice
available through digitalization, the first justification will be weaker; thus
`[t]he rationale for a continued investment in PSB is that only with such an
intervention would TV serve UK citizens adequately'.17 This approach is
also adopted in the later stages of the review,18 summarized in the final
document as:
we are not convinced that, in a fully digital world . . . undesirable outcomes
will necessarily be a direct result of consumer market failures ± rather, we
believe that this implies a situation in which a better-functioning market (i.e.
one which provides consumers with what they demand) would not meet all of
our needs as citizens. For this reason, we continue to believe that the public
intervention to deliver social purposes will still be needed, even with a fully-
functioning broadcasting market. But we do not believe that continued
consumer market failures are likely to be a major rationale for public
intervention in a mature, fully digital world.19
This approach is clearly much more sophisticated than the more traditional
market-failure approaches in that it accepts a range of citizenship-based
reasons for intervention rather than justifying it on the basis of a limited
number of economic grounds such as externalities or monopoly power. Rather
than providing a rationale for intervention, the concept of market failure is
here setting a threshold which, if passed, justifies the adoption of regulation on
non-economic, citizenship grounds. If markets can meet our goals (which may
be drawn from economic or social rationales), they should do so. It is only if
they cannot, through market failure, that regulation is justified. This might
appear to offer a welcome broadening out of the market-failure approach to
recognized broader social and citizenship values. However, the Ofcom
appoach is less open to non-economic values than it might at first sight appear.
It assumes that regulation is always a second-best solution, and the idea that
there is an independent, cultural or social, justification for regulation as an

15 id., para. 144.

16 id., para. 145.
17 id., para. 156.
18 Ofcom, Ofcom Review of Public Service Television Broadcasting: Phase 2 ±
Meeting the Digital Challenge (2004) para. 4.3 and background paper, `A
Conceptual Review of Public Service Broadcasting' ch. 2.; Ofcom, Ofcom Review
of Public Service Television Broadcasting: Phase 3 ± Competition for Quality
(2005) paras. 1.8, 2.22±3, 2.25±34.
19 id. (2005), para. 2.32.

368

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
alternative to markets has no place here. This is particularly surprising in the
context of public service broadcasting, where there has been extensive
discussion and advocacy of non-economic based justifications for regulation
to secure a wide range of social goals.20 The advantage of the Ofcom
approach, especially for regulating in practice, is that it seems to offer an easy
means of prioritizing different techniques for achieving goals; first the market,
and only if this will not work appropriately, regulation. The implication,
though, is that, from the outset, we must be defensive about adopting a
regulatory solution; if not, why do we need to bother with market failure rather
than regulating directly on the basis of citizenship?
The difficulty is that the market-failure approach is both too narrow and
too wide to provide a proper understanding of regulation. It is too narrow in
that it suggests a single type of methodological approach in assessing a wide
range of regulatory activities, one which assumes that in principle market
solutions are always the first-best outcomes to decisions on the allocation of
goods and services. It is too wide because it assumes that other justifications
for regulation are essentially arbitrary, for example, the whim of politicians
or political majorities rather than themselves being based on defensible
social theory. As once critic has put it, the economists' `words ``merits'' and
``equity'' are names for black holes. You have no substantive account of
these . . .'.21 There is thus an assumption of a radical divide between market
allocations and social justice, the first being a matter for technocratic
regulators mimicking market outcomes, the second for politicians; wherever
possible, the first should be preferred.22
This radical separation between the economic and the political appears
questionable given the increasingly wide definitions of regulation now being
used (at least amongst legal and sociological writers), which are far removed
from traditional command-and-control regulation. Indeed one of the reasons
for the influence of the market-failure model is that it adopts as the central
model of regulation command and control limiting the natural working of
markets; as such, it is to be minimized where possible. Black has recently
proposed as a definition:
regulation is the sustained and focused attempt to alter the behaviour of others
according to defined standards or purposes with the intention of producing a
broadly identified outcome or outcomes, which may involve mechanisms of
standard-setting, information-gathering and behaviour-modification.23

20 For a proposal of such cultural and social justifications for public service
broadcasting, see G. Born and T. Prosser, `Culture and Consumerism: Citizenship,
Public Service Broadcasting and the BBC's Fair Trading Obligations' (2001) 54
Modern Law Rev. 657.
21 Brown, op. cit., n. 11, p. 19. See, also, Black, op. cit., n. 8, pp. 21±2, 27.
22 For a strong statement of such a divide, see G. Majone, `Regulatory Legitimacy' in
his Regulating Europe (1996) 284.
23 Black, op. cit., n. 8, p. 20 (emphasis retained). Compare Brownsword, op. cit.
(2005), n. 4, p. 4.

369

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
Such a wide definition is proposed partly to ensure that decentred regulation,
not applied by public authorities, is included, but it also has merits both in
recognizing the pervasiveness of regulation and in opening up the plurality
of different regulatory objectives. On these latter grounds it is the definition
of regulation adopted in this article.
Such plurality is very evident when one thinks beyond the economic
regulators of the utilities and competition law.24 To give a very striking
example, the Human Fertilisation and Embryology Authority is hardly a
surrogate for market forces (although one prominent law and economics
scholar has considered the possibilities of a market in orphans).25 Indeed, the
House of Lords has recently emphasized the breadth of its discretion in
deciding difficult ethical questions.26 Moving closer to more conventionally
economic territory, the duties of the Environment Agency do not suggest that
its role is limited to the correction of market failures: thus, its primary duty is
to discharge its functions so as to protect or enhance the environment in
order to contribute towards achieving sustainable development.27 Of course,
economic considerations are relevant to how these goals are achieved, for
example, through the controversial duty on the Agency to have regard to
costs and benefits in exercising its powers, and in the use of trading schemes
for carbon emissions and biodegradeable municipal waste.28 However, the
use of economic techniques as part of regulation is quite different from either
treating market failure as the overriding rationale for regulation or as a
necessary threshold for regulatory intervention: it is a means, not an end. In
the case of the Health and Safety Commission and Executive, legal duties are
mainly concerned with enforcement of the law, but in the Commission's
strategy the stress is very much on health and safety as `a cornerstone of a
civilised society'; it is accepted that an economic case exists for health and
safety, but the values which are emphasized in the document are clearly not
limited to those of correcting market failure.29 This confirms a conclusion in
an examination of nine areas of risk regulation; it found serious limitations to
the utility of the market failure account of regulation in them:

24 A case can be made, of course, that public utility regulation is also based on a
variety of different economic, political, and social rationales; see T. Prosser, Law
and the Regulators (1997).
25 For discussion, see D. Campbell and S. Picciotto, `Exploring the Interaction
Between Law and Economics: The Limits of Formalism' (1998) 18 Legal Studies
249, at 255±6.
26 R (on the application of Quintavalle) v. Human Fertilisation and Embryology
Authority [2005] UKHL 28, for example, per Lord Hoffman at para. 26.
27 Environment Act 1995, s. 4. See, also, Secretary of State for Environment, Food and
Rural Affairs, The Environment Agency's Objectives and Contributions to
Sustainable Development: Statutory Guidance (2002).
28 Environment Act 1995, s. 39; Waste and Emissions Trading Act 2003.
29 Health and Safety Commission, A Strategy for Workplace Health and Safety in
Great Britain to 2010 and Beyond (2004).

370

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
 [t]he limited predictive power of the [market failure] hypothesis may surprise
those who stress the inexorable forces of global capitalism playing on modern
liberal democracies.30
If we move to a narrower view of regulation in which it is equated with
the governance of markets, an example would be that of financial services
regulation. Here it might be assumed that a market-failure rationale would be
much stronger, given the financial nature of the products regulated, the
absence of any direct distributive implications of regulatory decisions, and
the lack of a clear non-economic public interest rationale of the sort which
very obvious in environmental and health and safety regulation. However,
even here, a sustained and convincing case against the market-failure
rationale has been made recently, preferring a theory of financial services
regulation rooted in citizenship.31 Similarly, in the case of the public utilities
it appears that social regulation based on citizenship grounds plays an
important role in both regulatory duties and regulatory practice.32

BETTER REGULATION AND REGULATION AS SECOND-BEST33

A further problem with the market-failure approach was hinted at in my

coverage of Ofcom's work above. This is that it may confuse means and
ends because of an inadequate conception of the variety of relevant ends for
regulatory action; as a result, it assumes that regulation is always second best
to markets, and this assumption has contributed to the challenging of
regulatory legitimacy referred to in my introduction. Such an assumption has
been apparent in some of the `better regulation' initiatives at both United
Kingdom and European level.34 An extreme version of this approach was the

30 Hood, Rothstein, and Baldwin, op. cit., n. 8, p. 89.

31 McVea, op. cit., n. 8.
32 T. Prosser, `Public Service Law: Privatization's Unexpected Offspring' (2000) 63
Law and Contemporary Problems 63.
33 For a recent summary, see R. Baldwin, `Is Better Regulation Smarter Regulation?'
[2005] Public Law 485
34 See, for example, `Chancellor Launches Better Regulation Action Plan', HM
Treasury Press Release 50/05, 24 May 2005, at: <[Link]
newsroom_and_speeches/press/2005/press_50_05.cfm> (consulted 11 April 2006);
Cabinet Office Better Regulation Executive, `A Bill for Better Regulation:
Consultation Document' (2005), at: <[Link]
documents/pdf/consultation_doc.pdf> (consulted 11 April 2006). Some of the
proposals are to be implemented by the Legislative and Regulatory Reform Bill,
before Parliament at the time of writing. For European developments, see C.
Radaelli, `Regulatory Impact Assessments ± a New European Governance?' in
Centre for the Study of Regulated Industries, Regulatory Review 2004/2005 (2005)
293, and for European and other international developments, see Baldwin, id., pp.
487±90. Key Community documents are Communication from the Commission,
Simplifying and Improving the Regulatory Environment COM(2002) 278 final, and
Communication from the Commission to the Council and the European Parliament,

371

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
Australian National Competition Policy requiring cost-benefit analysis of
regulatory strategies against the yardstick of maximizing market com-
petition.35 Under this policy:
[f]or example, when justifying the regulation of professional qualifications of
immigration advice agents, bureaucrats in the Department of Immigration
initially framed the issue as one of protecting vulnerable clients from being
given poor advice. At a later stage of the review, the shaping influence of
National Competition Policy led to a reframing of the issue as one of
correcting the market failure of information asymmetry between advisor and
client.36
The danger (not only in Australia) of such an approach is of seeing the
task of regulation as essentially a technical matter of making rules; these
rules are then seen as constraints on the freedom of business to compete in
open markets, and so to be minimized. Again we see the reduction of the
legitimate role for regulation to a technical matter of correcting market
failure. Yet the issues raised, for example, relating to employment condi-
tions, in fact revolve around deep conflicts of values and are not merely a
matter of technicality. These misunderstandings show a basic confusion
between the means of regulation by various administrative techniques
(including but by no means limited to rules) and the goals of regulation as
reflections of different social values.37
In the United Kingdom, some of the more influential reports on regulatory
reform have accepted this distinction, for example, those by Philip Hampton
on reducing administrative burdens and by the Better Regulation Taskforce on
reducing burdens and improving outcomes.38 The first distinguishes carefully
between policy and administrative costs: the former are matters for political
decisions, the latter are administrative overheads to be reduced.39 This shows
a welcome acknowledgement that better regulation should not be about
restricting the policy options available to regulators, but about streamlining
administrative costs (and, outside the artificial world of bureaucrats cari-
catured by public choice theory, it is hard to think of anyone who would be
against such streamlining). The Better Regulation Taskforce report similarly
distinguishes policy and administrative costs, and goes further in recom-
mending the setting of quantitative targets for the reduction of the latter.40 The

Better Regulation for Growth and Jobs in the European Union COM (2005) 97
final.
35 For detailed analysis, see B. Morgan, Social Citizenship in the Shadow of
Competition: The Bureaucratic Politics of Regulatory Justification (2003).
36 id., pp. 3±4.
37 For a radical criticism of such confusion, see F. Ackerman and L. Heinzerling,
Priceless: On Knowing the Price of Everything and the Value of Nothing (2004).
38 P. Hampton, Reducing Administrative Burdens: Effective Inspection and Enforce-
ment (2005); Better Regulation Taskforce, Regulation ± Less is More: Reducing
Burdens, Improving Outcomes (2005).
39 Hampton, id., paras. 1.6±1.7.
40 Better Regulation Taskforce, op. cit., n. 38, p. 23.

372

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
report also proposes a `one-in, one-out' approach to regulation, in which a
balance has to be struck between new regulatory requirements and the removal
or simplification of existing ones.41 Such a highly quantified approach is
likely to downplay the importance of policy reasons for regulation in favour of
administrative balancing and the meeting of targets. Again, it portrays
regulation as a technical process rather than as the outcome of competing
policy values: a matter of means rather than ends.
These questions have arisen most controversially and vividly in the
adoption of cost-benefit analysis before regulatory decisions are made; this
has been a central element in recent moves to better regulation through the
use of regulatory impact assessments, required in the United Kingdom since
1998 for all regulatory policy proposals impacting on business, charities or
voluntary bodies.42 In one, weak, sense the adoption of cost-benefit analysis
or regulatory impact assessments has an important role in augmenting the
range of material available for the decision-maker, and, if properly done, it
can broaden political participation through increasing the range of interests
required to be involved in decision-making: it is a form of proceduralization.
Thus, in the EU context this means that:
the [regulatory impact assessment] becomes a tool to identify and discuss
openly the major trade-offs in the formulation of EU policy. This is why the
integrated [regulatory impact assessment] is also about raising the right
questions and illustrating the major tradeoffs . . .43
It is not going too far to suggest that, in this sense, cost-benefit analysis is of
value where it complicates decision-making by increasing the range of
interests to which the decision-maker must have regard; it is a means of
opening-up policy making without attempting to reduce it to mechanistic
formulae. The United Kingdom arrangements for regulatory impact assess-
ments might appear to fit this approach as they merely constitute recom-
mendations to ministers; they are `intended to inform decision-making, not
to determine decisions or to substitute for political accountability.'44 They
also fit Sunstein's claim that:
cost-benefit analysis should not be taken as undemocratic but, on the contrary,
should be seen as a means of fortifying democratic goals, by ensuring that
government decisions are responsive to well-informed public judgments.45

41 id., pp. 32-40.

42 See Baldwin, op. cit., n. 33, pp. 490±511. There is no space here for me to adopt a
full analysis of the issues of principle surrounding the regulatory impact assessment
process, for which see, for example, C. Sunstein, Risk and Reason (2002), and for
cogent critiques of the process, see Hood, Rothstein, and Baldwin, op. cit., n. 8,
especially pp. 179±84; Ackerman and Heinzerling, op. cit., n. 37.
43 Radaelli, op. cit., n. 34, p. 298. See, also, the early versions of the Australian policy
at Morgan, op. cit., n. 35, pp. 58±9, and compare the later National Competition
Policy itself, at pp. 67±8.
44 Baldwin, op. cit., n. 8, p. 491.
45 Sunstein, op. cit., n. 42, p. 35, also pp. ix, 64±5, 111±12, 124, 169, 178, 292.

373

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
This leaves room for debate on ends of regulation, not simply different
means.
Cost-benefit analysis may be treated quite differently however; it may
instead be seen as a means of simplifying decision-making by reducing the
relevant variables to a quantifiable form which can be weighed directly
against each other to determine an outcome. The potential danger of
shoehorning essentially qualitative principles into quantitative form has been
noted in the context of financial services regulation, and in the Australian
experience where `the aggregative nature of cost-benefit analysis tends to be
insensitive to distributive goals . . .'.46 The same danger can be seen more
broadly in elements of the better regulation initiatives, in particular, the use
of regulatory impact assessments. Thus the Cabinet Office Regulatory
Impact Assessment Guidance stresses that `costs and benefits must be
quantified wherever possible' and must be shown as monetary benefits using
economic valuation techniques wherever possible. Qualitative evaluation is
very much a second best:
[r]ecord costs and benefits in qualitative terms only when the above are not
possible. Without quantification it will be less clear what the differences
between the options are and whether the benefits of an option justify the
costs.47
This is of very considerable importance as a regulatory impact assessment must
be completed for all policy changes, whether European or domestic, which
could affect the public or private sectors, charities, the voluntary sector or small
businesses. The danger is of course that qualitative benefits of regulation could
be crowded out by more quantifiable costs.48 Indeed, the National Audit Office
in an evaluation of regulatory impact assessments has noted that:
[t]he expected benefits of many regulations provide changes or outputs for
which no market exists, and this makes quantification difficult. Many costs
tend to be easier to estimate in money terms; for example, a business will
know how much certain activities cost, so will have an idea how much these
costs may change as a result of changes due to new regulations.49
As an academic study has put it:
[i]n comparing monetized benefits to monetized costs, and in making this
comparison the criterion for judging public policies protecting people and the
environment, the analysis also stacks the deck against such policies.50

46 McVea, op. cit., n. 8, pp. 414, 423±6, 429±30, 444; Morgan, op. cit., n. 35, p. 72.
47 Cabinet Office, Regulatory Impact Assessment Guidance (2005) s. 5, at: <http://
[Link]/regulation/ria/ria_guidance/[Link]> (consulted 11
April 2006).
48 For a sustained critique of such crowding-out, see Ackerman and Heinzerling, op.
cit., n. 37.
49 National Audit Office, Evaluation of Regulatory Impact Assessments Compendium
Report 2003±4, HC (2003±4) 358, para. 2.52.
50 Ackerman and Heinzerling, op. cit., n. 37, p. 40.

374

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
 The problem of seeking to overquantify is particularly strong given the
limits of the regulatory impact assessment procedure; these have been
summarized as discouraging innovative alternative methods of regulation,
and as employing an approach too focused on single, given, regulatory
proposals rather than combinations of approaches or regulatory processes.51
This is, of course, not to deny that better regulation processes may be
extremely useful in tailoring regulatory interventions to their environment,
and indeed improvements may be made to the current process with that in
mind.52 The problem is, however, that the current processes do not see
regulation as an organic process that requires a balancing of competing
values setting out the sort of society we wish to live in, but rather as a set of
individual interventions that impose technical limitations on the functioning
of markets and are thus always a second best. Rules limiting the operation of
markets are justifiable if they are necessary to correct market failure, but are
not perceived as representing broader social values quite different from those
expressed through markets themselves.

REGULATION AND SOCIAL THEORY

At this stage it will be helpful to move back from the practical reality of
regulation as it has developed over the last few years to ask what we can
learn about different approaches to regulation from social theory. This is a
particularly pressing need given the expansion of the concept of regulation
discussed above; no longer can it be seen as the state issuing commands to
correct market failure, but is instead an all-pervasive social phenomenon. As
such, it cannot be isolated from broader developments in social theory. To
some extent this link has been made already, for example, in autopoiesis and
governmentality approaches. However, relatively little has been said about
the roots of different accounts of regulation in classical social theory.
The market-failure approach is, of course, rooted in welfare economics; as
Black has put it, `[a]lthough it is quintessentially an interdisciplinary subject,
those lawyers, political analysts, and sociologists who look at regulation
seem often to have surrendered the value debate to welfare economists'.53
Sociologically, the roots of this view lie in Weber and the role of formal
rather than substantive rationality.54 Points in common include the
methodological individualism of the welfare economics and Weberian
approaches, and the use of ideal types.55 A further striking point of inter-

51 Baldwin, op. cit., n. 33, p. 503.

52 id., pp. 507±11.
53 J. Black, `Proceduralising Regulation ± Part I' (2000) 20 Oxford J. of Legal Studies
597±614 and `Part II' (2001) 21 Oxford J. of Legal Studies 33±58, at 598.
54 For an important analysis of Weber's work in this area, see R. Swedberg, Max
Weber and the Idea of Economic Sociology (1998).
55 id., p. 28.

375

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
section between Weber's work and the more traditional approaches to
regulation is the stress on calculability of the legal order as a requirement for
modern or rational capitalism:
[t]he concept of `calculability' covers a host of phenomena in this context. For
one thing, because rational capitalism is sensitive to disturbances of any sort, it
is imperative that surprises be kept to a minimum. The state must not make
arbitrary interventions in the economy through its legal system, but respect its
autonomy. Property must be respected and contracts must not be broken (pacta
sunt servanda). The law must also be clear and logical, and it should be
administered in a professional and predictable manner. Calculability further-
more means that the legal system ± in contrast to religious law ± is predictable,
when it comes to matters that are not covered by the law.56
This theme is similar to much of the economics-influenced writing on
regulation, proposing that regulatory performance will only be satisfactory:
so long as three complementary mechanisms are in place to restrain arbitrary
administrative action: substantive restraints on discretionary actions by the
regulator, formal or informal restraints on changing the regulatory system, and
institutions to enforce the restraints.57
Ironically, some basis for the market-failure approach to regulation can
also be found in the Marxist tradition where regulation is seen as linked to
the developing process of capital accumulation, and, in effect, acting as its
handmaiden. As Bob Jessop has put it:
[t]he regulation approach analyses capitalism very broadly . . . critically
examining its anatomy as an `integral economy' or `economy in its inclusive
sense'. Seen in integral or inclusive terms, specific forms of capitalism can be
interpreted as an `accumulation regime + mode of social regulation'. This
comprises an ensemble of socially embedded, socially regularized and
strategically selective institutions, organizations, social forces and actions
organized around (or at least involved in) the expanded reproduction of capital
as a social relation.58
Alongside these approaches which root regulation in economic require-
ments is a much broader tradition of thinking of regulation in social rather
than economic terms. This is true even in the case of writers who come from
an economic tradition or are heavily influenced by welfare economics. To
give some examples from the most authoritative sources, Anthony Ogus,
whilst associating social regulation with market failures, also accepts that it
has a range of non-economic goals, including distributional justice, paternal-
ism, and community values.59 Baldwin and Cave also accept as rationales for

56 id., p. 104. For further discussion of calculability of law, see A.T. Kronman, Max
Weber (1983) 94±5, 122±4, 132±4, 135±7, 171.
57 B. Levy and P. Spiller, `A Framework for Resolving the Regulatory Problem' in
Regulations, Institutions and Commitment, eds. B. Levy, T. Eggertsonn, P. Spiller,
and R. Calvert (1996) 1, at 1.
58 B. Jessop, The Future of the Capitalist State (2002) 5; see, also, 225±36.
59 Ogus, op. cit., n. 3, pp. 4, 46±56.

376

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
regulation distributional justice and social policy and planning.60 Breyer,
whilst again starting from market failure, refers to fairness and income
distribution as potential regulatory rationales.61 However, these accounts
lack developed social theory comparable in sophistication to the economic
arguments for regulation. In Black's words:
[i]n the standard treatments of `regulation', the `why regulate?' question is
nearly always answered in terms of the correction of market failures, with the
occasional nod to distributional or other ancillary aims.62
However, reference to these goals is not sufficient to provide a coherent
theory alongside that of market failure; the different goals tend to be seen as
the result of arbitrary political decisions rather than reflecting any general
philosophy of what is required for a good society. This also is a source for
the weakness of these other goals in the political debates about what justifies
regulation.
How might these other goals be justified? Other accounts come from very
different sets of traditions, reflecting both proceduralist approaches to
regulation and the civic republican tradition. Thus Black, in a very important
contribution to regulatory theory, has distinguished between `thin' and
`thick' forms of proceduralism. The former is characterized by the exogen-
ous formation of preferences which are then aggregated through an electoral
system, and the determination of negative rights protecting citizens from
government and defining private spaces of action: `a ``thin'' conception of
proceduralization would thus involve participation in which preferences
remained exogenous, unchanged, and which was discourse-less.'63 By
contrast, `thick' proceduralization is based on deliberative democracy and
concepts of rational discourse derived from Habermas. Thus, in addition to
pragmatic discourse concerned with meeting goals of efficiency, there is a
place for ethical and moral discourse concerned with questions like `what
kind of society do we want to be' and `what is good for us all'.64 It is through
this that regulation can address the broader, substantive concerns which the
welfare economics approach has sidelined in regulatory studies.65
A not dissimilar result can be found in attempts to develop the `public
interest' approach to regulation away from being simply a `black hole' into
which any political preference can be dropped into something more
concerned with substantive values.66 This proposes a core meaning for the
`public interest' in the values of equality of citizenship within democracy,

60 Baldwin and Cave, op. cit., n. 3, pp. 14±15.

61 S. Breyer, Regulation and its Reform (1982) 15±35, especially at 20, 22.
62 Black, op. cit., n. 8, p. 7. See, also, J. Braithwaite, `The New Regulatory State and
the Transformation of Criminology' (2000) 40 Brit. J. of Criminology 222.
63 Black, op. cit., n. 53, p. 607.
64 id., p. 42.
65 id., p. 598.
66 See, in particular, M. Feintuck, `The Public Interest' in Regulation (2004).

377

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
strongly derived from the civic republican tradition.67 In an earlier
contribution making similar points, Sunstein has noted that:
[a]pproaches that begin from laissez-faire premises and rely exclusively on
neoclassical economics are bound to misinterpret the modern regulatory state,
relying as they do on criteria that cannot capture the diverse legitimate reasons
for regulatory controls.68
Instead he proposes `a series of interpretative principles, all with support in
current law, that can help accomplish some of the goals of deliberative
government . . .'.69 The principles are constitutional norms (including
federalism, political deliberation, accountability, the rule of law, and
property and contract rights); institutional concerns (such as the presumption
in favour of judicial review and deference to administrative discretion); and
efforts to correct statutory failure (for example, construing statutes so that
politically unaccountable actors are prohibited from deciding important
issues, collective action problems do not subvert statutory programmes, and
irrationality and injustice, measured against the statute's own purposes, are
avoided).70 This is probably the most fully-developed suggestion we have
for a set of substantive regulatory principles, though of course it is much
easier to ground such principles in a society with the constitutional base of
the United States of America than would be the case in the United Kingdom.
In addition, although they are based on substantive values, many of these
principles remain essentially procedural, for example, those relating to
judicial review. If we are to find a more thoroughgoing set of substantive
regulatory principles, they need supplementing from other sources.

SOCIAL SOLIDARITY

In finding such principles, it will be helpful to look outside the `Anglo-Saxon'

world towards continental Europe. Here there is a well-developed and
influential tradition of regulation as a means of implementing the values of
public service and as providing a contribution to social solidarity. This should
provide us with a useful means of supplementing a market-failure approach
by suggesting an independent source of regulatory legitimacy, and one which
does not necessarily make regulation a second best to market allocation. The
public law implications in relation to public service have been discussed in
detail elsewhere and so will not be covered in any depth here.71 Instead, the

67 id., p. 248.
68 C. Sunstein, After the Rights Revolution: Reconceiving the Regulatory State (1993)
at 229.
69 id., p. 231.
70 id., ch. 5.
71 See C. Graham and T. Prosser, Privatizing Public Enterprises ± Constitutions, the
State and Regulation in Comparative Perspective (1991); Prosser, op. cit., n. 2, pp.
35±7 and ch. 5.

378

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
major concern will be the theoretical sources of this approach, which can be
found in Durkheim, as further applied by the French legal theorist and
disciple of Durkheim, LeÂon Duguit. Some of the implications of this work for
regulation will then be discussed in more detail.
Roger Cotterrell has summarized the relevant strands in Durkheim's
writing as:
[w]hile subscribing to a view of sociology as the scientific and, in some sense,
value neutral, study of social facts, Durkheim saw it also as a form of
enlightenment. Moral concerns could never be far from the centre of a science
whose primary object was the study of society as a moral phenomenon. . . . for
Durkheim and some writers in a renewed Durkheimian tradition, a pressing
issue is how to symbolize social unity and create for modern complex societies
a moral framework in which regulation is effective, and the regulated are able,
in some way, to participate as moral actors in a solidary society which is more
than an economic free for all.72
This emphasizes the strongly moral elements in Durkheim's work (despite
the partial positivism of his methodology) and its contrast with the more
economics-based models derived from other social theorists: he strongly
rejected utilitarianism.73 The approach sounds primarily proceduralist
through emphasizing the participation of the regulated; however, there are
also strong substantive themes built around the concept of social solidarity.
This forms a major theme throughout Durkheim's writing, and will merely
be exemplified below.74
Within Durkheim's work, two particular aspects of social solidarity were
especially relevant. The first, and the most celebrated in discussions of his
work on law, was his use of law as an external index symbolizing the nature
of social solidarity.75 Social solidarity is thus a social phenomenon to be
explained and law provides part of the explanation. In one context, this has
been most discussed in relation to criminal law, in another, in relation to
contract. Thus he contrasts pre-modern societies characterized by `mech-
anical solidarity' and penal law, with the `organic solidarity' of modern
societies based on the division of labour and `restitutive law', of which the
central example is contract.76 This would seem at first sight to be similar to

72 R. Cotterrell, Law's Community: Legal Theory in Sociological Perspective (1995)

193±4.
73 For Durkheim's moral concerns see R. Cotterrell, Emile Durkheim: Law in a Moral
Domain (1999) chs. 2, 4, 7, 13, and for his rejection of utilitarianism, pp. 18±21, 29.
74 For an authoritative overview, see S. Lukes, Emile Durkheim: His Life and Work
(1992, rev. edn.). Ch. 7 of this book is an excellent summary of Durkheim's work on
social solidarity. For more specific discussion in relation to law, see Cotterrel, id.,
and S. Lukes and A. Scull (eds.), Durkheim and the Law (1983), especially the
editors' introduction and ch. 1. For a radical, non-positivistic interpretation of
Durkheim see, also, F. Pearce, The Radical Durkheim (1989) especially chs. 5 and 9.
75 See Lukes and Scull, id., especially editors' intro. and ch. 1; Cotterrell, op. cit., n.
73, pp. 31±4.
76 See H. Collins, Regulating Contracts (1999) 104.

379

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
the Weberian approaches discussed earlier through emphasizing the role of
predictable private law, but an essentially moral concept of social solidarity
(and later a peculiarly moral version of individualism) was also proposed in
relation to civil and administrative law. Indeed, in the context of the earlier
Rules of Sociological Method, `[p]olitical divisions, he there argued, are
moral, not merely material and geographical, and social organization can
only be studied via public law, which determines it . . .'.77
The second important aspect of social solidarity in Durkheim's work is its
role not simply as a matter to be explained, but as a moral phenomenon
limiting the potentially fragmenting effects of contractual relations. One
striking statement of this, particularly relevant to modern regulatory con-
cerns, is contained in ch. VII of The Division of Labour in Society, entitled
`Organic Solidarity and Contractual Solidarity'.78 Its importance is in
adopting a radically different view of social organization from one based on
private law contracts and exchange. The analysis begins with a critique of
Herbert Spencer's model of a spontaneous order in an atomized society with
cooperation automatically produced by the fact that each person follows his
or her own interest. In this model, the sole link between them is free
exchange, as exemplified particularly by contracts between individuals:
[t]hus social solidarity would be nothing more than the spontaneous agreement
between individual interests, an agreement of which contracts are the natural
expression. The type of social relations would be the economic relationship,
freed from all regulation, and as it emerges from the entirely free initiative of
the parties concerned. In short, society would be no more than the establish-
ment of relationships between individuals exchanging the products of their
labour, and without any social action, properly so termed, intervening to
regulate that exchange.79
Such a society would be unstable as `every harmony of interests conceals
a latent conflict, or one that is simply deferred'.80 Nor was there an evolution
towards such contractually-based relations as industrial societies developed,
for private relations were taking on a public character, and `[w]herever a
contract exists, it is submitted to a regulatory force that is imposed by society
and not by individuals: it is a force that becomes ever more weighty and
complex.'81 Regulation of contracts is needed, but its extent cannot be
delimited in advance.82 Indeed:
the rules of professional morality and law are categorical, like the others. They
force the individual to act in accordance with ends that are not for his own, to

77 Lukes, op. cit., n. 74, p. 10.

78 E. Durkheim, The Division of Labour in Society (1984, tr. W.D. Halls). The debate
is well discussed in H. Jones, The French State in Question: Public Law and
Political Argument in the Third Republic (1993) 154±9.
79 id., pp. 151±2.
80 id., p. 152.
81 id., p. 158.
82 id., p. 163.

380

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
 make concessions, to agree to compromises, to take into account interests
superior to his own. . . . Thus altruism is not designed to become, as Spencer
would wish, a kind of pleasant ornament of our social life, but one that will
always be its fundamental basis. How indeed could we ever do without it?
Men cannot live together without agreeing, and consequently without making
mutual sacrifices, joining themselves to one another in a strong and enduring
fashion. Every society is a moral society.83
Though including an essential moral theme, this is not a mere moralistic
appeal, rooted as it is in social solidarity as a social fact rather than only a
moral ideal.84
A further relevant theme within Durkheim was an institutional one; the
role of occupational associations as regulatory, or rather self-regulatory
institutions; this also foreshadows much recent discussion of `decentred'
regulation.85 Such organizations would work alongside a strongly inter-
ventionist state, which:
had special responsibility to impose rules of justice on economic exchanges, to
ensure that `each is treated as he deserves, that he is freed of all unjust and
humiliating dependence, that he is joined to his fellows and to the group
without abandoning his personality to them'. The State was `above all, the
organ par excellence of moral discipline'.86
I shall say more below on the implications of this work for regulation. A
more direct link with public law, and in particular the concept of public
service, can be found in that of Durkheim's disciple, LeÂon Duguit.87 In this
work the theme of social solidarity is centred around the quintessentially
French concept of public service, which is conceived as replacing the older
concept of sovereignty as the basis of public law. Reflecting Durkheim, the
real basis of public service was social interdependence.88 The central
elements contained in it were `the existence of a legal obligation of the rulers
in a given country, that is to say of those who in fact possess power, to
ensure without interruption the fulfilment of certain tasks.'89 Thus:

83 id., pp. 172±3.

84 See Cotterrell, op. cit., n. 73, pp. 5, 201±3.
85 See Black, op. cit., n. 8, and her `Decentring Regulation: Understanding the Role of
Regulation and Self-Regulation in a ``Post-Regulatory'' World' (2001) 54 Current
Legal Problems 103-46. Durkheim insisted that labour be represented in such
associations, not a universal characteristic of more recent decentred regulation; see
Cotterrell, id., pp. 165, 176±80.
86 Lukes, op. cit., n. 74, p. 272 (footnotes omitted); for more on the role of professional
associations, see pp. 265±8, 271±2.
87 See, in particular, L. Duguit, Law in the Modern State (1970, tr. F. and H. Laski, first
published in English, 1919) especially ch. II; for discussion of his relationship with
Durkheim, see Cotterrell, op. cit., n. 73, pp. 39±41; for fuller discussion of the bases
of his theory, see Jones, op. cit., n. 78, pp. 159±79.
88 id., p. 43; L. Duguit, `The Law and the State' (1917) XXXI Harvard Law Rev. 1, at
183±4.
89 id., p. 39.

381

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
 [a] public service . . . may be defined as follows: any activity that has to be
governmentally regulated and controlled because it is indispensable to the
realisation and development of social solidarity is a public service so long as it
is of such a nature that it cannot be secured save by government intervention.90
Examples of activities where this would be the case were communications,
transport, and electricity.91 This sounds unfashionably statist. However,
Duguit also accepted that public service could be delegated through
concessions to both public and private institutions; the principles of public
service would apply equally to both the government and to private actors.92
The state had the positive obligation to `intervene to protect and to guarantee
against all obstacles the manifestations of individual activity working to the
realization of social solidarity.'93

SOME IMPLICATIONS FOR REGULATION

I would suggest that in the work of Durkheim and Duguit, summarized very
briefly here, there are roots for an approach to regulation based on social
solidarity. This in turn can have two roles: the first, in those areas where we
adopt market-based allocation, is to provide the necessary legal and moral
foundations for them. Thus, rather than law correcting market failures, law
serves to constitute market relations. This theme has been much discussed
recently and there is no need to rehearse it once more here.94 Much of the
discussion concerns the role of law in allocating property rights and provid-
ing techniques or instruments which are needed to make markets work, for
example, through contract or tort law.95 Durkheim's work goes further than
this, as shown in the debate with Spencer referred to above: a major role for
law and regulation will be to provide the essential social underpinning of
mutual trust and expectation which is necessary for markets to function. This
approach thus reverses the assumptions underlying many of the critiques of
regulation discussed above; rather than seeing it as an external constraint on
markets, it is a necessary source for those conditions which markets require
for their operation.
The second role of regulation based on social solidarity is to prevent or
limit the socially fragmenting role of markets. This is important as a cor-
rective to the market-failure approach, because it does not relegate regulation
to a second best where markets fail. Rather, there may be scope for treating

90 id., p. 48.
91 id., pp. 46±7.
92 id., pp. 52±4.
93 id., p. 184.
94 For example, C. Shearing, `A Constitutive Conception of Regulation' in Business
Regulation and Australia's Future, eds. P. Grabosky and J. Braithewaite (1994) 70.
95 See, for example, T. Daintith, `Regulation' in International Encyclopedia of
Comparative Law, Vol. XVII, eds. R. Buxbaum and F. MaÂdl (1997) 10±16.

382

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
regulation as a first-choice means of social organization where we value
social solidarity more than we value the undoubted benefits of markets in
terms of promoting efficiency and consumer choice. Moreover, it suggests
that the apparently simple solution of market allocation corrected by
distributional means through social welfare benefits or by directly targeted
public support to disadvantaged consumers may not be the most appropriate
solution where we wish to avoid social fragmentation.96 Regulatory inter-
ventions may be justified to ensure equal treatment on the grounds of
citizenship and inclusivity, and this may, for example, justify geographically
averaged tariffs for public utilities rather than those based directly on costs.
What this theoretical discussion has suggested, then, is that alternative
rationales for regulation may be found which are quite independent from
welfare economics and market failure. In particular, social solidarity may
provide an appropriate rationale, and one which has greater substantive
content than the proceduralist approaches discussed above. The important
point is that it does not assume that regulation is always a second best; instead
it is based on a mix of different values. This then raises the difficult question
of which rationale is most appropriate in different areas of regulation.
The answer will depend on which values we wish to promote in a
particular regulatory environment. Thus where competition is feasible and
where we are predominantly interested in maximizing efficiency and, espe-
cially, consumer choice, we are likely to adopt a version of regulation which
is concerned only with market failure. This regulation will be especially
concerned with abuse of a dominant position, and social concerns will take
the form of a safety net protecting universal service. This is, in fact, what we
have in the European Union's new regulatory framework for electronic
communications.97 On the other hand, where rights are at issue or social
solidarity concerns about equal access to essential services in social or
geographical terms predominate, then a different approach to regulation will
be more appropriate; health and education would provide fruitful examples
where equal standards of service and guaranteed provision are required.
This might appear to be a somewhat bland conclusion simply involving
the pragmatic balancing of different values. However, although I have
suggested that there will be different rationales for regulation which will
apply in different situations, this is not to say that everything is a form of
relativistic mix in which no substantive values can be found. It is in those
very substantive values that we find the guides we seek for the most
appropriate form of regulation. For example, Roger Brownsword has argued
that what he terms `techno-regulation' is unacceptable as a regulatory
technique because it by-passes practical reason and prioritizes control over

96 Compare Morgan, op. cit., n. 35, p. 158.

97 For a useful summary, see R. Richardson, `The New Electronic Communications
Regime: Ofcom One Year on, a Missed Opportunity or the Foundations for
Success?' (2004/5) 14 Utilities Law Rev. 41.

383

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
choice. By suppressing individual autonomy, such regulation compromises
fundamental human values, especially human rights and human dignity.98
This indeed points to an irony in the economic view of regulation. Despite
the stress on consumer choice as an underlying goal, the individual who is
subject to this view of regulation, and indeed of the wider view of the way in
which markets work, is a curiously non-autonomous entity. His or her
preferences are fixed exogenously rather than being developed through a
learning process and participation in social cultures, and those preferences
are limited to the satisfaction of material wants through self-interest (a vision
of social life vehemently criticized by Durkheim99). The position has been
put well by Hollis in his discussion of:
the lurking presence of a rational individual behind every allocation of
resources. On the one hand he is an actor who seeks to maximise his utility by
rational choices which bring him nearer to the margin of indifference. On the
other hand there is a marked tendency to a behaviourism, in which revealed
preference is sufficient evidence of desire and rationality assumptions are used
to eliminate all actual differences between men placed in economically similar
settings.100
Hollis fruitfully reconstructs the rationally autonomous elements in the model
so as to transcend the assumptions of neo-classical economics. Nevertheless,
for our purposes, if we accept the different rationales for regulation set out
above, the rational economic model in unreconstructed form may be a
seriously misleading guide for regulators.101
From the 1980s onwards the predominantly economic approach central to
regulatory literature and to the politics of regulation has seen regulation as
essentially a means of mimicking markets through the creation of
appropriate incentives to shape the behaviour of economic actors, who are
seen as rational agents responding to such incentives in the way which best
promotes their own preferences and self-interest. Thus the regulated are
isolated, calculating individuals. My approach suggests that individualized
incentives may not always be the best technique to adopt; appreciating and
supporting cultural norms, including those of public service, may often be
more effective.102 The point has been put extremely vividly by Amartya Sen,
who criticizes within it:
the assumption that when asked a question, the individual gives that answer
which will maximize his personal gain. How good is this assumption? I doubt

98 Brownsword, op. cit. (2005), n. 4 and op. cit. (2004), n. 4, at p. 204.

99 Cotterrell, op. cit., n. 73, pp. 18±21.
100 M. Hollis, Models of Man: Philosophical Thoughts on Social Action (1977) 17, also
37±8; Hollis and Nell, op. cit., n. 1, especially pp. 205±32, 243±5, 264±6.
101 For Durkheim's position, see Cotterrell, op. cit., n. 73, ch. 2, especially at p. 30.
102 There is of course a considerable literature relating to issues of regulatory
compliance which also accept this point: for the complexity of compliance issues:
see, for example, B. Hutter, Compliance: Regulation and Environment (1997); K.
Yeung, Securing Compliance: A Principled Approach (2004).

384

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
 that in general it is very good (`Where is the railway station?' he asks me.
`There' I say, pointing at the post office, `and would you please post this letter
for me on the way?' `Yes', he says, determined to open the envelope and
check whether it contains something valuable.)103
He argues instead for admitting into economic theory behaviour based on
commitment, in the sense of an individual choosing an act which will yield a
lower level of personal welfare than an available alternative, for example, on
moral grounds. Indeed (reminiscent of Durkheim and of the decentred
regulation literature), `[g]roups intermediate between oneself and all, such as
class and community, provide the focus of many actions involving commit-
ment.'104 This approach suggests that regulation must be based around a
variety of different values which are appropriate to the context in question,
and that there is no overarching, first-choice economic logic that can form
the basis for regulatory decisions. This in turn suggests that the tools by
which we implement regulation should adopt a more sophisticated model of
what is needed to influence the behaviour of those regulation seeks to
influence. In particular, there needs to be an acceptance that the behaviour of
those regulated, especially where regulation affects organizations which
deliver public service goals, may be profoundly affected by culture and ethos
rather than responding directly to economic incentives.105 The fostering of,
and support for, appropriate cultures may be an important means of
regulating for social solidarity.
The approach I have suggested also has important implications for the
concept of consumer choice. Clearly, in many areas we will want to promote
consumer choice through established mechanisms of regulation for com-
petition. However, in others the values of social solidarity I have outlined
above may press for a different form of regulation based around common
values which, by their nature, may limit the scope for consumer choice in
favour of the provision of common standards for all; this is particularly
important in relation to current debates on healthcare and education,
themselves public services in which both the role of markets is controversial,
and the different forms of regulation potentially available are increasingly
debated.
There are a number of further implications for institutional design in
regulation that can be drawn from the theoretical approach I have outlined
above. The first is that there is no single correct regulatory approach which
can be based on an economic or other logic. Regulators have to weigh
competing values which will often be in conflict; in this sense they are

103 A. Sen, `Rational Fools: A Critique of the Behavioural Foundations of Economic

Theory' (1977) 6 Philosophy and Public Affairs 317, at 331±2.
104 id., p. 344.
105 For discussion of the public service ethos, see Public Administration Select
Committee, The Public Service Ethos, HC (2001-2) 263; for a similar proposal to
my own, see C. Scott, `Services of General Interest in the European Union:
Matching Values to Techniques' (2000) 6 European Law J. 310, at 325.

385

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
`governments in miniature'.106 Obviously, deciding fundamental value
conflicts will not be a matter for regulators themselves unaided by guidance
from government, something which now seems broadly accepted in relation
to United Kingdom regulators, for example, in the form of guidance to the
utility regulators in relation to social and environmental policies and that to
the Environment Agency on its objectives and contributions to sustainable
development.107
The second point is one that has also been made frequently recently: the
need for regulation to be open to its environment. The discussion above
makes it clear that there is no single set of values which can be imposed by
regulators on the regulated; rather, what we will want will depend on the
context of the regulation and the particular values which we may wish to
promote in that particular situation. This is something that has of course been
central to the literature on regulation derived from autopoiesis; however,
unlike in that literature, I am suggesting here that sensitivity to the environ-
ment can be combined with the selection of fundamental regulatory values
whose appropriateness will each depend on the situation in question. After
all, environments are notoriously tricky conceptual tools; there are many
different aspects of any environment and inevitably we shall have to perform
a process of selection of what is relevant from them. Indeed, there may be
some regulatory contexts in which we wish deliberately to irritate an
environment and inject into it broader public interest values. In other words,
regulation is a task which is indeed highly complex but is not wholly
context-dependent. There is also no incompatibility between the approach
proposed here and the important insight that it is helpful to conceive of
regulation widely, involving public and private actors as regulators and a
wide range of different institutional arrangements, not only those purpose-
fully designed by the state but also those which arise spontaneously.108 The
institutions for regulation will be many and varied, but the values that are to
be applied, both to process and to substance, are more important than the
specific institutional form.

106 Prosser, op. cit., n. 24, pp. 305±6. Compare the review of this book by Colin Scott
([1997] Public Law 740±2) which rightly emphasizes the limits on the powers
delegated to the utility regulators by government. This is, however, not to deny that,
within the powers which they do have, decision making has to respond to multiple
values.
107 See Prosser, op. cit., n. 2, 74±6; Defra, op. cit., n. 27.
108 See C. Scott, `Spontaneous Accountability' in Rethinking Public Accountability, ed.
M.W. Dowdle (2006, forthcoming).

386

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School
 CONCLUSIONS

I hope that my approach suggests a potentially fruitful research agenda

through the creation of varied models of regulation derived from different
values. Thus it should be possible to construct ideal types of regulation based
on different social values: these may be those of maximizing competition,
choice, and efficiency (already well catered for in the economics-based
work). However, they may also be the protection of human rights or the
maintenance of social solidarity. Each ideal type will have different implica-
tions for regulatory design and regulatory technique. They can then be
applied to different regulatory contexts to assess the best types of design and
technique to suit the values which are to be applied there. For example, we
should expect that very different techniques will be appropriate for
regulating human embryology (where human rights and social solidarity
will predominate) from those in competition law (the relevant values here
being choice and efficiency). The development of a range of distinct ideal
types may also have the advantage of countering the danger of the expansive
concept of regulation employed earlier, this danger being that a range of very
different activities are included within the same conceptual umbrella and so
perhaps made subject to inappropriately uniform technique.
At first sight this excursion into regulation and social theory might appear
somewhat fanciful and far-removed from the reality of everyday regulatory
decision-making. However, I would argue that it represents a more accurate
view of what regulators actually do in practice than the market-failure
approach (though showing this will involve further empirical study). As I
argued at the outset, it is just not possible to conceive accurately of many
major regulators as correcting market failure and, even for those where this
would appear a more accurate view, notably the utility regulators, a sub-
stantial amount of their work has been directed towards social goals.109 The
same is even more true at the European level. It may be possible to `trans-
late' social goals into market-failure terms, as was documented in relation to
the Australian experience.110 If we accept the arguments above, however,
this should not be necessary as social regulation is not a second best to
economic regulation. Moreover, the `translation' could water down social
objectives: `[i]n this translation, redistributive equity values were most at
risk, though environmental values were also vulnerable.'111 Recognition of a
wider range of regulatory rationales may help to resolve the paradox of
regulation that is both pervasive and of controversial legitimacy; indeed
specification of a wider range of values to which regulation must respond
also makes more acceptable a wide concept of regulation.

109 See Prosser, op. cit., n. 2, ch. 4.

110 See Morgan, op. cit., n. 35, pp. 146, 184±5, 217.
111 id., p. 146.

387

ß 2006 The Author. Journal Compilation ß 2006 Cardiff University Law School