You are on page 1of 1073







Prof. Migai Akech Law, Democracy and Governance Syllabus 2016



This course examines the role of law in shaping democracy and governance, that is, the
involvement of citizens in determining how the state rules them, and how they rule
themselves. In both contexts, the course seeks to establish how governance manifests itself,
whether it is democratic, and whether law limits it. The course, therefore, proceeds on the
premise that governance should not only be democratic, but should also adhere to the rule of
law ideal, meaning that law should limit it. The course seeks to realize its three objectives by
examining the manifestations of governance in the regulation of various facets of public and
private/personal life, namely politics, finance, morality, the media, and the resolution of


Class 1 - 3: Concepts: Democracy, Governance, Rule of Law
Themes: Democracy; the contribution of liberal theory (or liberalism) to the
practice of democracy; the rule of law; critiques of liberal theory.



CHALLENGES TO LIBERAL DEMOCRACY 3-14, 37-70 (Rowman & Littlefield
Publishers, 2008).


32-72, 91-101, 114-126, 137-141 (Cambridge University Press, 2004).

Prof. Migai Akech Law, Democracy and Governance Syllabus 2016

David Williams and Tom Young, Governance, the World Bank and Liberal

Francis Fukuyama, What is Governance?, 26 GOVERNANCE 347 (2013).

Class 4: The Idea of Regulation
Themes: What is regulation; objectives of regulation; types of regulation state
versus self/market regulation; economic v social regulation.


University Press, 1980).

B. Rasmusen, Economic Regulation and Social Regulation, American Law &
Economics Association Annual Meeting, Paper No. 47, 2005.

Eric Windholz and Graeme Hodge, Conceptualizing Social and Economic
Regulation: Implications for Modern Regulators and Regulatory Activity,
Monash University, Faculty of Law, Research Paper No 2012/05.

Andrei Shleifer, Understanding Regulation, 11 EUROPEAN FINANCIAL
MANAGEMENT 439 (2005).

Migai Akech, Public Law and the Neoliberal Experiment in Kenya: What Should
the Public Interest Become?, Unpublished JSD Dissertation, New York
University School of Law 211-222, 224-234, 260-286 (2004).

Class 5 - 6: Regulating Politics
Themes: Democracy in divided societies the perennial quest for legitimate
governance; electoral management bodies; political parties; the mis(use) of
law in politics.


Arend Lijphart, Constitutional Design for Divided Societies, 15 JOURNAL OF
DEMOCRACY 96 (2004).

Richard S. Katz, Democratic Principles and Judging Free and Fair, in Michael
D. Boda, (ed), Revisiting Free and Fair Elections, An International Round Table on
Election Standards organized by the Inter-Parliamentary Union, Geneva,
November 2004, 17-39.

Prof. Migai Akech Law, Democracy and Governance Syllabus 2016

Jorgen Elklit and Andrew Reynolds, Judging Elections and Election

Management Quality by Process, in Michael D. Boda, (ed), Revisiting Free and
Fair Elections, An International Round Table on Election Standards organized by
the Inter-Parliamentary Union, Geneva, November 2004, 53-73.

Joel D. Barkan, James D. Long, Karuti Kanyinga, Karen E. Ferree, and Clark
Gibson, Kenyas 2013 Elections: Technology is Not Democracy, 24 JOURNAL
OF DEMOCRACY 156 (2013).

Migai Akech, Constraining Government Power in Africa, 22 JOURNAL OF
DEMOCRACY 96 (2011).

Class 7-8: Regulating Finance
Themes: regulatory objectives; regulation of bank and non-bank financial
institutions; regulation of mobile banking.


Darryl Biggar and Alberto Heimler, An Increasing Role for Competition in the
Regulation of Banks, International Competition Network, Antitrust
Enforcement in Regulated Sectors Sub Group 1, June 2005.

Francis M. Mwega, Financial Regulation in Kenya: Balancing Inclusive Growth
with Financial Stability, Overseas Development Institute (ODI), Working Paper
407, November 2014.

Abd Elrahman Elzahi Saaid Ali, The Regulatory and Supervision Framework of
STUDIES 123 (2015).

Matu Mugo, Regulation of Banking and Payment Agents in Kenya, 6

Catia Batista, Felix Simione and Pedro C. Vicente, International Experiences of
Mobile Banking Regulation, International Growth Centre, Brief 36012, January

Class 9-10: Regulating Personal/Private Life
Themes: rationales, forms and consequences of the regulation of personal life
in a plural legal system.


Prof. Migai Akech Law, Democracy and Governance Syllabus 2016

Collette Akoth Suda, Formal Monogamy and Informal Polygyny in Parallel:

African Family Traditions in Transition, University of Nairobi, inaugural lecture,

Johanna E. Bond, Culture, Dissent and the State: The Example of
Commonwealth African Marriage Law, 14 YALE HUMAN RIGHTS AND

Winifred Kamau, Law, Pluralism and the Family in Kenya: Beyond Bifurcation of
AND THE FAMILY 133 (2009).

Jackton B. Ojwang and Emily Nyiva Kinama, Woman-to-Woman Marriage: A
Cultural Paradox in Contemporary Africas Constitutional Profile,

L.N.W. v Attorney General and 3 others, Nairobi High Court, Constitutional
Petition No. 484 of 2014.

Okoth Okombo, Judges Verdict on Birth Records May Lead to More Abortion
Cases, Saturday Nation, 4 June 2016, 13.

The Marriage Act 2014, No. 4 of 2014.

Class 11-12: Regulating the Media
Themes: Nature of the media old and new; role of the media in
democratization; forms and politics of media regulation; media and citizenship
the question of hate speech.


Sheila S. Coronel, The Role of the Media in Deepening Democracy, United
Nations Public Administration Network, 2001.

BELONGING 25-80 (Zed Books, 2005).

Fredrick Ogenga, Political Economy of the Kenyan Media: Towards a Culture of

Levi Obonyo and Clayton Peel, Media Regulation in Emerging Democracies:
The Example of Kenyas Hybrid Model, 5 AFRICAN COMMUNICATION
RESEARCH 139 (2012).

Prof. Migai Akech Law, Democracy and Governance Syllabus 2016

Grace Mutungu, New Media in Kenya: Time for Regulation?, Article 19, 13
September 2012.

Alexander Tsesis, Dignity and Speech: The Regulation of Hate Speech in a
Democracy, 42 WAKE FOREST LAW REVIEW 497 (2009).

Class 13-14: Regulating Dispute Resolution Institutions
Themes: judicial independence and accountability; the judiciary as a neutral
arbiter among competing political interests; legitimacy of formal versus
informal justice systems; regulating informal justice institutions.


Nuno Garoupa and Tom Ginsburg, Guarding the Guardians: Judicial Councils
103 (2009).

Migai Akech and Patricia Kameri-Mb0te, Kenyan Courts and the Politics of the
Rule of Law in the Post-Authoritarian State from 1991 to 2010, 18 EAST

Leigh T. Toomey, A Delicate Balance: Building Contemporary Customary and
State Legal Systems, 3 LAW AND DEVELOPMENT REVIEW 156 (2010).

Tanja Chopra, Dispensing Elusive Justice: The Kenyan Judiciary Amongst
Pastoralist Societies, 2 HAGUE JOURNAL ON THE RULE OF LAW 95 (2010).


Political Studies (1994), XLII, 84-100

Governance, the World Bank and

Liberal Theory


School of Oriental and African Studies, University of London

We examine the recent debates about governance, focusing particularly on the World
Bank and identify certain factors which have in recent years moved the Bank's
thinking beyond narrowly economic notions of development. Our account is tentative
and we suggest further avenues of research. We try to connect the Bank's thinking
systematically with key features of liberal discourse and suggest that thiscan do much
to illuminate practice. We illustrate this with a discussion of the growing relationship
between the Bank and NGOs, to contribute to forms of analysis which go beyond the
ideas vs. interests polarities that still inform so much of contemporary social and
political theory.

There ought not to be two histories, one of political and moral action and one of
political and moral theorizing, because there were not two pasts, one populated
only by actions, the other only by theories. Every action is the bearer and
expression of more or less theory-laden beliefs and concepts; every piece of
theorizing and every expression of belief is a political and moral action.
Alasdair Maclntyre, After Virtue, p.61

Governance: What's in a Name

Good government and governance have become almost an obsession in current
debates about development.' There has been a flood of academic texts, an
increasing number of conferences and a growing focus upon them by bilateral
donors and multilateral institutions such as the Commonwealth, the ECA, and
the World Bank.* Not that this implies any clear consensus amongst those who
have promoted these terms. At one extreme are those for whom they simply
mean competitive elections. The Director of the Carter Centre of African
Governance has argued that accountability, the most crucial element of gover-
nance, can be best achieved, 'through the requirement that a government's
continuation in office depends on the active approval of the people as expressed
in competitive elections'.' Others see them as more indirectly related to democ-
I For a general discussion of the term see G. Hyden, 'Governance and the study of politics', G .

Hyden and M. Bratton (eds), Governance and Politics in Africa (Boulder, Lynne Rienner, 1992).
2 For the Commonwealth see E. Anyaoku 'The commonwealth and the challenge of democracy'.
Development Policy Review, 10 (1992), 99-106.
3 The Carter Centre of Emory University, African Governance in the J990s,Working papers from
the second annual seminar of the African Governance Program (Atlanta, Carter Centre, 1990),
,(". Political Studies AEsaiation 1994. Published by Blackwell Publishers. 108 Cowley Road, Oxford OX4 IJF. U K and 238 Main Strat.
Suite 501, Cambridge. M A 02142. USA.

racy, in whatever sense of that protean term.4 British Foreign Secretary Douglas
Hurd sees governance as embodying pluralism, public accountability respect, for
the rule of law and human rights, and market principles; while for Linda
Chalker it comprises competition, accountability and respect for the individuaY.5
The former British Shadow Minister for Overseas Development, Ann Clwyd,
suggested that respect for human rights, popular participation and pluralism in
civil society were essential for development.6Francois Mitterand has said that
France would link its aid contributions to efforts designed to lead to greater
liberty and democracy. The German Ministry for Economic Co-operation has
introduced criteria for assessing the kind and volume of aid for a country. These
include popular participation in the political process, responsible and accountable
government and respect for human rights.8 USAID is funding programmes
designed to increase participation in development, to support and strengthen
democratic institutions and civil society and support human rights9
The World Bank has always been concerned with good government in a
rather narrow sense: broadly, managerial or institutional issues relating to
bureaucratic reforms, policy analysis, improving co-ordination and what it calls
the efficiency of public services.1 This concern lies behind such projects as the
African Capacity Building Initiative and is reflected in what the Bank and
others call technical issues. There are those who would see the Bank as
concerned mainly with these issues.I2 Officially the Bank is constrained by its
Articles of Agreement which expressly forbid taking non-economic consider-
ations into account.13 The authors of the early drafts of the Articles, including
Keynes, were at pains to emphasize the neutrality of the institution when it
came to political ideologies and interests.14The World Banks focus upon a new
range of issues, not simply technical and managerial, under the heading of
Claude Ake The case for democracy, and Anyang Nyongo Democracy and political institu-
tions. Carter Centre, African Governance.
D. Hurd, Speech given to OD1 1990; L. Chalker, speech given to the Wilton Park Conference on
Good Government in Africa 1991. Close students of Mr. Hurds political thought might note the
following remark from a fictional work. The setting is a future (presumably Tory) conference at
which the Foreign Secretary must explain why a British soldier has died in a foreign land - he was
doing something new. something in a way more daring and ambitious. He was joining with others in
an attempt to deal with wickedness and cruelty, to establish decency and order not just where the
Union Jack flew (sic) but throughout the world. In the story the speech turns the conference and
receives a standing ovation. Ten Minutes to turn the Devil. A Short Story by Douglas Hurd. The
Observer Review, 31 January 1993.
A. Clwyd The Labour Partys policy on overseas aid, Journal of International Developmenr.
(1992). 94-102.
Interview with FranGois Mitterand, Le Monde, 20 June 1990.
K. Van de Sand and R. Mohs, Making German aid more credible. Development and Coope-
ration, 1 (1992), 4
African Bureau of Information Centre USAID, African Voices. 1 (1992). 4.
Iu See World Bank, Accelerated Developmenr in Sub-Saharan Africa: an Agenda for Action (Wash-

ington DC, World Bank, 1981) pp.3 1 4 ; World Development Report (Oxford, Oxford University
Press, 1983), pp. 115-24; World Development Report (Washington DC, World Bank, 1988).
World Bank, A Frameworkfor Capacity Building in Policy Analysis and Economic Management
in Sub-Saharan Afiica (Washington DC, World Bank, 1989).
I* M. Bratton and D.Rothchild. The institutional bases of governance in Africa. Hyden and
Bratton (eds),Governance and Politics in Africa. p.265; R. Charlick. Governance Working Paper.
given at the Wilton Park Conference on Good Government in Africa.
I) Articles 111 5b. IV 10. and V 5c. See also I. Shihata, The World Bunk in a Changing World:

Selected Essays (Dordrecht, Martinus NijholT, 1991) p.61. Of course in its operations the Bank has
often ignored these.
I4 Shihata, World Bank, p.71.

C Political Studies Asoocidtion, 1994

86 Governance, the World Bank and Liberal Theory

governance thus seems to be an important departure from previous policy,

prompted in large part by its experience in Africa. Indeed it is acknowledged by
a number of writers that the World Bank report Sub-Saharan Africa: from
Crisis to Sustainable Growth, which characterized the crisis in the region as a
crisis of governance has been of crucial importance in stimulating this intense
focus.15 The new focus has more explicitly political concerns with legitimacy,
participation, pluralism, a free press and human rights, and these issues have
become for the Bank a key factor in explaining Africas current crisis.
However, as with the florid statements by politicians, there are problems in
deciphering exactly what the Bank means by governance. The Banks position is
a rather confused one. This is only to be expected perhaps for such a new area of
concern, but it makes for certain analytical difficulties. The Bank is a large and
diverse organization, so much so that even to talk of the Banks position at all
is to oversimplify. Even during the period when most observers would have
considered the Bank as having an unbroken commitment to pricism, there
were departments within it which did not reflect orthodox thinking.16 It is likely
there will be differences over governance as well. To overcome these difficulties
for the purposes of this analysis the official position will be gleaned from Bank
publications, and will focus upon common themes rather than differences of
emphasis. This will be supplemented with an analysis of an unofficial position
as reflected in the publications of the Senior Policy Adviser in the Africa Region
Technical Department, Pierre Landell-Mills.

Governance and the World Bank

As already suggested, the 1989 report, Sub-Saharan Africa: from Crisis to
Sustainable Growth, clearly marked a watershed in the emergence of governance
issues in the Banks thinking. While it continued to register a concern for the
appropriate economic policies that had characterized earlier reports, it also
argued that a crisis of governance underlay the litany of Africas development
problems.I7 What was needed in the region above all was political renewal
involving the creation of a pluralistic institutional History sug-
gests (the report argued), that political legitimacy and consensus are a precon-
dition for sustainable development. The two countries with the best economic
performance in Africa, Botswana and Mauritius, both had effective parliamen-
tary democracie~.~ There are two aspects to this analysis. The first concerns so-
called technical areas. This has particularly revolved around establishing a legal
framework for development and capacity building. The former is to involve
both an instrumental element which concentrates on the formal elements
necessary for a system of law to exist, such as a set of rules known in advance,
G. Hyden. Governance and the study of politics, Hyden and Bratton (eds), Governance and
Politics in Africa, p.5; M. Bratton and D. Rothchild, The institutional bases of governance in
Africa. p.265. World Bank, Sub-Saharan Africa: j i o m Crisis to Sustainable Growth (Washington
DC,World Bank, 1989), p.60. See also The emergence of the Good Government agenda: some
milestones, Good Government? IDS Bulletin, 24 (1993), p.7.
16 M. Lipton, The limits of price policy for agriculture: which way now for the World Bank?,

Development Policy Review, 5 (1987). 197-215; P.Mosley, J. Harrigan, and J. Toye, Aid and Power,
vol. 1 (London. Routledge, 1991). p.24.
17 World Bank. Sub-Saharan Africa:Jrom Crisis to Sustainable Growth, p.60.

18 World Bank, Sub-Saharan Africa: from Crisis to Sustainable Growth, p.6, p.61.

I v World Bank. Sub-Saharan Africa: from Crisis to Sustainable Growth, p.60 emphasis added,
I Political Siudies Asmation. 1994

and an independent judiciary, and a substantive element which refers to the

content of the law and concepts such as fairness, justice and liberty. The
World Bank sees its role as facilitating the communication of laws more effec-
tively, ensuring the consistency of laws, updating legal systems, and training the
judiciary. The latter is to involve improving policy analysis and budget disci-
pline, improving training and bureaucratic procedures, reforming the civil
service, particularly the reduction of overmanning, improving bureaucratic co-
ordination and establishing a distinction between public office and private
person. It is clear that the model for both the legal system and the bureaucracy
is a Western liberal one. Both of these sets of reforms are presented as free from
any political or ideological concerns; the existence of such a system is a basic
requirement . . . for a modem state.*
Second, governance is concerned with civil society. This involves support for
voluntary organizations and NGOs, the informal and formal economies as well
as for the more obvious groups in civil society such as universities, trade unions,
and professional organizations. It involves, building a pluralistic institutional
structure, and creating intermediaries between the government and the peo-
ple.22 These intermediaries have an important role to play; they can create links
both upward and downward in society and voice local concerns more effectively
. . . they can also exert pressure on public officials for better performance and
greater accountability.23The Banks promotion of civil society is linked to its
promotion of accountability, legitimacy, transparency and participation as it is
these factors which empower civil society and reduce the power of the ~tate.2~
Accountability is crucial to ensure congruence between public policy and actual
implementation, and the efficient allocation and use of public resource^'.^^ It is
to be encouraged at the macro and micro levels. At the macro level, there should
be effective systems of financial accountability and systems of accountability
for economic performance. At the micro level, accountability and legitimacy
should be encouraged by competition providing opportunities for exit, and
participation providing opportunities for voice. The Bank sees its role as
encouraging competition, providing comprehensive information, encouraging
auditing reforms, encouraging participation in the design and implementation
of development projects, and making use of non-governmental organizations to
ensure that the voice of the poor is heard.26 The Bank has also encouraged
decentralizing administration and strengthening local government.27

20 For the rule of law see Shihata, World Bank, pp.81-8. See also World Bank, Governance and

Development (Washington DC, World Bank, 1992), pp.29-39, and Sub-Saharan Africa: from Crisis
to Sustainable Growth, p.55. For building capacity see Shihata, World Bank, p.91; World Bank,
Sub-Saharan Africa: from Crisis to Sustainable Growth, pp.55-6; A Frameworkfor Capacity Building
passim; World Development Report 1988 (Washington DC, World Bank, 1988)passim; R. M. Lacey,
Managing Public Expenditure: an Evolving World Bank Perspective (Washington DC,World Bank,
2 Shihata. World Bank, p.85 emphasis added.

22 World Bank, Governance and Development, p.49; Sub-Saharan Africa: from Crisis to Sustainable
Growth, p.61; A Framework for Capacity Building, p.6.
23 World Bank, Sub-Saharan Africu:/rom Crisis to Sustainable Growth, p.61.
3 Shihata, World Bank, p.90, World Bank, Governance and Development, pp.13-26.
2) World Bank, Governance and Development, pp. 1 3 4 .
26 World Bank, Governance and Development, p.49.

27 World Bank, Governance and Development, pp.214; Sub-Saharan Africa: from Crisis to Sus-

tainable Growth, p.58.

C,Political Studies Association, 1994
88 Governance, the World Bank and Liberal Theory

Transparency is essential for economic efficiency, and for the prevention of

corruption.28.This focus upon the construction of civil society is important for
understanding the Banks position on democracy. It has shied away from
outright advocacy of democracy preferring instead to focus upon the values
of liberal democracy (participation, accountability, legitimacy and so on)
and the construction of what we will later call a liberal public sphere.
Pierre Landell-Mills, speaking in his personal capacity, is freer to say things
the Bank could not publicly endorse. His conclusion is that, to be workable the
governance of African states needs to be systemarically rebuilt from the bottom
up. This implies effective empowerment of a variety of interest groups - not just
grass roots, community and womens organizations, but also professional and
business associations and other intermediary private voluntary organisations.*9
An articulate and empowered middle class he continues, echoing Barrington
Moore, is a necessary precursor to stable democratic politics, and NGOs are
their training ground.30The donor community should be involved in empower-
ing these groups in civil society at a national and local level.31 Particularly, the
Bank should promote institutional pluralism and foster NGOs and grass roots
organizations. He says bilateral donors and NGOs should press for more
explicit political concerns such as respect for human rights and elections.32 This
unofficial view is also, in its rhetoric at least, concerned to build on the
indigenous and respect national sovereignty.
It is clear then that these are new areas of concern for the Bank. It has moved
away from a narrowly economic notion of development and a concern with
good government in the old-fashioned sense. What has helped to produce this
shift? There seem to be at least five key factors in explaining the emergence of
the governance issue within the Bank. Four of these are discussed briefly now.
We return to the fifth, relations with NGOs, at the end of this article.
Flavourof the month Syndrome
The new Bank President, Lewis Preston, has admitted that the Bank has a
flavour of the month syndrome. The crucial questions are of course what
flavour, and perhaps more importantly, whose flavour. The Bank has always
had a complex relationship with what we might call the development ortho-
doxy in the wider development community of Western governments, NGOs
and academics. In some ways the Bank reflects the orthodoxy and in others the
Bank has helped to produce it. The new focus upon governance is in part based
upon a growing consensus in the wider development community that political
considerations play a crucial role in determining development. Clearly though
this consensus is itself partially the result of the Banks own focus on gover-
nance. As to whose fashion there can be little doubt that the Bank is influenced
by Western countries and in particular the United States. In 1980, when the
?* World Bank. Governance und Development, pp.39-47.
3 P. Landell-Mills. Governance,Civil Society and Empowerment in Sub-Saharan Africa, paper
prepared for the Annual Conference of the Society for the Advancement of Socio-Economics 1992,
emphasis added.
IO Landell-Mills, Governance, Civil Society and Empowerment.
1 P. Landell-Mills and 1. Serageldin, Governance and the external factor, Proceedings of the
World Bank Annual Conference on Development Economics (Washington DC, World Bank, 1991).
2 Landell-Mills and Serageldin, Governance and the external factor, and Governance and the
development process, Finance and Development, 28 (1991). 14-7.
I Landell-Mills and Serageldin, Governance and the external factor.
Independent on Sunduy. (London) 10 May 1992.
I Political Sludm Associalion, 1994

Reagan administration came to power, it criticized the Bank for promoting

socialism and undermining capitalist development, and pressured the Bank to
come into line with the US governments foreign policy aims.3s The Reagan
Budget Director said the Bank, has not been vigorous in using the leverage
inherent in its large lending programmes to press recipients to redirect their
economies towards a market ~ r i e n t a t i o n . ~ ~
It would be wrong however to see the Bank as entirely the tool of the US or
Western governments. Clearly the relationship is more complex than that and in
many ways the Bank is an important influence upon these governments poli-
cies. Yet of course this influence is important, and is sure to be felt as the US and
other Western governments re-orientate their foreign policies in response to the
changes in Eastern Europe. The US has always had a tradition of focusing on
human rights in its foreign policy and it seems likely that this will increase in
coming years.37 USAID has an African Bureau Democracy and Government
programme. Other Western governments such as France, Britain and Germany
are increasingly concerned to tie aid to democratic reforms.38It is important not
to be deterministic (and also to recognize that there may be important differ-
ences between Western governments on this issue), nonetheless it is indisputable
that with the US holding 17.59% of the votes in the Bank, Germany 6.19%, and
Britain and France 5.93%, these countries are bound to have an influence on
Bank policy.39

Experience with Adjustment Lending

We will not be concerned here directly with criticizing the Banks adjustment
policies. These criticisms include the view that such policies are not consistent
with Africas long-run development needs; that there is no theoretical basis for
advocating wholesale economic liberalization; that structural adjustment fails
to stimulate a supply-side response; and that adjustment unduly hurts the
poor.O Of course, the economic effects of adjustment are, and will probably
remain, conte~table.~ What is important is that it became clear to the Bank that
part of the reason for the limited (at least) success of structural adjustment were
political considerations that went beyond the size of the public sector or increas-
ing the technical capacities of states.42 Whether it was viewed as a success or a
failure the experience of adjustment lending led the Bank to take account of
political factors such as interest group pressure and government legitimacy as
somehow important. The second point is that the Bank has recognized some of

35 R. Ayres, Banking on the Poor: the World Bank and World Poverty (Cambridge, Mass., MIT
Press, 1983) pp.11-2, p.231.
36 Quoted in Ayres, Banking on the Poor, p.12.
17 See C. Lancaster, Governance and development: the views from Washington, Good Govern-

ment? p. 13.
38 For an overview see S. Riley, The Prospects for Democracy in Africa, Talk given at Chatham
House 1991.
39 Figures from World Bank, Annual Report 1991 (Washington DC, World Bank, 1991) p.235.
00 F. Stewart, Are adjustment policies consistent with Africas long-run development needs?,
Development Policy Review, 9 (1991), 413-39; Mosley et al.. Aidand Power, p.92; J. Toye. Dilemmas
of Developmenf (Oxford. Blackwell. 1987) pp.74-5; M. Lipton, Requiem for adjustment lending,
Development Policy Review, 8 (l990), 43743.
41 See for example C. Stoneman, The World Bank Versus Zimbabwe: Structural Adjustment or

Development?, paper presented at University of York CSAS 1991.

42 World Bank, World Development Report 1990, p.115; Shihata, World Bank, p.53.

Politicdl S:udies Association. 1994

90 Governance, the World Bank and Liberal Theory

the shortcomings of previous policy. It has in recent years concentrated more

upon infrastructure, health and education in an attempt to reduce supply-side
problems, and on alleviating some of the social problems associated with
. ~ ~ led to the rather obvious conclusion that the state has an
a d j ~ s t m e n tThis
indispensable role in development, but it did focus attention on to the kind of
state that was needed and what its relations with the populace should be.

Internal Bank Factors

These are difficult to assess, as are the internal workings of any large organiza-
tion and it is difficult to come to any firm conclusions, but certain features do
seem important. First the decline of the Economic Research Staff which used to
dominate report writing during the years of the pricist orthodoxy, but which
disappeared as a separate vice-presidency in 1987.45As a result, in part at least,
the Bank became much more open to outside influence, and adopted a more
consensual style. Second, in 1985 the Economic Development Institute (EDI)
was brought within the operations complex of the Bank and had its financial
allocation increased with the object of building country capabilities for econ-
omic management, policy reforms and senior level training.& This has been
reflected in programmes such as the African Capacity Building Program and
other technical reforms. The ED1 organized a set of political economy semi-
nars in 1986 which focused upon the political constraints to structural adjust-
ment and has continued to produce discussion papers and reports on the issue
of It is clear that in some way the Africa region of the Bank has
had an extraordinary influence. It was the 1989 report which focused Bank
thinking, and has become the basis for a general Bank position on governance!*
What the dynamics of this adoption are is unclear (it is perhaps that Africa
seems the most pressing area of concern), but it would be intriguing to see what
the South East Asia region, for example made of the focus upqn legitimacy,
participation, and consensus as the basis for economic d e ~ e l o p m e n tWithin
the Africa region the senior policy adviser, Pierre Landell-Mills seems to have
been important. He was staff director on the 1983 World Development Report
which focused upon Management in Development, and more importantly he
was task manager and co-ordinating author on the 1989 Sub-Saharan Africa
report. He continues to produce provocative pieces on the issues of governance.
Of course much of this is speculation. The Bank has a reputation for secrecy,
and by their very nature these internal politics are difficult to study. Many
questions remain unanswered. For example it is unclear how, and how long it
takes for these governance ideas to feed into operations, or what the relations
3 World Bank, Sub-Saharan Africa: from Crisis to Sustainable Growth, and World Development
Report 1990, passim. See also World Bank The Social Dimensions of Adjustment in Africa (Wash-
ington DC, World Bank, 1990).
World Bank, Sub-Saharan Africa: from Crisis to Sustainable Growth, p.55; Governance and
Development, pp.6-I.
J Lipton, Limits of price policy; Mosley et al. Aid and Power, p.24.
16 G. de Lusignan, The Banks Economic Development Institute, Finance and Development, 23
(1986), 28-3 I.
41 See for example L. Adamolekun, Issues in Development Management in Sub-Saharan Africa
(Washington DC,World Bank, 1989).
a Shihata, World Bank, p.55; World Bank, Governance and Developmeni, pp.45.
Something hinted at in M. Moore, Declining to learn from the East? The World Bank on
governance and development , Good Government? 39-50.
0 Polrtlcal Sludus Assmarion. 1994

are between the various departments and regions within the Bank. All of these
issues would repay further investigation.

Academic Influences
These seem to have been of two kinds. Firstly, influence has been exerted in the
general sense that academic trends are part of the operational environment of
the Bank, though of course only some trends have influence. Beyond that the
Bank also hires and consults with individuals or groups of academics. In regard
to governance these influences have clearly been at work in at least two areas.
First, in the preparation of the 1989 Sub-Saharan Africa report the Bank
commissioned various background papers which have been published in four
volumes.50The 1989 Report particularly acknowledges the contributions of
Claude Ake, Goran Hyden, Jaques Giri and Janet Macgaffey among others to
these volumes. Drawing on the work of these authors has enabled the Bank to
develop two themes which were not present before. One is illustrated in Volume
3 (Institutional and Socio-political Issues) where the contributions of Ake
(Sustaining Development on the Indigenous) and Hyden (Creating an Ena-
bling Environment), among others are concerned with the theme of respecting
Africas indigenous roots in attempting to build its fut~re.~ The second is the
focus upon civil society particularly the informal economic sector, grassroots
organizations, and NGOs.52 This focus upon civil society has become a common
theme of writing on governance, and perhaps reflects in part the growth of
academic interest in civil society in recent years.53These two strands are proble-
matic and we discuss them further below.
The second area has been that of providing a theory of African politics which
can most easily be summed up as negative politics. This assessment has been
derived from the New Political Economy (Rational Choice or Public Choice)
framework. Here traditional micro-economic assumptions about the primacy
of self-interest are applied . . . to the political claims of citizens, to the actions of
politicians and policy makers, to the behaviour of bureaucracies and even to the
actions of states more One influential commentator on the Bank
has suggested that it has been strongly influenced by these ideas, particularly
the general move away from seeing the Third World state as either economically

y, World Bank, The Long-Term Perspective Study of Sub-Saharan Africa (LTPS) 4 vols (Wash-
ington DC,World Bank, 1990).
World Bank, LTPS. vol. 3, p.v.
s2 See for example J. Macgaffey, The endogenous economy, and J. Gin, Formal and informal
small enterprises in the long-term future of Sub-Saharan Africa, both in World Bank, LTPS vol. 3.
53 See Carter Centre, African Governance, and Bratton and Rothchild, Institutional bases of

governance. For civil society see M. Bratton, Civil society and associational life in Africa, World
Politics, 49 (198&1989), 407-390, J-F. Bayart, Civil society in Africa in P. Chabal (ed.),Political
Dominarion in Africa (Cambridge, Cambridge University Press, 1986); and D. Woods, Civil society
in Europe.and Africa: limiting state power through a public sphere, African Studies Review, 35
(1992). 77-100.
M. Grindle, The new political economy: positive economics and negative politics, in G . Meier
(ed.), Politics and Policy Making m Developing Countries: Perspectives on the New Political Economy
(San Francisco, ICS Press, 1991).
JJ Grindle, The new political economy, p.44. See also R. Bates, Markets and Stares in Tropical
Africa (Berkley, University of California Press, 1981) pp.2-3; M. Staniland, Whar is Polifical
Economy? (New Haven, Yale University Press, 1985) ch.3. For the Bank and new political
economy see Mosley er al., Aid and Power, pp.13-21.
8 Political Studia Association, 1994
92 Governance, the World Bank and Liberal Theory

or politically benign.56 There have also been more specific influences at

This aspect would repay further research into what and how various influences
are absorbed into Bank thinking. It does seem clear however that this concep-
tion of African politics has influenced the Bank: individuals whether in or out
of government, use the resources at their disposal to further their own private
interest; and, appropriation of the machinery of government by the elite to
serve their own ends is at the root of this crisis of governan~e.~~
The importance
of this for us is that it generates prescriptions which aim to make political actors
more accountable, and their actions more transparent. Actors are assumed to
respond rationally and hence will provide what the people want for fear of
losing their jobs.

Governance and Liberal Theory

Liberal theory is broad and rich and there can be no question of offering a full-
scale analysis, or even an exhaustive definition of liberalism here. Yet it would
also be analytically impoverishing not to pursue an approach that explores the
many links between political theory and practice. Michael Sandel has sug-
gested people often see political philosophy residing at a distance from the real
world of politics. But for the study of governance political theory is of crucial
importance. Sandel went on to argue that political philosophy inhabits our
practices and institutions; they are in a real sense embodiments of theory.59
This section follows his suggestion. It first proposes that the World Bank
constructs governance, in part at least, from liberal theory. Second, this con-
struction of governance reproduces some important ambiguities and tensions
that exist in that liberal theory.
Even to identify liberalism is fraught with difficulties. It has been said that
the word itself is a trouble-breeding and usually thought-obscuring term, and
that it has no single unchanging nature or essence.@Here we are concerned
with liberalism as what might be called a working ideology - how it informs
political and other practices, rather than in its most sophisticated formal intel-
lectual statements, though we do try to link the two. Despite the difficulties, the
view taken here is that it does seem to denote or delimit a field of arguments or a
bounded area of disagreement. Within this field there are critical issues such as
liberty, choice and rights, and dispute over their various justifications and
limits. It is arguments over these particular issues which distinguish liberalism
from other intellectual traditions. Here we will do no more than assert (conven-
tionally enough) that a central feature of modern liberalism is a distinction
between the right and the good and the apparent prioritization of the first

56 J. Toye, Interest group politics and the implementation of structural adjustment policies in
SubSaharan Africa,Journal of International Development, 4 (1992). 184.
57 The title of an early paper by D. Lal, The Political Economy of the Predatory State, DRD
Paper 105 (Washington DC, World Bank, 1984) gives the flavour of this kind of analysis. La1 was an
influential figure in the Research Department of the Bank.
World Bank, Annual Report 1988, pp.49-51; Sub-Saharan Africa: from Crisis to Sustainable
Growth, p. 197.
~9 M. Sandel, The procedural republic and the unencumbered self, reprinted in S.Avineri and A.
De-Shalit (eds), Communitarianism and Individualism (Oxford, Oxford University Press, 1992) p. 12.
M Arthur 0. Lovejoy quoted in S . Lukes, Individualism (Oxford, Blackwell. 1990) p.43; J. Gray,
Liberalism (Milton Keynes, Open University Press, 1986) p.ix.
0Political Studies Association, IW
over the second.6i It is, so the argument goes, philosophically impossible to
choose between notions of the good, and attempting to do so, to establish an
overriding notion of the good, will only end in conflict and violence. This in turn
generates interlocked propositions about the nature of the state, civil society
and the self. Thus modem liberals argue that the State should be a neutral
framework within which competing conceptions of the good can be equally
pursued.62Linked to such notions of a neutral state, indeed required by them, is
the notion of civil society characterized as a realm of freedom in which
individuals engage in formally uncoerced transactions. Finally this complex of
concepts requires a certain notion of the self, a free choosing individual who is
the best, indeed the only judge, of his own interests. Arguably it is this free
individual who is at the core of liberal belief. To what extent is World Bank
thinking shaped by such conceptions?

TechnicalReforms and Neutrality

William Clausen (World Bank President during 1980-6) has been quoted as
saying that the Bank is not a political organisation, the only altar we worship at
is pragmatic economic^'.^^ In the preface to the Berg Report Clausen argued
that its focus upon markets and prices was, not a recommendation which
derived from any preconceived philosophy of ownership. It derives from con-
siderations of efficiency. The General Counsel of the Bank has stated that,
technical considerations of economy and efficiency rather than ideological and
political considerations should guide the Banks work at all times.64 This
entirely familiar (and not very sophisticated) separation of technical issues
from politicsis a common theme of elite discourse in liberal capitalist societies
which nonetheless rests on notions of neutrality. Thus the Bank subscribes to
such notions for entirely conventional liberal reasons, namely that the state (and
by extension the Bank) must be neutral between competing conceptions of the
Yet there are good reasons to suppose that this notion of neutrality (in all its
variants) is unsustainable within liberal theory as well as the distinction between
the right and the good from which it is derived. As Mendus shows, neutrality is

6i Classically J. Rawls, A Theory of Jusfice (Oxford, Oxford University Press, 1971). We are

mindful of skating over important differences within the liberal tradition, Arguably that tradition,
in modem times at least, takes either a utilitarian or rights-based form. For an argument that these
traditions have much in common see I. Shapiro, The Evolution of Rights in Liberal Theory (Cam-
bridge, Cambridge University Press, 1986).
See R. Dworkin. Liberalism,in S . Hampshire (ed.),Public and Privare Morality (Cambridge,
Cambridge University Press, 1978). Waldron points out that while the term neutralityis relatively
novel in this context the term is only the most recent attempt to artculate a position that liberals
have occupied for centuries. J. Waldron, Liberal Rights Collected Papers 1981-1991 (Cambridge,
Cambridge University Press, 1993) p. 143. There are of course sophisticated analysesin the literature
both of the form of such neutrality (e.g. as between intentions and consequences) and indeed
arguments as to whether liberals should abandon the idea. We do not think that this detracts from
the analysis presented here. See the discussion in W. Kymlicka, Liberal individualism and liberal
neutrality, Erhics, 99 (1989), 883-905, and R. Douglas, G. Mara and H. Richardson (eds),
Liberalism and the Good (London, Routledge, 1990).
61 Quoted in T. Hayter and C. Watson, Aid: Realify and Rhetoric (London, Pluto, 1985) p.196.
Shihata, World Bank,p.95.
There are of course other factors in play concerning relations between states and sovereignty
which we neglect here. The imperatives of politeness towards the sovereignty of Third World states
are in any case becoming more threadbare by the day.
0Poliucal Studies Association, 1994
94 Governance, the World Bank and Liberal Theory

an extremely thin principle which must be sustained by some other value usually
(and distinctively within the liberal tradition) autonomy.66Even if neutrality is
taken as a guiding principle rather than a foundational one, it generates neutra-
list conclusions only with respect to those who already accept liberal principles.
In all the constructions of neutrality in contemporary liberal theory, heavily
loaded assumptions about what are valuable ways of life are in fact smuggled
in.67Arguably this sleight of hand has characterized liberal thought from its
beginnings. Salkever suggests that from Hobbes onwards, the liberal preference
for a peacable and comfortable life is established subliminally, as it were, by
repetition that appears to seek the readers complicity, rather than by an explicit
argument that would risk contradicting the neutrality condition.68It is indeed
difficult to avoid the conclusion that, the concept of right, far from being (as the
liberal insists) independent of and anterior to any conceptions of the good, will
in fact be a function of our conception of the good.@
The Banks discourse exactly reflects these theoretical arguments. Its so-
called technical reforms are necessarily guided by a prior conception of the
good. Governments, it argues, must provide rules to make markets work, and
ensure property rights.O The existence of a system of stable law is, a basic
requirement for a stable business environment and the success of investment.
Without it, the fate of enterprises like that of individuals will be left to the
whims of the ruling elite or clique. Legal reform confers stability on contractual
transactions and ensures predictability to property rights. Bureaucracies are to
be judged on, the degree and quality of. . . [their] intervention in the running of
the ec~nomy.~ Clearly there is here a conception of the good of social organiza-
tion and the role of the state within it, that, whatever else it is, is not neutral,
despite attempts to hide behind procedural norms (questions of right). This
concept and its attendant view of the state is only neutral between social and
political forms that already concede its principles. The good for which the
World Bank stands is a market economy and a neutral state which ensures the
proper functioning of that economy by means of the enforcement of property
rights and contractual obligation^.^^ What has changed in the context of gover-
nance is that the Bank, or at least some within it, have come to the view that
such a neutral and effective state, cannot be sustained without a corresponding
liberal public sphere.

Pluralism, Tolerance and Civil Society

We have already suggested that the most innovative part of the World Banks
recent positions is the concern with civil society and especially with the centra-
lity of civil society to development. Traditionally of course such a civil society
* S. Mendus, Toleration and the Limits ofLiberalism (London, Maanillan, 1989) ch.4.
67 See R. Bellamy, Liberalism and Modern Society (Cambridge, Polity, 1992) ch.5 for a useful
demonstration of this.
a S. Salkever, Loppdand bound: how liberal theory obscures the goods of liberal practicesin R.
Bruce Douglas et al. (eds). Liberalism und the Good, p. 17 I .
69 Mendus, Toleration pp. 1 IsC20. There is a useful parallel argument in C. Taylor, Atomism
Philosophical Papers. 2 (1986), p.205 er seq.
World Bank, Governance and Development, p.6.
7l Shihata. World Bunk, pp. 87-90.
72 Of course liberal thinkers notoriously disagree about what the state may do after this point in
terms of patterneddistributions and so on. Their general agreement up to this point does not of
course make it neutral.
@ Political Studies Association, 1994

was seen as a sphere of interactions free of state interference and characterized

by pluralism and tolerance. Much of the governance literature heeds the histori-
cally familiar liberal imperative of tolerance. Landell-Mills has argued that,
Africa will only emerge from its current difficulties if it can progressively
remodel its institutionsto be more in tune with the traditions, beliefs, and structure
of its component ~ocieties.~3 This view is echoed in official Bank documents:
each country has to devise institutions which are consonant with its social
values; . . . for change to be effective it must be firmly rooted in the societies
concerned; World Bank programmes must reflect national characteristics and
be consistent with a countrys cultural values.74The Bank cites the examples of
Japan and the South-East Asian NICs to sustain this view.7s As one Bank
commentator has said, Japan, the Republic of Korea and Taiwan are examples
of economies that have achieved high levels of modem production and
advanced technology while maintaining their unique national traits.76The
assumption is that Africa can do the same. This view may have firm practical
justifications. But it also has theoretical underpinnings in the language of
tolerance and pluralism. Outside models not only cannot but should not be
imposed: The Bank should not prescribe a particular political system.77
Such sentiments would seem to require that close attention is paid to how in
different societies culture, language, social practices and so on interact with
institutions such as the market, state or bureaucracy. Indeed the Bank itself has
both encouraged and carried out such investigations. These appear however to
produce rather paradoxical conclusions. Of course, certain features of African
life may prove to be highly beneficial, and they may embody, to a certain extent,
themes of governance such as participation and legitimacy; but they may not, or
at least they may embody them in rather different ways from Western under-
~tandings.~s Thus Landell-Mills himself has suggested, African managers can-
not easily set aside their loyalties to their community . . . African managers
cannot easily escape the heavy social obligations that take up a large proportion
of their time.79The Bank has said that, family and ethnic ties that strengthen
communal actions have no place in central government agencies where staff
must be selected on merit, and public and private monies must not be con-
fused.80Many other experts, including some working for the Bank, appear to
agree. Diouf has argued that the lifestyle of Senegaleseelites which has had such
deleterious economic effects remains rooted in African culture.8 Bayart has
argued that part of the reason for corruption within the African state is the

7J Landell-Mills, Governance, civil society and empowerment.

74 World Bank, Governance and Development, p.12, and Sub-Saharan Africa: from Crisis to
Sustainable Growth, p.60, p. 193.
World Bank, Governance and Development, p.8.
76 M. Dia, Developmentand cultural values, Finance and Development, 28 (1991), 12.
77 Shihata, World Bank, p.93.
78 For examples of where they are thought to be beneficial see World Bank, LTPS vol. 2. For the
idea that participation and legitimacy can be found in non-official structures see P. Chabal, (ed.),
Political Domination, and D. Cruise OBrien, Saints and Polificians(Cambridge, Cambridge Univer-
sity Press, 1975).
79 P. Landell-Mills, Creating transparency, predictability and an enabling environment for
private enterprise, paper given at the Wilton Park Conference on Good Government in Africa
10 World Bank, Sub-Saharan Africa: from Crisis to Sustainable Growth, p.60.
81 World Bank, LTPS vol. 3, p.60.

0Political S w d m Associauon, 1994

96 Governance, the World Bank and Liberal Theory

impact of what he calls the cultural logic of African politics.82There is then on

the one hand an aspiration to build on the indigenous, and on the other a
recognition that the indigenous may be an obstacle.
This paradox can be better understood by an examination of the underlying
liberal assumptions. Liberals pride themselves on being tolerant and pluralistic.
Indeed notions of tolerance, particularly religious tolerance, have been of prim-
ary importance in the historical genesis of liberal thinking.83The concepts of a
neutral state and a pluralistic civil society are linked not least by the notion of
autonomy and its corollary of human diversity. In connecting these with deve-
lopment the Bank adopts a venerable liberal position rooted in John Stuart Mill
for whom the only unfailing and permanent source of improvement is liberty.
But it also recapitulates all the contradictions of this position and resolves them
in the same highly illiberal way. As Mendus shows, Mills position depends on
distinguishing those who are and those who are not autonomous and sanctions
breaches of the principle of liberty of the latter (typically in Mill children,
uncivilised nations and, in some moods at least, the working classes). Mills
position rests on a wholly problematic distinction between the condition and the
development of autonomy which is resolved in practice by his robust, and
highly loaded, notions of moral progress.*
The same may be said of the World Bank. Landell-Mills makes it clear that
the Bank is only keen to build on the indigenous insofar as it is compatible with
modernization.86It remains to decipher the basis on which this judgement about
what is compatible with modernization is to be made. The Banks favoured
academic experts are commendably clear on this. Hyden argues that, as long as
the economy of affection is able to influence behaviour in the civil public realm
. . . governments in Africa are likely to remain paralysed . . . it is a problem with
roots in society. To that extent it is clear that improvements in government
performance are dependent on the transformation of society.87Exactly so and this
is at the root of the Banks idea of building civil society. This is not to consist
of ethnic or other affective or community groups, but contractual, non-
community, non-affective groups, such as professional associations, chambers of
commerce and industry, trade unions and N G O S . ~ It ~is a society made up of
such groupings which will support the technical reforms; indeed it is this type
of society which will demand these technical reforms. The neutral state will be
supported by a liberal public sphere. As Partha Chatterjee has suggested the rise
of modem civil society is linked to what he calls the narrative of capital which
replaces the narrative of community. Community with its ties of kinship and
affection sits, very uneasily, he argues, with the demands of capital for
efficiency and homogenised individ~als.8~ The transformation suggested

J-F. Bayart, Finishing with the idea of the Third World: the concept of political trajectory, in
J. Manor (ed.),Rethvlking Third World Politics (London, Longman, 1991).
8 ) Mendus, Toleration,passim.
J. S. Mill, On Liberty (Harmondsworth, Penguin, 1988).
85 Mendus, Toleration.

Landell-Mills, Governance. civil society and empowerment.

B7 G . Hyden, No Shorrcuts to Progress (Berkley, University of California Press, 1983) pp.8-22.
p. 106. emphasis added.
a See World Bank, Sub-Saharan Africa: from Crisis to Sustainable Growth, p.191; A Framework
/or Capacity Building; and Landell-Mills, Governance, civil society and empowerment in Sub-
Saharan Africa.
09 P. Chatterjee, Responseto Taylors Modes of civil society , Public Culture, 3 (1990).

Q Politlcal Siudics Association, 1994

explicitly by Hyden, and implicitly by the Bank, is the destruction of those
affective or community ties which hinder development.

New Political Economy and the Liberal Self

We have already suggested that at the heart of liberal conceptions is a free self
or agent: all along, it has been a matter of emancipating selves from the
constraints imposed by pre-established values and identities, and the more
recent emphasis on individuality simply reflects the extension of this tendency
every more deeply - and widely - into the constitution of the se1f.m Most often,
and especially in relation to economic life, the characteristics of that agent have
been assumed to be universal underneath the superficial variety of culturally
conditioned behaviour. Arguably indeed much liberal social science functions
to sustain liberal political belief.9 Such conceptions of universal economic
behaviour, have been extended to the political and social realms as illustrated
by the NPE approach to politics, an approach which has met with some
sympathy from the Bank. It is clear that the Bank holds views about the
universality of economic behaviour and has been influenced by a view of
African politics derived from NPE.This leads to policies designed to enable the
public, to demand and monitor good perf~rmance.~~ In short, for the Bank this
means policies such as increased accountability and transparency, to which
politicians and bureaucrats are assumed to respond in a rational way by
improving their performance.
It is however well-known that in the field of liberal political theory (if not
economic theory) the idea of a universal liberal self has come under such
sustained assault that some liberal thinkers have been prepared to concede that
the liberal self is a product of particular historical circumstan~es.~~ It is at least a
strong inference that the liberal self far from lying dormant, universally waiting
emancipation from sinister oppression, is in fact a construct of recent and
Western provenance.% It would be wrong to suggest that the Bank or its
thinkers have devoted great attention to these issues but it seems a strong
implication (that perhaps parallels the more formal theoretical debates) that the
Bank is beginning to make the connection between the kind of civil society that
it wishes to foster, and the kind of agent or self necessary for that society to
function. The thought is undoubtedly there: the benefits of a participatory
approach is not simply the immediate advantage of a project better tailored to
the clients needs, but also the impulse it gives to the long term process of
changing men tali tie^'.^^ The mentalities to be brought about by long-term
change will be devoid of affective or communal ties and will be free to associate
within modern social organizations. On a broader level the Western legal and
bureaucratic models, and of course the market economy, are premised upon the
9o R. Bruce Douglas and G . M. Mara, The search for a defensible good: the emerging dilemma of
liberalism,Douglass et al (eds), Liberulism and the Good.
91 So that for example if all peasants are reallyeconomically rational policies to impose markets

or whatever on them may be justified. For some useful remarks on this see Stephen K.White, The
Recent Work of Jurgen Habermas, p.21.
92 World Bank, Governance and Development, p.14.

93 See J. Rawls, Justice as fairness: political not metaphysical, and D. Gauthier The liberal
individual, both reprinted in Aveniri and De-Shaiit (eds), Communilarianismand Individualism.
91 C. Taylors, Sources of the Self(Cambridge, Cambridge University Press, 1989) is perhaps the
most important recent analysis.
p5 Landell-Mills, Governance,civil society and empowerment,emphasis added.

0 Political Studies Association, 1994

98 Governance, the World Bank and Liberal Theory

individual who has no other public ties than contractual ones he chooses for
himself. For all the talk of building upon the indigenous it is this image which
dominates World Bank discourse.

Engineering Civil Society: the World Bank and NGOs

Finally, this analysis throws light on the relationship between the World Bank
and NGOs. NGOs have shifted their orientation from aid to development and
they have become more intellectually and operationally sophisticated and more
inclined to educate their home constituency. They may have also differentiated
politically, with many identifying themselves as progressive and seeking to
escape the perceived constraints of charitable activity. On the surface, there-
fore, the effects of these trends have been to produce considerable hostility
towards the World Bank and the network of Western interests, agencies and
discourses of which it is part. During much of the 1980s, the NGO sector was
very critical of the World Bank, particularly over structural adjustment in
general and its impact on the poor.% Thus at a big NGO conference in Sep-
tember 1990 it was argued that, World Bank and I M F programmes must be
placed within the larger context of the dominant economic and political para-
digm. This neo-liberal model is based on coercion, strength, control and cul-
tural as~imilation.~~ The basis for the claims to a critical stance seems to be
threefold. The first two concern poverty and the validity or otherwise of particu-
lar development models but the third is much more overtly political and
concerns local, democratic and participatory strategies in development. This
element of hostility is however tempered by a desire to talk to the Bank and even
cooperate with it, though publicly at least relations appear to be guarded rather
than warm.98 For its part the Bank appears to be taking a much more positive
attitude towards NGOs and to be actively searching for further areas and means
of cooperation with them as both implementers of projects and sources of
advice on policy.99In turn, NGOs are becoming more sympathetic towards the
Bank as it withdraws from some of its more excessive pricist policies and focuses
once again on reducing poverty, and on promoting health, education and
infrastructure.Im The Bank has also recruited ex-NGO staff.
There are doubtless many aspects to this rapprochement but we suggest that
there are considerable common elements in their respective agendas. The radical
noises of NGOs about Western interests should not obscure their common
vision of what development means which is rooted in Western notions of the
state, civil society and the self. The most radical part of NGO discourse ( and
clearly an important part of their self-esteem) is their emphasis on grass roots
participation and their strong demand that this become part of the development
process, an emphasis with which the Bank is not unsympathetic. But this
terminology is always to be understood entirely within Western preconceptions.
Social justice also demands the eradication of all forms of discrimination,
* J. Clark, Democratizing Development the Role of Voluntary Organisations (London, Earthscan,
1990). ch.12.
y7 Development gap continuing the challenge: Report on the 1990 International NGO Forum p.2.

We have drawn here on the Position paper of the NGO Working Group on the World Bank
December 1989.
World Bank. How rhe World Bank norks wrth NGOs (Washington DC, World Bank. 1990),and
World Development Reporr 1991, pp. 1 3 S 6 .
A. Williams, A growing role for NGOs in development, Finonce und Development, 27 (1990).
I Polilia1 Studies Arronaiion. 1994
whether on grounds of race, creed, tribe or sex.l0l Although NGO literature is
replete with contradictions (demanding the protection of cultures on one page
and the elimination of cultural bias on the next, which itself reflects wider
debates amongst liberals about such matters) the agenda is known in advance -
progressive, grassroots organizations are simply to be mobilised around
it.Io2 For its part the Bank has a vision of economic order which is rooted in
standard liberal conceptions and is not simply the plaything of particular
Western interests (cf. the Banks consistent opposition to tied aid). Its more
recent analysis implies the need for a capacity to reach much deeper into Third
World societies and mould them more than has ever been contemplated at least
in modem times. NGOs provide a conduit of micro-interference that comple-
ments the macro-level interference at state level. The radical vision of NGOs is
not fundamentally different - it shares the Banks doubts about the capacities of
Third World governments, increasingly shares its stress on civil society and
entirely shares its lack of interest in and contempt for cultural traditions that do
not square with Western notions of rights and justice.

The construction of governance is based upon three levels of transformation: at
the institutional level the creation of a neutral state; at the social level the
creation of a liberal public sphere or civil society; and at the personal level the
corresponding creation of a liberal self and modem patterns of behaviour.
These three transformations were also at the heart of the gradual transforma-
tion to modernity undergone in Britain, for example during the seventeenth to
nineteenth centuries, a parallel on which the Bank itself draws. In Governance
and Development, the Bank looks to the example of the transformation of the
English institutional sphere in the seventeenth century which saw the creation of
the Bank of England, a re-organization of public finance, the securing of
property rights and patent laws, and a freeing up of the market.lo3Landell-Mills
uses the example of the transformation of the English civil service in the
nineteenth century from being highly corrupt and inefficient to being accoun-
table, meritocratic, honest and non-partisan. He argues that the powerful
message from this example is that reform of institutions was, underpinned by a
steady growth in public morality that led to the popular demands for public
accountability, and by the growth of a social conscience. In short the growth of a
liberal public space which demanded a transformation of the institutional
sphere.104 The Bank and Bank staff are then making connections in the first two of
the suggested transformations. It might be added that as to the third the work of
Foucault and others has cast light upon the mechanisms for the control of
populations, for regulation of new forms of conduct and the creation of new

J . Clark, Democrafising Developmenf.p.30.


The parallels with those involved in development work within capitalist societies is surely an
important topic for research. Bellah et at., Habits ofthe Heart (London. Hutchinson, 1988) note
that. for all the lip service given to respect for cultural differences, Americans seem to lack the
resources to think about the relationship between groups that are culturally, socially or economi-
cally quite different. p. 206.
lo World Bank. Governance and Development, p.7.

Landell-Mills, Governance.civil society and empowerment.


( Political Studies Association. 1994

100 Governance, the World Bank and Liberal Theory

forms of subjectivity during this period.lo5Neither the Bank nor we are suggest-
ing the tranformation in Africa would be exactly the same, or that there are the
same institutions involved. However, it does seem plausible to suggest the
construction of governance by the Bank is akin to the historical tranformation
of the now developed world. Another insight to be gained from this historical
transformation is how much of a part liberal thought played in justifying it.J6
Liberal thought and practice historically, and now in the form of governance,
when faced with differencecannot be neutral or tolerant but does indeed have
its own broad conception of the good, which it is engaged in imposing politi-
cally, legally, socially and culturally wherever it has the power to do ~0.107
A final set of questions is raised by this analysis. What drives this relentless
urge in Western culture to do good to others by reshaping them? The sixteenth
century Spanish invasion of the Americas produced amongst other things
intense intellectual debates within a Christian framework which posits a single
god and enjoins a universalist religion applicable to human beings equally.l~
The dispute pitted those who thought the natives too debased to be worth more
than conquest at best or death at worst against those, the humanists, who
thought that they too had souls and could be brought to accept the Christian
faith. The intellectual logic in Connollys acute reconstruction is very clear - the
premises of singularity and universality press against affirming a plurality of
gods appropriate to the other in the name of cultural pluralism, while the
premise of human equality makes it sinful to practice benign neglect or indiffer-
ence towards pagan beliefs. If innocent, these others must be converted; if
hopelessly corrupted they must be conquered or eliminated so that the corrup-
tion will not spread. Perhaps tolerance can be added to the list of possible
stances . . . but it must be, as tolerance usually is, a circumscribed and tactical
tolerance. Tolerance, in this context becomes forbearance toward cultural prac-
tices thought to be intrinsically wrong or inferior, but also thought to contain a
glimmer of truth that might evolve, with proper prodding into realization of
Christian truth.Iw Exactly. At some point the textual analysis of the theoretical
implausibility of the universal lawIIomust be supplemented by the political and
anthropological analysis of what those who imagine they possess such a law are
enabled to do to others as a result. To track governance to its real lair, it is this
logic that needs further investigation.

lo) M. Foucault, The History of Sexuality: an Introduction (Harmondsworth, Penguin, 1990);

Discipline and Punish (Harmondsworth, Penguin, 1991). See also S. Clegg, Framework of Power
(London. Sage, 1989). pp. 167-78.
See J. Tully, Governingconduct, in E. Leites (ed.), Conscience and Casuistry in Early Modern
Europe (Cambridge, Cambridge University Press, 1988).
Io7 A. MacIntyre, Whose Justice? Which Rationaliry? (London, Duckworth, 1988) p.336.

lm W. Connolly. IdentifylDiflerence (Cornell, Cornell University Press, 1991). p.42.

lop Connolly, IdentifyIDifferencc pp.42-3.
P. Singer. Practical Ethics (Cambridge, Cambridge University Press, 1979) ch.1.
Q Political Siudia A m m i o n . 19w
What Is Governance?

This commentary points to the poor state of empirical measures of the

quality of states, that is, executive branches and their bureaucracies. Much
of the problem is conceptual, as there is very little agreement on what
constitutes high-quality government. The commentary suggests four
approaches: (1) procedural measures, such as the Weberian criteria of
bureaucratic modernity; (2) capacity measures, which include both
resources and degree of professionalization; (3) output measures; and (4)
measures of bureaucratic autonomy. It rejects output measures and sug-
gests a two-dimensional framework of using capacity and autonomy as a
measure of executive branch quality. This framework explains the conun-
drum of why low-income countries are advised to reduce bureaucratic
autonomy while high-income ones seek to increase it.

This commentary is the beginning of an effort to better measure gov-

ernance, which at this point will amount to nothing more than an
elaboration of the issues complexity and the confused state of current
discussions. Before we can measure good governance, however, we
have to better conceptualize what it is.

The state, that is, the functioning of executive branches and their
bureaucracies, has received relatively little attention in contemporary
political science. Since the onset of the Third Wave of democratizations
now more than a generation ago, the overwhelming emphasis in com-
parative politics has been on democracy, transitions to democracy,
human rights, transitional justice, and the like. Studies of nondemo-
cratic countries focus on issues like authoritarian persistence, meaning
that the focus still remains the question of democracy in the long run
or democratic transition. In other words, everyone is interested in
studying political institutions that limit or check powerdemocratic
accountability and rule of lawbut very few people pay attention to
the institution that accumulates and uses power, the state.

*Stanford University

Governance: An International Journal of Policy, Administration, and Institutions, Vol. 26, No. 3,
July 2013 (pp. 347368).
2013 Wiley Periodicals, Inc.

The relative emphasis on checking institutions rather than power-

deploying institutions is evident in the governance measures that have
been developed in recent years. There are numerous measures of the
quality of democracy like the Freedom House and Polity measures, as
well as newer and very sophisticated ones like the Varieties of Democ-
racy project led by Michael Coppedge, John Gerring, et al. We have
fewer measures of Weberian bureaucracythat is, the degree to which
bureaucratic recruitment and promotion is merit based, functionally
organized, based on technical qualifications, etc. One of the only studies
to attempt to do this was by Peter Evans and James Rauch back in 2000,
but their sample was limited to 30 odd countries and produced no
time-series data. The Varieties of Democracy project is also collecting
data on bureaucratic quality based on expert surveys. Other bureau-
cratic quality measures include the Bertelsmann Transformation Index,
which focuses on how effectively policymakers facilitate and steer
development and transformation processes, and the proprietary
Political Risk Services Group International Country Risk Guide. Four of
the six World Bank Institutes Worldwide Governance Indicators
purport to measure aspects of state capacity (government effectiveness,
regulatory quality, political stability and absence of violence, and
control of corruption), but these are aggregates of other existing mea-
sures and it is not clear how they map onto the Weberian categories. For
example, does a good absence of violence score mean that there is
effective policing? I suspect that there isnt much by way of street crime
or military coup attempts in North Korea. (These problems are also true
of the Banks internal CPIA scores.) Finally, Bo Rothsteins Quality of
Governance Institute in Gothenberg has developed a set of measures of
quality of governance for 136 countries worldwide, as well as a more
detailed survey of 172 regions within the European Union. It is based
again on expert surveys focusing on the degree of a states impartiality,
which Rothstein argues is a proxy for overall state quality.

The bias against thinking about state capacity is particularly strong

among rational choice institutionalists. Most in this school begin with
Mancur Olsons assumption that states are predatory and that the chief
aim of political development is the creation of institutions like rule of
law and accountability that limit the states discretion. This school
assumes that all states have the power to be predatory, and seldom
raise the question of where state capacity comes from in the first place,
or how it increases or decreases over time. Frankly, it would be very
hard to develop a rational choice theory of state capacity, as capacity in
any organization is so heavily influenced by norms, organizational
culture, leadership, and other factors that do not easily fit into a model
based on economic incentives.

In addition, there has been a large literature on public sector reform

coming out of institutional economics, public administration, and from
the communities of practice surrounding development agencies
seeking to improve governance. The approaches favored by economists
sought to conceptualize governance in a principalagent framework,
and have sought to control corruption and bad administration through
manipulation of incentives. Many of the new approaches under this
framework sought to bring market-like incentives into the public
sector through creation of exit options, competition, manipulation of
wage scales, shortening of accountability routes, and better methods of
monitoring and accountability. Many of the techniques of New Public
Management was in some sense an outgrowth of these approaches,
though their applicability to developing world contexts has been ques-
tioned (Grindle 2004; Heady 1991; Pollitt and Bouckaert 2004; Schick
1998; World Bank 2004).

The existing measures of state quality or capacity have a number of

limitations. There is an inherent weakness in expert surveys, espe-
cially when trying to create time-series data. As the concept of good
governance is not well established, different experts may intend dif-
ferent things when responding to the same survey question. For
example, there is an important difference between clientelism and
outright corruption; in the former there is true reciprocity between
patron and client, whereas in the latter there is no obligation on the
part of the corrupt official to give anything back. The economic
impact of corruption varies tremendously depending on whether the
corruption tax is 10% or 50%, and the quality and nature of the
services that clients get in return. In China, for example, corruption
seems to be pervasive, but the tax rate is lower and the service pro-
vision rate much higher than in, say, sub-Saharan Africa. None of the
existing corruption surveys are, as far as I know, able to make dis-
tinctions of this sort.

Bo Rothstein makes a number of persuasive arguments that impar-

tiality ought to be the core measure of the quality of government.
However, it would seem entirely possible that a state could be
highly impartial and still lack the capacity and/or autonomy to
effectively deliver services. Rothstein argues that impartiality implies
the existence of sufficient capacity. This may be true, but it is some-
thing that needs to be empirically verified rather than simply

In addition, there are a number of Rule of Law measures that relate to

bureaucratic quality, such as those published by the ABA Rule of Law
Initiative and the World Justice Projects Rule of Law Index. Some

Chinese scholars have tried to measure the spread of rule-based

decision making by measuring the number of administrative cases
filed against government agencies, as well as the percentage of such
cases that are won by the plaintiffs.
The rule of law is defined differently by different scholars and can
mean, alternatively, law and order, property rights and contract
enforcement, observance of substantive Western norms of human
rights, and constitutional constraints on the power of the executive
(Kleinfeld 2006). Some scholars have distinguished between rule by
law, in which the executive uses law and bureaucracy as an instru-
ment of power, and rule of law, in which the executive is itself con-
strained by the same laws that apply to everyone else. In many
respects rule by law overlaps with state quality, as we want states to
operate by general, transparent, impartial, and predictable rules. Rule
of law in the narrow sense of constitutional constraints on the execu-
tive, on the other hand, is closely associated with democracy. The
Prussian/German Rechtsstaat in the 19th century, Meiji Japan, and
contemporary China were all authoritarian states that could be said to
have rule by law but not rule of law. This means that certain aspects
of rule of law would be useful as measures of state quality. On the
other hand, many rule of law measures measure what is measurable
rather than the underlying quality of law, so we would need to be
careful in selecting them.

As a starting point, I am going to define governance as a governments
ability to make and enforce rules, and to deliver services, regardless of
whether that government is democratic or not. I am more interested in
what Michael Mann labels infrastructural rather than despotic
power (Mann 1984). The reason I am excluding democratic account-
ability from the definition of governance is that we will later want to be
able to theorize the relationship between governance and democracy.
The current orthodoxy in the development community is that democ-
racy and good governance are mutually supportive. I would argue that
this is more of a theory than an empirically demonstrated fact, and that
we cannot empirically demonstrate the connection if we define one to
include the other.
In this initial conceptualization, the quality of governance is different
from the ends that governance is meant to fulfill. That is, governance
is about the performance of agents in carrying out the wishes
of principals, and not about the goals that principals set. The
government is an organization that can do its functions better or

worse; governance is thus about execution, or what has traditionally

fallen within the domain of public administration, as opposed to
politics or public policy.1 An authoritarian regime can be well
governed, just as a democracy can be maladministered. (As we
will see below, this distinction cannot always be maintained quite
so neatly; principals can set self-undermining tasks for their
As Rothstein (2011) points out, it is not so easy to separate governance
as implementation from the normative ends that government is meant
to serve. It is not clear that a well-governed state is one that has
ruthlessly efficient concentration camp guards as opposed to bribable
ones. On the other hand, once one starts to introduce substantive ends
as criteria for good government, it is hard to know where and when to
stop. As Rothstein points out, the existing Worldwide Governance
Indicators embed a number of normative policy preferences (e.g., less
rather than more regulation) that color the final results. Would we want
to argue that the U.S. military is a low-quality one because it does
things we disapprove of, say, invading Iraq?
Rothstein argues that use of the criterion of impartiality solves this
problem as it is both normative and embeds what most people under-
stand by good government. However, for reasons I will elaborate
below, I dont think that impartiality by itself is a sufficient metric.2 I
want to put the normative question to the side for the time being,
however, particularly because I am interested in developing measures
that will work for both authoritarian and democratic regimes. Focus-
ing on an extreme case like concentration camps should not distract us
from the fact that there are many valence issues like provision of
education, health, or public safety that are shared by virtually all gov-
ernments, in which a more instrumental view of quality of governance
will suffice.
If we accept this definition of the object we are trying to study, then
there are at least four broad approaches to evaluating the quality of
governance: procedural measures, input measures, output measures,
and measures of bureaucratic autonomy.

Procedural Measures
The most classic effort to define governance in terms of procedures was
Max Webers famous characterization of modern bureaucracy in
Economy and Society (Weber 1978, 220221). We continue to use the
term Weberian bureaucracy as an ideal type to which we hope
highly corrupt, neo-patrimonial states will eventually conform. It
might be useful to review Webers conditions here:

1. Bureaucrats are personally free and subject to authority only within

a defined area.
2. They are organized into a clearly defined hierarchy of offices.
3. Each office has a defined sphere of competence.
4. Offices are filled by free contractual relationship.
5. Candidates are selected on basis of technical qualifications.
6. Bureaucrats are remunerated by fixed salaries.
7. The office is treated as the sole occupation of the incumbent.
8. The office constitutes a career.
9. There is a separation between ownership and management.
10. Officials are subject to strict discipline and control.

Conditions 15 and 9 are probably at the core of what people think

of when they talk about modern bureaucracy: They clearly delin-
eate such an organization from the kinds of venal or patrimonial
office that existed in Europe under the Old Regime, or that exist in
contemporary neo-patrimonial developing countries today. However,
characteristics 6, 7, 8, and 10 are more problematic. Condition 6,
fixed salaries, is not compatible with the kinds of incentives often
offered bureaucrats under New Public Management. Conditions 7
and 8 are not true of many mid-level officials in contemporary
America, in both the public and private sectors. One could say that
the United States fails to live up to the Weberian ideal, but it does
not seem likely that the quality of bureaucracy in the United States
would improve if it were impossible for talented individuals from
the private sector or the academy to serve in government for periods
of time. And condition 10 is incompatible with civil service protec-
tion, which during the Progressive Era was seen as a hallmark of
the modern bureaucracy that was replacing the patronage system.
More importantly, condition 10 suggests that bureaucrats are simply
robotic agents whose only purpose is to do the bidding of principals.
The idea of bureaucratic autonomythe notion that bureaucrats
themselves can shape goals and define tasks independently of the
wishes of the principalsis not possible under the Weberian

Nonetheless, certain procedural measures would remain at the core of

any measure of quality of governance. One would want to know
whether bureaucrats are recruited and promoted on the basis of merit
or political patronage, what level of technical expertise they are
required to possess, and the overall level of formality in bureaucratic

Capacity Measures
The problem with all procedural definitions of bureaucracy is that the
procedures, however defined, may not actually correlate with the posi-
tive outcomes expected from governments. We assume that a Webe-
rian bureaucracy will produce better services than one that is highly
discretionary and patrimonial, yet there may be circumstances where
the latters lack of rules result in faster and better tailored responses.
Enforcement power is not part of Webers definition; it is possible to
have an impersonal, merit-based bureaucracy that nonetheless is
extremely poor at getting things done. To say that a bureaucrat is
selected on the basis of merit does not define merit, nor does it
explain whether the officials skills will be renewed in light of chang-
ing conditions or technology.
The most commonly used measure of capacity is extractive capacity,
measured in terms of tax extraction. Tax extraction measures capacity
in two ways: First, it takes capacity, however generated, in order to
extract taxes; second, successful tax extraction provides resources that
enable the government to operate in other domains. Tax extraction
rates can be measured both by the percentage of taxes to gross domes-
tic product, as well as by the nature of taxationthat is, whether it is
based on income or wealth, or indirect taxation (as income and wealth
taxes are much more difficult to extract than indirect taxes).
While tax extraction is a reasonable starting point for measuring capac-
ity, it has several important limitations:

1. There is a difference between extractive potential and actual extrac-

tion rates. Actual tax rates are set not just by extractive potential, but
by policy choices regarding the optimal rate and types of taxation.3
The United States proved it had the potential to extract significantly
higher levels of taxes during the two World Wars, because it had an
overriding national interest in doing so. The peacetime level reflects
normative preferences for the optimal size of government, which
may vary between countries of identical potential capacity.4
2. A given level of taxation does not necessarily translate into the
efficient use of tax revenues. Revenues can be wasted on poor
administration, unproductive transfers, or outright corruption.
Bureaucratic outputs are the result not just of resource inputs, but of
things like organizational culture. Judith Tendler has written about a
poor and underresourced state in Northeast Brazil that nonetheless
achieved very good governance outcomes (Tendler 1997).5
3. For many countries, government revenues are based on resource
rents or international transfers rather than domestic taxation. In
many countries such rents and transfers constitute the vast majority

of government revenues. One could argue that if taxation is going to

be used as a measure of state capacity, then resource rents ought to
be excluded.
Tax extraction rates are hardly the only possible measures of state
capacity. States perform a whole variety of functions, any one of which
can be used as a proxy for state capacity as a whole. Taxation is a useful
proxy for general capacity because it is a necessary function of all
states, and one for which considerable data exist. In an ongoing doc-
toral research, Melissa Lee and Nan Zhang have suggested using the
capacity to generate accurate census data as an alternative proxy for
capacity, as population registration is a very basic state function.
Beyond taxation, another critical measure of capacity is the level of
education and professionalization of government officials. Central
banks in the early 21st century across the developing world are incom-
parably better run than they were in the lead-up to the debt crises of
the 1980s in Latin America and sub-Saharan Africa, due in part to the
significantly higher degree of professionalism in their staffing. A key
aspect of state building in the United States during the Progressive Era
was the replacement of incompetent political patronage appointees
with university-trained agronomists, engineers, and economists.
A focus on the degree of professionalization of the bureaucracy par-
tially solves the problem of how to measure levels of corruption, a
measure that is not dependent on expert or perception surveys. All
professional education (with the possible exception of business
schools) embeds a strong normative element in which service to ones
profession and broader public goals is paramount. A doctor, for
example, is supposed to act primarily in the interests of the patient
rather than seeking first to maximize his or her individual benefit. Of
course, all professionals are also selfish individuals who can act in a
corrupt manner. But in modern organizations we trust highly edu-
cated professionals with a much higher degree of discretion because
we assume or hope that they will be guided by internal norms in cases
where their behavior cannot be monitored from the outside.
As state capacity varies substantially across functions, levels of govern-
ment, and regions, one would ideally want capacity measures for all
major government agencies. In Brazil, for example, it has been widely
recognized that certain islands of excellence exist within the Brazil-
ian state that would be missed by an aggregate measure. Thus, an
article by Katherine Bersch, Sergio Praa, and Matthew Taylor devel-
ops capacity measures across more than 300 different Brazilian federal
agencies (Bersch, Praa, and Taylor 2012). Obviously, this kind of data
does not exist for many countries, and even in Brazil the authors do

not have similar statistics for capacity at the state, local, and municipal
levels where a great deal of governance happens. For evaluating a
country like China, it would be very important to generate this type of
disaggregated data, as there is a widely held perception that the quality
of governance varies enormously across the different levels and func-
tions of government.
As a kind of compromise between an unachievable ideal of fully dis-
aggregated capacity data and a limited aggregate measure, it might be
possible to specify a subset of government functions on which data
should be collected. This could be a set of functions theoretically per-
formed by all governments (e.g., macroeconomic policy management,
basic law and order, primary and secondary education, population
registration), or it could incorporate data on how expansive the func-
tions performed are (e.g., giving extra credit if a government is able to,
say, regulate pharmaceuticals).

Output Measures

Good procedures and strong capacity are not ends in themselves. We

want governments to do things like provide schooling and public
health, public security, and national defense. This suggests an alterna-
tive measure of government quality, a measure of final output. One
could look at literacy, primary and secondary education test scores, or
various measures of health to get some idea as to how governments are
Attractive as output measures sound, there are several big and, in my
view, decisive drawbacks to their use. First and most important,
outputs like health or education are not simply the consequences of
public action; the public sector interacts with the environment around
it and the society it is dealing with to produce results. For example, the
Coleman Report on U.S. education in the 1960s showed that educa-
tional outcomes depended much more strongly on factors like friends
and family of students than they did on public sector inputs to edu-
cation (Coleman 1966). Joel Migdals model of weak states and strong
societies suggests that a governments ability to penetrate or regulate a
society depends on the ratio of two factors, state capacity and the
self-organization of the underlying society. Two states could have equal
regulatory capacity but unequal regulatory outcomes because the
society in one is far better organized to resist state penetration than the
other (Migdal 1988).
A second problem is that measuring output is itself problematic meth-
odologically. Public sectors produce primarily services, which can be

notoriously hard to measure. For example, standardized test scores, a

common way of evaluating educational outcomes, have long been
under attack as poor measures of education, and for creating incentives
to teach to the test. Measures of rule of law, like time to trial, rate of
case clearances, etc., say nothing about the quality of the justice being
produced by a legal system.
Finally, outcome measures cannot be so easily divorced from proce-
dural and normative measures. A police state may succeed in control-
ling street crime by massively arresting and torturing suspects, yet
most believers in liberal democracy would accept a higher degree of
crime in exchange for procedural protections of individual rights. Even
if one is morally neutral about whether torture is justified as a police
method, one would want to know whether it is employed routinely in
evaluating a government.
My own sense is that the problem of the tainting of output measures by
exogenous factors means that they should not be used as state quality
measures in the first place. One could employ a variety of econometric
techniques to control for these exogenous factors, but that entails
another layer of complexity and problems. In fact, it might be better to
leave output as a dependent variable to be explained by state quality,
rather than being a measure of capacity in itself. If output is not a valid
measure of state quality, it implies that we also cannot generate useful
measures of government efficiency as a measure of state quality, as the
latter represents a ratio of state inputs to outcomes.

Bureaucratic Autonomy

A final measure of the quality of government is the degree of bureau-

cratic autonomy possessed by the different components of the state.
Samuel Huntington makes autonomy one of his four criteria of
institutionalization; highly institutionalized political systems have
bureaucracies with high autonomy. The opposite of autonomy in Hun-
tingtons terminology is subordination (Huntington 2006).6
Autonomy, properly speaking, refers to the manner in which the politi-
cal principal issues mandates to the bureaucrats who act as its agent.
No bureaucracy has the authority to define its own mandates, regard-
less of whether the regime is democratic or authoritarian. But there are
a wide variety of ways in which mandates can be issued. Ideally, the
principal should set a broad mandate to the agent, for example, pro-
curement of an advanced strike fighter. But the principal can also issue
many other mandates as well regarding the way in which to carry out
the broad mandate, such as purchasing a strike fighter using contrac-

tors that increase employment in Congressional districts X and Y, or

through minority and women-owned businesses, or to achieve Z
degree of performance desired by a rival service. In other cases the
principal can issue mandates regarding the bureaucracys recruitment
and promotion of personnel, requiring that they hire certain individu-
als, or else setting detailed rules for personnel management.
Political principals often issue frequently overlapping and sometimes
downright contradictory mandates. Indeed, there can be multiple prin-
cipals in many political systems, that is, political authorities with equal
legitimacy able to issue potentially contradictory mandates. State-
owned utilities, for example, often have mandates to simultaneously
do cost recovery, universal service to the poor, and efficient pricing to
business clients, each promoted by a different part of the political
system. These different mandates obviously cannot be simultaneously
achieved, and generate bureaucratic dysfunctionality. Amtrak could
become a profitable and efficient railway if it were not under Congres-
sional mandates to serve various low-volume rural communities. In
China there are often duplicate functional agencies, one reporting
through a chain of command that goes through national ministries, the
other reporting to municipal or provincial governments; it is not
always clear how conflicts between them are to be resolved.
Autonomy therefore is inversely related to the number and nature of
the mandates issued by the principal. The fewer and more general the
mandates, the greater autonomy the bureaucracy possesses. A com-
pletely autonomous bureaucracy gets no mandates at all but sets its
own goals independently of the political principal. Conversely, a non-
autonomous or subordinated bureaucracy is micromanaged by the
principal, which establishes detailed rules that the agent must follow.
An appropriate degree of bureaucratic autonomy does not mean that
bureaucrats should be isolated from their societies or make decisions at
odds with citizen demands. Indeed, if the general mandate is to
provide high-quality services in health or education, the bureaucracy
would need considerable feedback and criticism from the citizens that
it is trying to serve. It also does not exclude extensive collaboration
with private sector or civil society organizations in service delivery.
Indeed, an appropriately autonomous bureaucracy should be able
to make judgment calls as to when and where to engage in such
It would seem that the relationship between autonomy and quality of
government would look like an inverted U (see Figure 1). At one
extreme, that of complete subordination, the bureaucracy has no room
for discretion or independent judgment, and is completely bound by

Bureaucratic Autonomy and Quality of Government

Government Quality

Bureaucratic Autonomy

detailed rules set by the political principal. At the other end of the
x-axis, that of complete autonomy, governance outcomes would also
be very bad, because the bureaucracy has escaped all political control
and sets not just internal procedures but its goals as well. This is
basically the idea contained in Peter Evans concept of embedded
autonomy: Bureaucrats need to be shielded from certain influences of
social actors, but also subordinate to the society with regard to larger
goals (Evans 1995).
There are myriad examples of excessive subordination leading to poor
performance. One of the worst forms is when bureaucracies lose
control over internal recruitment and promotion to the political
authorities and are staffed entirely by political appointees. This is in
effect what happens in clientelistic political systems. But even in the
absence of clientelism, bureaucracies can be excessively slow moving
and indecisive because they are excessively rule bound. However, the
curve in Figure 1 slopes downward at the left end of the x-axis, which
represents full autonomy. Both Imperial Germany and Japan in the
periods before World War I and World War II, respectively, suffered
from this problem. Both countries had developed very high quality,
autonomous bureaucracies, particularly their military services, which
then took over from the political authorities the task of formulating
foreign policy.
The inflection point of the curve in Figure 1 is shifted to the right,
however, due to a general recognition that the dangers of excessive
micromanagement are greater than those posed by excessive

autonomy. A high degree of autonomy is what permits innovation,

experimentation, and risk taking in a bureaucracy. In Daniel Carpen-
ters The Forging of Bureaucratic Autonomy, both the Post Office and the
U.S. Forest Service are portrayed as high-quality autonomous bureau-
cracies during the Progressive Era precisely because they innovated
and devised agendas not strictly spelled out by Congress (Carpenter
2001). This same insight is embedded in the evolution of the U.S.
Armys basic field manual for combined arms operations, FM 100-5. In
rethinking combined arms doctrine in light of the Vietnam War, the
drafters of the manual shifted emphasis from centralized command
and control to more flexible Mission Orders under which the com-
mander only set broad goals, and devolved implementation to the
lowest possible echelon of the command structure. The latter were in
other words agents who were permitted a high degree of autonomy,
which included toleration of failure if they sought to innovate or
experiment.7 More broadly, one could argue that modern private sector
organizations have evolved over time away from rigid Taylorite hier-
archies reflecting strict Weberian criteria to more flexible, flatter orga-
nizations that delegate far more authority to lower levels of the
If an appropriate degree of bureaucratic autonomy is an important
characteristic of high-quality government, then neither the Weberian
nor the principalagent models can stand intact as frameworks for
understanding how bureaucracies ought to work. The Weberian
model, as noted earlier, assumes that bureaucrats are essentially rule-
bound implementers of decisions made by political authorities; they
may have technical capacity but they do not have the authority to set
agendas independently. The principalagent framework is inadequate
as well because it, too, assumes that agents are simply tools of the
principals, whereas in a good bureaucracy authority often flows in the
reverse direction, from the agent to the principal (the latter point, basic
in an older tradition of public administration, is made by Herbert
Simon; Simon 1957).
How do we measure bureaucratic autonomy? I believe that this is one
of the most central and difficult issues in constructing a good measure
of government quality. The most common approach is to use expert
surveys in which experts are asked to evaluate the autonomy of a given
bureaucracy. Expert surveys are particularly problematic in this area
because the very concept of autonomy has been poorly specified, and
it is not clear exactly what it is that experts are being asked to judge. Do
they have adequate criteria for judging what proper and improper
mandates are, or to look for multiple or conflicting mandates as a
measure of subordination?

It would be nice therefore to have a more objective measure of

autonomy. As autonomy is the opposite of subordination, one could
use the degree of clientelism or political interference in bureaucratic
operations as a measure. One could look, for example, at the relative
number of classified versus political positions in a bureaucracy.
However, this gets at only one type of subordination related to per-
sonnel. Political principals can hamstring bureaucracies by issuing
multiple contradictory mandates that have nothing to do with staffing,
or by setting excessively detailed rules for bureaucratic behavior.

Capacity and Autonomy

It would seem to be the case that the quality of government is the result
of an interaction between capacity and autonomy. That is, more or less
autonomy can be a good or bad thing depending on how much under-
lying capacity a bureaucracy has. If an agency were full of incompetent,
self-dealing political appointees, one would want to limit their discre-
tion and subject them to clear rules. The assertion embedded in
Figure 1 that the optimal amount of autonomy is shifted to the right is
true only in high-capacity countries. In very low capacity countries, the
opposite would be the case: One would want to circumscribe the
behavior of government officials with more rather than fewer rules
because one could not trust them to exercise good judgment or refrain
from corrupt behavior. This is why Robert Klitgaard coined the
formula Corruption = Discretion - Accountability (Klitgaard 1988).
This is also why development agencies have been advising poor coun-
tries to limit bureaucratic discretion in recent years. On the other hand,
if the same agency were full of professionals with graduate degrees
from internationally recognized schools, one would not just feel safer
granting them considerable autonomy, but would actually want to
reduce rule boundedness in hopes of encouraging innovative behavior.
Figure 2 illustrates how the optimal autonomy curves would differ for
four hypothetical countries of differing levels of capacity. For each, the
curve slopes downward at the extremes, because every bureaucracy
can have too much or too little autonomy. But the lower-capacity
countries have their inflection points shifted to the left, while they are
shifted right for higher-capacity countries.
One can control the behavior of an agent either through explicit formal
rules and incentives or through informal norms and habits. Of the two,
the latter involves substantially lower transaction costs. Many profes-
sionals are basically self-regulated, due to the fact that (1) it is hard for
people outside their profession to judge the quality of their work and
(2) part of their education, as noted previously, consists of socialization

Optimal Levels of Autonomy for Differing Levels of Capacity

Government Quality
level 4

level 3

level 2
level 1

Bureaucratic Autonomy

to certain professional norms that seek to preclude certain types of

self-seeking behavior. The higher the capacity of a bureaucracy, then,
the more autonomy one would want to grant them. In judging the
quality of government, therefore, we would want to know about both
the capacity and the autonomy of the bureaucrats.
One would want, in other words, to be able to empirically locate the
agency on the matrix shown in Figure 3. The line sloping downward
and to the left represents the line drawn through the inflection points
of Figure 2, representing optimal levels of autonomy for a given level
of capacity. Bureaucracies that were to the left of the line would be
hobbled by excessive rules; those to the right of it with excessive
discretion. For the past decade, international donors have been advis-
ing developing countries to decrease the amount of discretion in the
behavior of their bureaucracies. From Figure 3 it would appear that
this is only contingently good advice; in a high-capacity state, one
would like to have more rather than less discretion.
The framework in Figure 3 suggests that there are two quite separate
approaches to public sector reform. One always wants to move up the
y-axis to higher capacity, particularly with regard to the professional-
ism of the public service. This, however, is not something that can be
done easily, and it is not something that can in any case be accom-
plished in a short period of time. If a country cannot significantly
upgrade capacity in the short run, one would want to shift the degree
of autonomy toward the sloping line. This would mean moving
toward the left in a low-capacity country, and toward the right in a

Autonomy and Capacity

High capacity
Sweet spot

too many rules excessive discretion

Subordination Autonomy

Low capacity

Reform Paths

High capacity

Germany Singapore

too many rules excessive discretion

Subordination Autonomy


Low capacity

high-capacity country. Figure 4 contains some hypothesized positions,

aggregated across the whole government, for different countries, and
suggests that while Nigeria and China need to move left, the United
States needs to move right. However, China needs to end up at a point
with significantly more autonomy than Nigeria because of its much
higher capacity.
Trying to locate India on this matrix demonstrates some of the com-
plexities of this analysis. India is famous both for high levels of cor-
ruption and clientelism, and for simultaneously having excessive rules
and bureaucratic red tape. India clearly needs much greater state
capacity across the board. But does it need more or less autonomy? The
answer to the latter question is probably both, dependent on specific
context. Given the recent scandal, the agency handling spectrum auc-
tions needs to be subjected to much stricter rules; on the other hand,
the Hyderabad Municipal Water Authority needs to be relieved of its
multiple and conflicting political mandates if it is to function properly.
This then suggests why devising single aggregated measures for
quality of governance can be inadequate and misleading.


It is clear that in evaluating the quality of governance in large, complex

countries like China or the United States, the existing quantitative
measures are woefully inadequate. If we are to establish desiderata for
better ones, we would have to answer the following questions:

If we are to use procedural measures of government quality, which

on Webers list do we want to keep?
For how many countries could we collect disaggregated capacity
If we cannot collect a full set of capacity measures, what are the best
proxies for aggregate capacity? Beyond tax extraction levels, can we
come up with measures of bureaucratic professionalization?
How do we distinguish between actual and potential capacity, with
regard to a commonly used measure like tax extraction?
How, exactly, are we going to define bureaucratic autonomy, and
what measures are available as proxies for it?
How important is it to have quantitative measures at all, as opposed
to qualitative descriptions of process, or else case studies of particu-
lar areas of governance?

This commentary does not pretend to answer these questions, but only
to serve as a basis for discussion. As we cannot measure what we

cannot adequately conceptualize, we have to start with the concept

first. I have laid out two separate dimensions of governance, capacity
and autonomy, and suggested some of the components that make
them up. Capacity, in particular, consists of both resources and the
degree of professionalization of bureaucratic staff. I have further
posited that quality of governance is ultimately a function of the inter-
action of capacity and autonomy, and that either one independently
will be inadequate as a measure of government quality. Finally, I have
also suggested that states need to be disaggregated into their compo-
nent parts, by function, region, and level of government, and that we
need both capacity and autonomy measures for all of these compo-
nents. Obviously, this volume of data does not exist for most countries,
and may not exist for any country. How much could be generated? It
might be useful to start with a large, relatively data-rich country like
the United States, and see how far we could get.


This commentary has been extensively revised based on presentations

and discussions generated by the Governance Project at Stanford Uni-
versitys Center on Democracy, Development, and the Rule of Law. I
would like to also thank the reviewers at the Center for Global Devel-
opment for their comments.

1. This distinction was made in Woodrow Wilsons famous article (Wilson
1887). It is also made in Max Webers equally famous essay, Politics as a
Vocation (Weber 1946).
2. One would have to say that a concentration camp guard who executed
everyone he was ordered to kill was more impartial than one who played
favorites or spared certain individuals in return for bribes or sexual favors.
This points to the difference between impartiality of policies compared to
the impartiality of the way in which policies are executed.
3. Mancur Olson and others in the rational choice tradition argue that states are
predatory and that all states will seek to tax at a maximal rate, subject only to
limitations on capacity, and the time discount rates of the sovereign. There is,
however, considerable historical evidence that this is not true, and that states
have deliberately taxed well below their theoretical capacities for a variety of
reasons (Fukuyama 2011, 303305; Olson 1993).
4. Marcus Kurtz in his forthcoming book on the state in Latin America makes
use of this distinction (Kurtz 2013).
5. On the general importance of organizational culture, see DiIulio (1994) and
Wilson (1989).
6. One could use the term accountability as the antonym for autonomy;
however, accountability has certain normative implications that subordina-
tion does not.

7. This field manual was based on the operational doctrine that had been
developed by the German army from the end of World War I through the
beginning of World War II; Mission Orders are an American version of
Aufstragstaktik (Fukuyama and Shulsky 1997).

Bersch, Katherine, Sergio Praa, and Matthey Taylor. 2012. An Archipelago of
Excellence? Autonomous State Capacity among Brazilian Federal Agencies.
Unpublished paper.
Carpenter, Daniel. 2001. The Forging of Bureaucratic Autonomy: Reputations, Net-
works, and Policy Innovation in Executive Agencies, 18621928. Princeton, NJ:
Princeton University Press.
Coleman, James S. 1966. Equality of Educational Opportunity. Washington, DC: U.S.
Department of Health, Education and Welfare.
DiIulio, John J. 1994. Principled Agents: The Cultural Bases of Behavior in a
Federal Government Bureaucracy. Journal of Public Administration Reseach and
Theory 4 (3): 277320.
Evans, Peter. 1995. Embedded Autonomy. Princeton, NJ: Princeton University Press.
Fukuyama, Francis. 2011. The Origins of Political Order: From Prehuman Times to the
French Revolution. New York: Farrar, Straus and Giroux.
Fukuyama, Francis, and Abram Shulsky. 1997. The Virtual Corporation and Army
Organization. Santa Monica, CA: Rand Corporation.
Grindle, Merilee S. 2004. Good Enough Governance: Poverty Reduction and
Reform in Developing Countries. Governance 17 (4): 525548.
Heady, Ferrel. 1991. Public Administration: A Comparative Perspective. New York:
Marcel Dekker.
Huntington, Samuel P. 2006. Political Order in Changing Societies. New Haven, CT:
Yale University Press.
Kaufmann, Daniel, and Aart Kraay. 2009. Governance Matters VIII: Aggregate and
Individual Governance Indicators, 19962008. Washington, DC: World Bank
Kleinfeld, Rachel. 2006. Competing Definitions of the Rule of Law. In Promoting
the Rule of Law Abroad: In Search of Knowledge, ed. T. Carothers. Washington, DC:
Carnegie Endowment, 3174.
Klitgaard, Robert. 1988. Controlling Corruption. Berkeley, CA: University of
Kurtz, Marcus. 2013. Latin American State Building in Comparative Perspective. New
York: Cambridge University Press.
Mann, Michael. 1984. The Autonomous Power of the State: Its Origins, Mecha-
nisms, and Results. European Journal of Sociology 25 (2): 185213.
Migdal, Joel S. 1988. Strong Societies and Weak States: State-Society Relations and State
Capabilities in the Third World. Princeton, NJ: Princeton University Press.
Olson, Mancur. 1993. Dictatorship, Democracy, and Development. American
Political Science Review 87 (9): 567576.
Pollitt, Christopher, and Geert Bouckaert. 2004. Public Management Reform: A Com-
parative Analysis. 2nd ed. New York: Oxford University Press.
Rothstein, Bo. 2011. The Quality of Government: Corruption, Social Trust, and Inequal-
ity in International Perspective. Chicago: University of Chicago Press.
Schick, Allen. 1998. Why Most Developing Countries Should Not Try New
Zealand Reforms. World Bank Research Observer 13 (8): 123131.
Simon, Herbert. 1957. Administrative Behavior: A Study of Decision-Making Processes
in Administrative Organization. New York: Free Press.
Tendler, Judith. 1997. Good Government in the Tropics. Baltimore, MD: Johns
Hopkins University Press.

Weber, Max. 1946. From Max Weber: Essays in Sociology. New York: Oxford Uni-
versity Press.
Weber, Max. 1978. Economy and Society. Berkeley, CA: University of California
Wilson, James Q. 1989. Bureaucracy: What Government Agencies Do and Why They Do
It. New York: Basic Books.
Wilson, Woodrow. 1887. The Study of Administration. Political Science Quarterly
2 (2): 197222.
World Bank. 2004. World Development Report 2004: Making Services Work for Poor
People. Washington, DC: World Bank.

Appendix: The Inadequacy of Existing Measures of Chinas Quality

of Government

If we accept the fact that quality of government is a mixture capacity and

autonomy, and that governments are themselves complex collections of
organizations, then it becomes clear that existing measures of governance
are highly inadequate. The Worldwide Governance Indicators produced
by the World Bank Institute (Kaufmann and Kraay 2009), as well as finer-
grained measures like Transparency Internationals Corruption Percep-
tion Index, treat single sovereign nations as the unit of analysis. Yet it is
obvious that the quality of governance varies enormously within coun-
tries, both by specific government function and by region. Moreover, one
cannot look at governance problems at one level only; many occur
because of interactions between levels of governments. A poor national
government can reduce the performance of a good local one, and vice

The problem of single country indicators is evident when we consider

something like Transparency Internationals Corruption Perception Index.
The 2011 index lists China as the 75th most corrupt country in the world.
It does a bit better than Brazil and Tunisia (both #73), it tied with Romania,
and it is just slightly better than Gambia and El Salvador. Yet this number
is virtually meaningless because it does not take account of the diversity of
outcomes within China. It is widely believed in China, for example, that
local governments there are much more corrupt than higher-level ones.
We do not in fact know whether this is true or not. Corruption varies not
just by level of government, but by region and by function; the railroad
ministry is very different from, say, the Central Bank.

There is also something very strange about the Worldwide Governance

Indicators rankings of China (see Table A1).

Chinas low rankings for Voice and Accountability and Rule of Law are not
surprising, given that no one argues either of these are Chinas strong suit.
The other four measures relate to what we are defining as governance.
While both the score and ranking for government effectiveness are higher

Chinas 2010 Performance, Worldwide Governance Indicators (Scores Range
from -2.5 [Weak] to 2.5 [Strong])
Category Score Percentile

Voice and accountability -1.6 5

Political stability/no violence -0.77 24
Government effectiveness 0.12 60
Regulatory quality -0.23 45
Rule of law -0.35 44
Control of corruption -0.60 33

than for any other measure, China still places only in the 60th percentile.
But what possible meaning can such a figure have? Clearly many local
Chinese government authorities have huge problems; on the other hand,
others perform far better. In my purely subjective estimation, the effec-
tiveness of Chinas national government with regard to macroeconomic
management of a hugely complex modernization process over the past
three decades has been nothing short of miraculous, given the fact that
China was not just managing an existing set of institutions, but also
transforming them in a more market-friendly direction. Its performance
since the Asian financial crisis has arguably been better than that of the
United States, which nonetheless ranks in the 90th percentile.

In terms of the three categories above, what do existing measures of

governance measure? In the case of the WBI Worldwide Governance Mea-
sures (WWGM), it is hard to say, because they are an aggregate of many
other measures. Many of them are perception surveys or expert estimates,
which often reflect output measures, but may also include evaluations of
procedures and capacity. It is not clear whether any of the WWGM com-
ponents explicitly seek to measure bureaucratic autonomy. Presumably
categories like Political Stability/Control of Violence are exclusively
output measures (where Chinas low 24th percentile ranking seems a bit
bizarre). The Rule of Law measure has big problems, beginning with the
lack of definition of what is being measured. If rule of law is defined as
constraints on the executive, China should rank even lower than it does as
there are no real legal constraints on the behavior of the Chinese Commu-
nist Party. If on the other hand this category means something more like
rule by law (which would make it a component of governance), the
ranking should be considerably higher. Most Rule of Law measures tend
to be related to procedures or capacity rather than output, because the
output of a legal system is so hard to measure. But we actually have no
idea what the Chinese numbers actually mean or purport to measure.

Francis Fukuyama is the Olivier Nomellini Senior Fellow at the

Freeman Spogli Institute for International Studies at Stanford Univer-
sity. Dr. Fukuyama has written widely on issues relating to democra-
tization and international political economy. His book, The End of
History and the Last Man, was published by Free Press in 1992. His most
recent book, The Origins of Political Order, was published by Farrar,
Straus, and Giroux in 2011. Other books include America at the Cross-
roads: Democracy, Power, and the Neoconservative Legacy, and Falling
Behind: Explaining the Development Gap between Latin America and the
United States.
Copyright of Governance is the property of Wiley-Blackwell and its content may not be copied or emailed to
multiple sites or posted to a listserv without the copyright holder's express written permission. However, users
may print, download, or email articles for individual use.
American Law & Economics
Association Annual Meetings
Year Paper

Economic Regulation and Social

Eric B. Rasmusen
Indiana University, Kelley Sch. Bus.

This working paper site is hosted by The Berkeley Electronic Press (bepress) and may not be
commercially reproduced without the publishers permission.
Copyright 2005
c by the author.
Economic Regulation and Social Regulation

3 April 2005

1social.tex, file 1

Eric Rasmusen


Health, safety, morals, and the general welfare are the traditional subjects of the
police power of the state. When we think of government regulation we usually think of
economic regulation. This is generally ecient only for a narrow range of activities and
is subject to abuse by private parties who can profit from it. Social regulation is another
area of government regulation, however, and one where the presumptive eciency of laissez
faire disappears, because market imperfections are more common and capture by special
interests is less profitable. This does not immediately imply that the government should
engage in social engineering, because our current knowledge of social processes is primitive.
It does imply that courts should be reluctant to strike down social regulation on the grounds
that it is irrational.

Indiana University Foundation Professor, Department of Business Economics and Public

Policy, Kelley School of Business,Indiana University, BU 456, 1309 E. 10th Street, Blooming-
ton, Indiana, 47405-1701. Oce: (812) 855-9219. Fax: 812- 855-3354.,

I thank Lucien Bebchuk, George Bentson, Robert Ellickson, and David Westfall and partic-
ipants in seminars at George Mason, Harvard, Michigan, Vanderbilt, and Yale Law Schools and
the Emory Department of Economics for comments. My thanks does not imply their agreement
with what I write. Footnotes starting with xxx are notes to myself for future drafts.

Hosted by The Berkeley Electronic Press


II.A. Standard Market Failure and Economic Regulation 1
II.A.1. Externalities . . .
II.A.2. Monopoly.
II.A.3. Imperfect Information
II.A.4. Multiple Equilibria
II.B. Government Failure and Economic Regulation III. SOCIAL REGULATION.
III.A. Social Regulation and Standard Market Failure
III.B. Special Reasons for Market Failure in Social Markets
III.B.1. Mental Externalities
III.B.1.a. Ideological
III.B.1.b. Sympathetic
III.B.1.c. Altruistic
III.B.1.d. Aesthetic
III.B.1.e. Private Sector Responses to Mental
III.B.2. Paternalism
III.B.2.a. Poor Judgement
III.B.2.b. Self-Control
III.B.3. Improving Tastes
III.B.3.a. Altruism
IV.A. Rent-Seeking
IV.B. Government Mistakes
IV.B.1. Social Ecology
IV.C. The Usefulness of Tradition
IV.D. Explaining Past Social and Economic Regulation

xxx Find out about the page label command in latex.


What you say is very fine indeed, Cephalus, I said. But as to this very thing,
justice, shall we so simply assert that it is the truth and giving back what a man
has taken from another, or is to do these very things sometimes just and sometimes
unjust? Take this case as an example of what I mean: everyone would surely say
that if a man takes weapons from a friend when the latter is of sound mind, and the
friend demands them back when he is mad, one shouldnt give back such things, and
the man who gave them back would not be just, and moreover, one should not be
willing to tell someone in this state the whole truth.
What you say is right, he said.
Then this isnt the definition of justice, speaking the truth and giving back what
one takes.2

Maintaining contracts and property rights- speaking the truth and giving back what one
takes are important functions of government. It is as appealing to modern Americans as to
ancient Greeks that people should be free to do what they want with their property, so long
as they respect other peoples property rights. There exists, at the same time, a strong feeling
that the economy would work better if the government intervened in it to control the prices
people charged when selling their property or the kinds of property they are allowed to sell, but
this feeling diminished sharply from the 1930s to the 1990s. After experiencing deregulation of
industries such as air travel, trucking, and telephone sales, both the educated and the uneducated
public seem to be less interested in government regulation of prices. The virtues of laissez faire
have come to be part of the conventional wisdom.

Not all life is monetized, even in America. Political issues are conventionally divided into
the Economic, the Social, and Foreign Aairs. Of Foreign Aairs we will say nothing, but in the
Social as in the Economic spheres, there is intense controversy over not only what the government
should make people do, but whether it should intervene at all. Table 1 lists a number of issues
in each sphere that have been controversial in America in the past decade, whether in political
campaigns or in the courts.


Plato, The Republic, translated by Allan Bloom, Basic Books, (1968) at 331c.

Hosted by The Berkeley Electronic Press
Social Issues Economic Issues

Abortion The minimum wage

Legalizing marijuana Outsourcing (protectionism)
Homosexuality National health insurance
Gun control Social Security privatization
Pornography Medicare drug benefits for the elderly
Gambling Living wage ordinances

Most of the issues here are not what is conventionally called regulation, but they all involve
using the threat of the power of the State to restrict peoples behavior, rather than simply to raise
money for public goods. The idea of social insurance whether it be national health insurance,
Social Security, or Medicare is to require people to buy insurance, and whether they are required
to buy it from the government or from private companies is a secondary issue.

It is not entirely clear how these are divided into the social and the economic, either. Mar-
ijuana sales are, after all, market transactions, so why not call the ban on marijuana economic
regulation? The reason, I think, is that although this is a monetized market, the aim of the
regulation is not to improve the workings of the market, but to eliminate marijuana, to remove
it from the economy rather than to enhance its role.

In America there is a curious dierence between liberals and conservatives in their positions
on the role of government. On social issues, conservatives tend to support heavier government
regulation. This is only a tendency, of course it is liberals who want stricter regulation of guns,
for example but it is a tendency nonetheless. On economic issues, liberals tend to support
heavier government regulation. Libertarians, who would like to restrict government impositions
on freedom in both the economic and social spheres, enjoy criticizing liberals and conservatives
alike for inconsistency.

Table 2 lays out these tendencies in a two-by-two grid.3 There are four possible combinations
of attitudes towards regulation. I have mentioned those occupied by conservatives, liberals, and
Neoconservatives are perhaps defined by their foreign policy attitudes than anything else, but they
have been an important libertarian strand in American thought. Classic liberals might belong either
with American conservatives or libertarians, since most Victorians (e.g., Macaulay, Gladstone, Disraeli)
favored free trade but not free love, but John Stuart Mill is an important exception, who led the way for
modern liberalism he opposed Sabbatarian legislation and prohibition of strong drink, bigamy, and opium;
John Stuart Mill, On Liberty, 194, 196, 198, 203 (1859), in Utilitarianism, Liberty, and Representative
Government, E.P. Dutton and Co., Inc., 1951.

libertarians already; what remains is the attitude that the government ought to regulate both
social and economic behavior. This is less prominent in America, but is common in the rest
of the world, on both left and right. I have put European conservatives and Marxists in
that box, two groups that are hostile to each other as to particular regulations but which agree
that people behavior ought to be restricted. In Anglo-American thought, this attitude might be
labelled Tory, with the idea that the slowing of change in all spheres of life is an important
function of government.


Social Regulation


FOR European conservatives (Tories) American liberals

Economic Regulation
American conservatives Libertarians
AGAINST Classic liberals (Whigs) Neoconservatives

Libertarians can claim human freedom as their highest good. Liberals put emphasis on a
variety of rights to decide behavior for oneself, most of which fall in the social sphere rather
than the economic. The European conservatives, Tories, and Marxists who support government
intervention broadly each have their own visions of the nations they are trying to construct. But
what of the Whigs? Is there a way to make their position consistent? We shall return to that
question later.

In both the economic and social spheres, deregulation has existed simultaneously with new
regulation. The general picture is one of increased tolerance, with libertarianism on the rise in
both the economic and social spheres among intellectuals, but with growing acceptance, also, of
government environmental, safety, and quality regulations. The success of the has been parallelled
by the success of the law- and- economics movement, many of whose adherents have a strong
tendency towards libertarianism. The economic approach to law has been applied in great detail to
economic regulation and deregulation, to the point where knowing some economics is indispensable
for anyone trying to do serious policy research in the area. It has been much less applied to social
regulation, although by 1992 it seems clear that in comparison with other developed countries,
the social problems of the United States are much more severe than the economic problems.

Is this simply because economics has nothing to say about nonmonetary problems? No.

Hosted by The Berkeley Electronic Press
The two driving ideas of economics are that of the rational maximizing individual, who chooses
actions depending on costs and benefits; and that of the competitive marketplace, which generates
predictable collective consequences from the independent actions of many individuals. These ideas
have been applied to many topics traditionally allocated to other disciplines such as political
science, law, sociology, and biology.4 The economic approach seems to explain behavior in a variety
of cultural and historical settings, and even the behavior of animals in laboratory experiments.5
Social interactions call for somewhat dierent tools of analysis than economic interactions, even
starting from the base of maximization and markets, but the rise of game theory and information
economics in the 1980s has vastly expanded the number of tools at our disposal, and we can
expect continued progress.

In the traditional interactions studied by economics, individually rational behavior tends

towards collectively optimal consequences. Smiths famed Invisible Hand guides property to be
traded until no one can benefit from future voluntary trades. At that point, almost by definition,
no intervention can make any one person better o without making someone else worse o.
Economists therefore favor the libertarian view that government regulation is unnecessary and
pernicious if society is to reach results that are optimal given the constraints of resources and
technology. People are often surprised by economists policy conclusions, because economists they
thought were liberal turn out to oppose the minimum wage and rent control, and economists they
thought were conservative turn out to favor legalization of marijuana and homosexuality. The
explanation is that in each case, the economist supports legalization of voluntary interactions
between consenting adults, which makes both parties to the transaction better o.

Government plays a role in recognizing and protecting property rights, but it decentralizes
authority by allocating those property rights to individuals. A property right is eectively the
right to take some action. If the horse, Dobbins, is allocated to A, A is allowed to take a number of
actions with respect to Dobbins to ride him, drown him, eat him, or sell him but B cannot do
any of those things without As permission. In the economic world of perfect competition, every
action is under the control of someone; there are no actions which a person cannot undertake if
he is willing to pay someone else enough to do so. In addition, the action is an alienable property
right; A can either ride Dobbins himself or sell that right to B. A government regulation or a
Gary Becker is the best known imperialist, e.g., Gary Becker(1974), A Theory of Social Interactions,
Journal of Political Economy, 82: 1063-1093 (November/December 1974); Gary Becker(1981), A Treatise
on the Family, Harvard University Press, 1981. He has been joined by many other scholars using rational
choice analysis in: Law, Richard Posner(1998) Economic Analysis of Law, 802 pp., 5th edition (1st edition,
1972), Aspen Law and Business, 1998; Sociology, James Coleman, Foundations of Social Theory (1990);
Political science, Peter Ordeshook, Game Theory and Political Theory: An Introduction (1986); Biology,
John Maynard-Smith(1982) Evolution and the Theory of Games, 224 pp., Cambridge University Press,
Raymond Battalio, John Kagel, Howard Rachlin & Leonard Green (1981) Commodity-Choice Behav-
ior with Pigeons as Subjects, Journal of Political Economy, 89:67-91 (February 1981).

social custom that forbade the riding of Dobbins by anyone, or made the right inalienable by
allowing only A to ride him would reduce welfare, because nobody would get the benefit of the
ride, or the benefit would go to A rather than to someone else that valued it more.

Economists are like Cephalus in The Republic: they view the role of government as to define
and protect property rights and to make people keep their promises. But Socrates dismisses
Cephalus and the economists definition of justice, only seven pages into the three hundred and
three pages of The Republic. He does not sco; this is the first definition of justice, and Cephalus
is the paradigmatic solid citizen, the pillar of society. Yet something is missing. This article will
try to determine whether given the assumptions of the economic approach the government does
serve any purpose other than to set property rights and settle contract disputes.

To these two functions, economists, and, I expect, Cephalus, would add the provision of
public goods such as national defense. Since the provision of public goods is independent of
regulation, in the sense that it imposes no constraints on individual behavior beyond requiring
that someone pay taxes, I will not discuss it further, except to briefly distinguish taxation from
regulation. Taxation is the use of government force to acquire funds with which to buy things.
Regulation is the use of government force to change peoples behavior. In both cases, governmen-
tion force is used, but the purposes dier. A pollution tax is really regulation, since its primary
purpose is to change behavior and only incidentally does it raise revenue. A requirement that
private employers provide health insurance is a tax since the end there is to buy things for people
that the government could pay for itself. This distinction has even been recognized in law, as the
celebrated Nollan case illustrates (Nollan v. California Coastal Com., 483 U.S. 825 (1987)). The
State of California would not let Mr. Nollan build a house unless he built a public path. The
U.S. Supreme Court ruled that this was not regulation, since the path did not solve any problem
created by building the house, but simply a tax, and a tax specific to the individual. This made
Californias requirement a taking without compensation. Since such takings are forbidden in the
U.S. Constitution, California was required to allow Nollan to build or else compensate him for
not building. In this paper, we will be concerned with regulation, not taxation.

I will argue that while the method of economics is useful for analyzing human interactions be-
yond the traditional economic markets, the default policy recommendation of laissez faire becomes
dubious if one goes too far afield. The old recommendation does not match local circumstances in
the newer provinces of economics empire. The analyst can still usefully assume that the individ-
ual acts rationally, but he cannot deduce that this leads to eciency in the absence of government
regulation and social custom. The positive predictions of economics will still be valid, but not
the usual normative conclusions. A presumption of the optimality of laissez faire is not a bad
thing for economic regulation, but for social regulation a better presumption is that the status
quo should be left untouched.6
The reader may be annoyed that I have not defined what is social and what is economic. A definition
may be very exact, and yet go but a very little way towards informing us of the nature of the thing defined;

Hosted by The Berkeley Electronic Press
This conclusion is important to the law because it implies that judges should not invali-
date statutes merely because they do not make sense on purely economic grounds. Judges who
pronounce on the reasonableness or rational basis of statutes and private covenants are too
self-confident, and they should be more deferential to the status quo of customary law and prac-
tice. The burden should be on those opposing the rules to show that they have a bad purpose,
not on those who support them to show that they have a good purpose. The argument will not
be based on making tradeos between standard economic goals and other values, as in discussions
that trade o eciency and equality. Here, eciency in the economists sense will be the only
goal, so the conclusions will be based on the utilitarian premises standard in economics, not on
religious or liberal premises. This requires some preliminary discussion to clarify societys objec-
tives. The objective, value maximization, will be to maximize utility by making the appropriate
tradeos. This is dierent from trying to reconcile absolute rules that conflict with each other or
from trying to discover the will of God. My hope is that by starting from the generally accepted
premise that it is better to make people happier, the analyst can come to useful conclusions about
how the government should restrict behavior. If it turns out that the resulting government regu-
lation oends an ethical or religious principle of the analyst, the analysis is nonetheless useful, in
showing that the principle cannot be imposed without reducing human happiness.

Even in a utilitarian framework, the individuals under analysis may have religious and ideo-
logical beliefs, and these will aect policy under value maximization. If some people believe that
God forbids people to work on Sunday or others believe that Kant forbids people to ban pornog-
raphy, the feelings of those people must be taken into account. The analysis will neither ignore
such beliefs nor take them as final: rather, they are data on utility to be fed into the utilitarian
calculus. If enough people are enough upset by Sunday work or pornography bans, those things
will be restricted.7 In evaluating both economic and social regulation, the objective will be to
examine taboos as means to increasing human happiness rather than as ends in themselves; the
but let the virtue of a definition be what it will, in the order of things, it seems rather to follow than to
precede our inquiry, of which it ought to be considered as the result. Edmund Burke, On the Sublime and
the Beautiful, p. 13. It is hard to define the boundary between the economic and the social, economics
and sociology, but in everyday language we seem to see a dierence. The economic sphere involves what is
private to the individual or an organization that produces goods for sale; the social sphere is what is left.
In law and economics, this approach is best known from the work of Richard Posner. See Michael Mc-
Connell & Richard Posner (1989) An Economic Approach to the Issues of Religious Freedom, University
of Chicago Law Review, 56: 1-xxx, Posner, Richard (1992) Sex and Reason, 458 pp., Harvard University
Press, 1992. I will not be addressing the concerns of those who object to universal commodification of
relationships or incommensurability of values. See Joseph Cropsey (1955) What is Welfare Economics,
Ethics, 65: 116-125 (xxx 1955). Reprinted in Political Philosophy and the Issues of Politics, University of
Chicago Press, 1977; or, for an overview, Matthew Adler, Law and Incommensurability: Introduction,
University of Pennsylvania Law Review, 146: 1169- xxx (xxx 1998). The utilitarian framework leaves
considerable room for analysis and is less liable to the criticism of assuming its conclusions than many
other approaches, as is evidenced by the dierences between Posners conclusions and my own.

Sabbath will be for man, not man for the Sabbath.8

The thrust of this book will be that laissez faire is the presumptive best policy in the economic
sphere, with, of course, many exceptions and special cases, but tradition is the presumptive best
policy in the social sphere. This is a modernized version of the argument of the English jurist
James Fitzjames Stephen.9 In Liberty, Equality, Fraternity, Stephen admires John Stuart Mills
Principles of Political Economy, with its careful explanations of the workings of the Invisible
Hand in economic markets, but he attacks Mills On Liberty as a confused muddling of utilitarian
principles. Mill, Stephen says, is willing to create human unhappiness merely for the sake of
abstract ideals (liberty, equality, and fraternity) that benefit no one and lack a rational basis.10
This was a dispute between two Liberals, each claiming to start from the same premises of
maximization of human happiness as the goal of society. This article will start from the same
premises, with the dierence that it will update the discussion using twentieth century economic

We thus will end up providing a justification for the attitude of the Whigs in Table 2,
who tend to favor social regulation and oppose economic regulation. The justification will be
a utilitarian one: that these tendencies will, if properly used, result in the greatest amount of
human happiness.

Here is an outline of how we will proceed. Chapter 2 sets out what economists mean by
eciency. Chapter 3 shows why laissez faire is good economic policy, and then sets out the usual
special cases where economic regulation is helpful. Chapter 4, the heart of the book, begins to
discuss laissez faire and regulation in the social context, pointing out the important role of mental
externalities. Chapters 6, 7, and 8 look at other reasons for regulation that apply more in the
social than in the economic context poor information, market power, and multiple equilibria.
Even if markets fail, governments fail too, and Chapter 9 looks at government failure in the
context of social regulation. Chapter 10 is about self-regulated communities, and discusses the
role of private norms. Chapter 11 concludes, with comments on the growing importance of getting
social regulation right, as opposed to economic regulation.
The intellectual ancestor of this approach is David Hume: But there is this material dierence between
superstition and justice, that the former is frivolous, useless, and burdensome; the latter is absolutely
requisite to the well- being of mankind and existence of society.... Were the interests of society nowise
concerned, it is as unintelligible why anothers articulating certain sounds, implying consent, should change
the nature of my actions with respect to a particular object, as why the reciting of a liturgy by a priest,
in a certain habit and posture, should dedicate a heap of brick and timber and render it thenceforth and
forever sacred. David Hume (1751) An Inquiry Concerning the Principles of Morals, p. 29, Bobbs- Merrill
Co. Inc. 1957.
James Stephen, A History of the Criminal Law of England (1883)
James Stephen, Liberty, Equality, Fraternity, 2nd edition (1874), page numbers from Chicago Univer-
sity Press edition, Richard Posner, ed.(1992).

Hosted by The Berkeley Electronic Press
The argument, put simply, will be that: (i) laissez faire works well in economic interactions,
with certain exceptions, and regulation is apt to be misused by the government, (ii) in social
interactions, the exceptions dominate, the case for laissez faire is much less compelling, and (iii)
government ocials face less temptation to use social regulation contrary to the public good, but
since it is dicult to predict the eects of social innovation, the best strategy is to maintain the
status quo.

2 Value Maximization

How should we decide whether a regulation is good, or bad? The first step is to choose
some valuation rule. Suppose, for example, that we are trying to decide whether a rule requiring
the arsenic level in drinking water to be less than 23 parts per billion is a good rule or not. A
strident environmentalist might say that the more stringent the rule, the better that a level of
23 is better than 30, but 4 would be an even better rule. Someone else might say that cost should
be considered too, and that reducing the level to 4 parts per billion would cost more than the
entire budget of the city government, requiring taxes to double.

The standard valuation rule used by economists is what I shall call value maximization,
but which is also variously called cost-benefit analysis, eciency, and wealth maximization.
The idea is simple. Add up how much each person who likes the regulation would pay to have it,
and subtract out how much each person who dislikes the regulation would require to be paid to
accept it. If the resulting number is positive, adopt the regulation.

A concrete example will illustrate this. Suppose Anderson and Brown want a stricter arsenic
regulation and would pay up to $30 and $70 to get it, whereas Corman and Daniels dont want
it, and would require payments of at least $20 and $10 to feel that the new regulation had not
hurt them. Since supporters would pay up to $100, and opponents would accept as little as $60,
adopting the regulation does maximize value.

This is not at all controversial if the payments actually take place. If we adopt the regulation
and make Anderson and Brown each pay $25 to Corman and Daniels, everybody is happy that
the deal went through. We call this a Pareto improvement, after the economist Pareto who came
up with the criterion that if everybody is better o, a policy is good.

What is more controversial is what whether we should say that the regulation is desirable
even if the new policy does not include payments to Corman and Daniels. The regulation still
maximizes value, but Corman and Daniels would oppose it. The standard position of economists
is that the regulation is nonetheless desirable, because the winners win more than the losers lose.
Absent other considerations, this surely makes sense. The alternative is to make Anderson and
Brown unhappy and more unhappy than Corman and Daniels would have been by rejecting the

2.1 A Single Market Transaction

Let us now apply the idea of value maximization to a market transaction. Smith approaches
Jones and asks if Jones will sell a bottle of whisky for $10. Jones agrees, and the whisky changes
hands. Is the transfer of the bottle from Jones to Smith a good thing? Value maximization says
that it is. Since Smith oered a price of $10, we know that his willingness to pay was at least
that high, and probably higher maybe $15. Since Jones accepted the price of $10, we know that
his value for the bottle was no more than that maybe it was $8. Using the figures of $15 and
$8, the net benefit from Jones giving the bottle to Smith is $7, an increase in value. It may be
that the true gain in value is more than $7 or less, but since both parties agreed to the trade, we
know that Joness value must be less than Smiths.

I will be calling this valuation rule value maximization, but we can use this example to
illustrate why the other terms are used for it. Suppose the values of $15 and $8 are correct, and
moving the bottle from Jones to Smith at a price of $10 has a benefit of $5 for Smith and $2 for
Jones. The eect on their satisfaction is then the same as if the trade were blocked, but by a
miracle $5 suddenly appeared in Smiths pocket, and $2 in Joness.11 This is why some people
call the idea wealth maximization: a value-maximizing trade increases the dollar amount at
which people value their possessions. It is also why some people call this an ecient allocation
of goods: by moving the bottle from the hands of Jones to those of Smith, it is as if the economy
had found a technology that increase the amount of output by $7.

It may seem obvious that it is good for Jones to sell the bottle of whisky to Smith. But value
maximization has an implication which is more troubling. Suppose Smith simply stole the bottle
from Jones. Value maximization says that this, too, is a good thing. The eect on total value is
exactly the same as the sale at the price of $10. The sale benefited Smith by $5 and Jones by
$2, a total of $7. The theft benefited Smith by $15 and hurt Jones by $8, which also makes for a
total benefit of $7. Value maximization treats these the same. All that matters is that the bottle
has moved from someone who values it less to someone who values it more.

This is an example of the power of the idea of value maximization, but also of its moral
neutrality. Economists do not take a stand on morality. To do so would not be controversial in
the case of theft, perhaps, but it would be in the case of most government policies. Is it moral to
tax rich people at a much higher rate than poor people? Is it moral to make consumers pay more
for sugar to benefit sugar producers? Is it moral to forbid racial discrimination or to require it, as
in the case of armative action or old-fashioned segregation? Those are important questions, but
the economist evades them. This evasion, moreover, is entirely reasonable. It is hard enough to
determine whether a policy maximizes value or not. We therefore separate that question from the
question of whether a policy is moral, which usually requires much dierent reasoning to answer.
I am ignoring wealth eects here; actually, adding $5 to Smiths wealth might have a minutely dierent
eect than giving him $5 in satisfaction from a bottle of whisky. True?xxx

Hosted by The Berkeley Electronic Press
In the case of theft, however, value maximization does provide a reason for why the action is
bad. In the particular case of Smith and Jones, value maximization says that the theft is good. It
is good, however, only because we started with a story in which Smith was willing to pay at least
$10 and Jones was willing to accept as little as $10. This information is crucial to whether moving
the bottle from Jones to Smith maximizes value. If, instead, we announced that we were going to
give the bottle to whoever valued it more, and asked each person for their values, imagine what
would happen. Smith might say he was willing to pay his entire wealth for it, and Jones might
say that he, too, valued that bottle more than anything else in the world. They would lie.

In actual practice, the idea of value maximization crucially requires us to know the value of
each person aected. The wonderful thing about market transactions is that market prices force
people to reveal something about their true values. Smith puts his money where his mouth is
when he oers $10. Jones reveals something about his value too, when he accepts $10. We do
not learn the exact values, but we know there is a net gain in social value.

After a theft, we do not know that value has increased. We do learn something that Smith
was willing to go to the trouble of stealing the bottle, and Jones was not willing to go to the
trouble of guarding it eectively but that is not enough to guarantee value maximization. And
in any case, Smiths stressful thieving and Joness worry and precautions are a social waste. If
Smith would have paid $4 to have gotten the bottle legally, and Jones invested $2 of his time in
trying to protect it, those $6 are a social waste, because transferring the bottle by open purchase
would have avoided them. Thus, theft, while it may be good in particular cases, is bad generally,
and it maximizes value to make it illegal.

I said that the idea of value maximization was powerful. One illustration is that we have
just used it to derive a reason why theft is bad, rather than having to accept the evil of theft
as an independent moral rule. If we command, Thou shalt value maximize, we can dispense
with Thou shalt not steal, Thou shalt not kill, and Thou shalt not commit adultery, which
become mere corollaries. Indeed, this is something like the rule of Jesus Christ (and Rabbi Hillel):
Thou shalt treat thy neighbor as thyself, a rule not unlike value maximization in its unbiased
treatment of everyone in society.

Another illustration of the power of value maximization is in how little information it requires.
Return to Smith and Jones, and suppose that Smith already owns 900 bottles of whisky, whereas
Jones only owns the one bottle for which Smith oers $10. How does that aect our opinion as
to whether the sale of the bottle is good or bad?

Presented with these facts, some people might say that the sale was bad. Smith, after all,
has plenty of whisky already. He does not need another bottle. Jones will be left whiskyless, and
we ought not to allow that to happen to him, poor fellow. Under this view, though, we really need
more information. If, in addition, we discovered that Jones had 2,000 bottles of gin, our decision
might reverse; Jones has plenty of liquor. If we then discovered that Smith was a millionaire, our

decision might reverse again; Smith can buy as many bottles as he wants, from other people. If
we then discovered that Jones only had one bottle of whisky because he had drunk the rest of
his stock and was too lazy to go and get more, our decision might reverse yet again; he is a lazy
pleasure-seeker, and does not deserve his remaining bottle.

Value maximization does not require any of that information. All it needs is that Smith
is willing to pay more than Jones for the bottle. It does not matter if Smith has more bottles
already he values that last bottle more than Jones. It does not matter if Jones has lots of gin
he does put a positive value on that whisky, and it should not be taken from him . It does not
matter if Smith is rich Jones still benefits from the transaction, and is happy to give up his lone
bottle for mere cash. It does not matter if Jones is lazy if we want to punish him for laziness,
do it with jail or a fine, not by preventing the innocent Smith from getting a bottle of whisky.

2.2 An Entire Market

The first step to understanding why free markets maximize value is to understand why the
transaction between Smith and Jones maximizes value. The second step is to look at how the
market price is chosen. Let us now look at the entire market for whisky, consisting of 300,000
potential buyers and 5,000 potential sellers, each of whom might sell 100 bottles. Buyers vary in
their willingness to pay from those who would pay at most $.01 per bottle to those who would
pay as much as $30. Sellers vary in their minimum acceptable price from as low as $4 to as high
as $20. The supply and demand diagrams in Figure 1 show this more precisely, as do the supply
and demand equations, P = 30 0.1Qd and P = 4 + 0.03Qs (with quantities of bottles measured
in thousands, as in the figure). The demand curve shows that there are 100,000 buyers willing to
pay at least $20 and 200,000 willing to pay at least $10.

(Let us assume that each buyer only wants one bottle.) The supply curve shows that there
are 1,000 sellers willing to take as little as $7 and 2,000 willing to take as little as $10. Since
each seller sells 100 bottles, this means that at the price of $10, the quantity supplied equals the
quantity demanded. This is the equilibrium price: the price generated by market forces.

Hosted by The Berkeley Electronic Press
Figure 1: Consumer and Producer Surplus

Why do market forces generate a price of $10? Think about what would happen if the price
were higher say, $20 per bottle. Sellers would be delighted to sell everything they had, but only
100,000 buyers would be willing to buy, so some sellers would end up unable to sell. Those buyers
would shave the price to $19.99, causing buyers to switch to them. That would leave other sellers
customerless, and those sellers would shave the price to $19.98. The price of $20 per bottle is thus
unstable, and the same reasoning shows that any price above $10.00 is unstable. At the price of
$10.00, however, each seller who is willing to sell find a buyer, and there is no incentive to cut
price further. A similar argument shows that any price below $10.00 is unstable. At any price
below $10.00, buyers are more eager to buy than sellers are to sell, and some buyers would be
unable to find a seller willing to sell. Those buyers would bid up the price, leaving other buyers
stranded, until the price was bid up to $10.00.

This argument shows that the free market equilibrium price is stable, but it says nothing
about whether it is optimal. It is, as it turns out, but to show that we need to think about the
costs and benefits to buyers and sellers. First, let us calculate them for the equilibrium price of
$10.00 and quantity of 200,000 bottles, bought and sold by the buyers and sellers at the left of

Figure 1, the buyers with the highest benefits and the sellers with the lowest values. You will
see immediately that whatever quantity is optimal, it ought to be bought by those high-valuing
buyers and sold by those low-valuing sellers. This is the result in the free market, but it is worth
mentioning because it might not be the result under government regulation, which might, for
example, result in the bottles going to the buyers who are most morally deserving or the best
connected politically.

Think about the benefit to the sellers who are selling those 200,000 bottles (some sellers are
inactive, selling nothing, so their benefit is zero). The gross benefit is their sales revenue, which
is simply (200,000 bottles) ($10/bottle) = $2,000,000. This is not their net benefit, because the
sellers did value those bottles, even if their values were not as high as the buyers. Their values
ranged from $4/bottle to $10/bottle, the height of the supply curve. Combined, their values
are the area labelled Seller Cost in Figure 1, since for typical sellers the value they place on
what they are selling is their acquisition or production cost (though the cost could be a dierent
opportunity cost that they cannot drink the whisky themselves!).

We can numerically calculate the size of the seller cost. Geometrically, it is the area of
the rectangle $4/bottle high and 200,000 bottles wide (which is $$800,000) plus the area of
the triangle with a height of ($10/bottle - $4/bottle) and a width of 200,000 bottles, which is
(1/2)($6/bottle)(200,000 bottles) = $600, 000. That sums to a seller cost of $1,400,000.

Since the sellers net benefit is their gross benefit (revenue) of $2,000,000 minus their lost
value (cost) of $1,400,000, their net benefit is $600,000. This is the area labelled producer
surplus in Figure 1, which is the standard name economists use for net seller benefit.12

Now think about the buyers. The gross benefit the active buyers get from the 200,000 bottles
is the sum of the values for each of the buyers. Some buyers have a value of $30, some of $29,
some of $28, and so forth down to the last active buyer, whose value is only $10. (There exist
other buyers who are inactive, but they will not be getting any benefit, so we can ignore them.)
The sum of the values is the area under the demand curve up to 200,000 bottles. This equals
the area of the rectangle $10/bottle high and 200,000 bottles wide, which is $2,000,000, plus the
area of the triangle above it with height ($30/bottle - $10/bottle) and width 200,000 bottles,
which is (1/2) ($20/bottle) (200,000 bottles)= $2,000,000. Adding up the two areas (which just
coincidentally have the same size) gives us the gross benefit of the buyers, their aggregate value
for the bottles, which is $4,000,000.

The net value for the buyers is less than the gross value, because they have to pay the sellers
$10/bottle. This is a payment of $2,000,000 for all 200,000 bottles, so the net value is $4,000,000
Readers who have studied economics have probably learned the concept of producer surplus a dierent
way, calculating the area of the producer surplus directly rather than as the revenue rectangle minus the
cost trapezoid. I use the slower method because soon we will look at atypical cases where the producer
surplus is not a simple triangle.

Hosted by The Berkeley Electronic Press
- $2,000,000= $2,000,000. In Figure 1, this is the area labelled consumer surplus.

Adding together the producer surplus and the consumer surplus gives us the total surplus,
the total value created by the existence of this market. When the quantity is 200,000 bottles and
the price is $10/bottle, the total surplus is thus $600,000 + $2,000,000=$2, 600,000.

Having calculated the value created by the free market, we must now show that this is
more than any government regulation could create. The first step is to understand a remarkable
fact: that it is the quantity of 200,000 bottles that determines the total surplus, not the price of
$10/bottle. To see this, suppose that we keep the quantity at 200,000, being bought by the same
buyers and sold by the same sellers as in the free market, but raise the price to $20/bottle. This
must be backed up by government force, as a two-part regulation. The first part is those 200,000
consumers must be forced to buy on pain of prison, something not all of them would do at such
a high price. The second part is that sellers must be forbidden to reduce their price, on pain of
prison, since otherwise they would shade their price to try to acquire more customers.

Having imposed the regulation, let us calculate the surpluses again. This will be easier than
before, because the the gross buyer benefit and the seller cost have not changed from their free
market levels! Since the same people are buying and selling the same 200,000 bottles, the buyers
still value the bottles at $4,000,000 and the sellers still value them at $1,400,000. All that has
changed is that the buyers now pay a much higher price a total amount of ($20/bottle) (200,000
bottles) = $4,000,000 and the sellers get that higher revenue. Thus, now the consumer surplus is
($4,000,000- $4,000,000) = $0, and the producer surlpus is ($4,000,000- $1,400,000)= $2, 600,000.
The total surplus is unchanged from its free market level of $2,600,000; all that has happened is
that now the sellers get all of it and the buyers get none of it.

Whatever price is chosen under this two-part regulation, the total surplus will stay the same.
When the quantity is fixed, the price is just a transfer from buyer to seller. The total surplus is

(Gross Buyer Benef it P rice Quantity) + (P rice Quantity Seller Cost).

The P rice Quantity terms cancel each other, so the total surplus is (Gross Buyer Benefit - Seller
Cost), which does not depend on the price.

The total surplus does depend on the quantity, however, which we were keeping fixed at
200,000 bottles. To see why this quantity maximizes value, consider increasing it. We will need
the most reluctant seller to become active, one whose value for a bottle is $10.00 and has already
sold some but not all of his 100 bottles, since sellers with lower values are already selling. We will
need a new buyer to become active too, one whose value is less than $10.00, since buyers with
higher values are already buying. Even if the new sellers value is $10.00 and the new buyers
value is $9.99, the highest possible, this new exchange reduces total value by $.01 rather than
increasing it. Increasing sales further would reduce total value by even more.

How about reducing the quantity below 200,000? If a buyer with a value of $10.01 or greater
stops buying from the most reluctant seller, with his value of $10.00, then total surplus drops by
at least $.01. Reducing the quantity below the free market equilibrium level loses some of the
gains from trade, from active sellers having lower values for the whisky than active buyers.

Thus, we see that the workings of the free market maximize value. First, the free market
arrives at the equilibrium price, without any government intervention necessary. Then, it turns
out that the equilibrium price elicits an equilibrium quantity which maximizes the sum of producer
and consumer surplus.

The fact that this is a two-step process is why the price matters more than it may have
seemed in my discussion earlier. Earlier, I showed that if the government required the quantity to
be at 200,000, with the same buyers and sellers as in the free market, then the government could
require any level it wanted to for the price without altering total surplus. Thus, it seemed that
this two-part regulation maximized value just as well as the free market could, even if it could not
do any better. But notice what I took for granted: that the government had detailed information
at its disposal and it could enforce its regulation costlessly.

How would the government know that the optimal quantity is 200,000? Supply and demand
curves are not written down in books that the government can consult. Economists measure
them with intricate statistical procedures, using market-generated data. Simply asking people
how much they would pay, partly because they do not think hard enough about it if they do not
really have to pay, and partly because they might not tell the truth. What would happen if, for
example, the government said it was going to set the price of a bottle of whisky to one penny per
bottle, and asked buyers to step forward if they were one of the 200,000 buyers with the highest
values? Moral scruples aside, all 300,000 of the consumers would step forward, since all of them
would like the chance to buy whisky at that price. If the government furthermore asked which
2,000 sellers had the lowest costs, no seller would step forward none of them would want to be
forced to sell at such a low price.

The free market, on the other hand, needs very little information. The process described
earlier moves the price to $10.00 without any need for people to tell the truth. The reason is that
in an actual market, buyers and sellers put their money where their mouth is. A buyer has no
reason to claim that he would only pay $5.00 for a bottle if his true value is $13.00 and the result
of his claim is that he loses the chance to buy at $10.00. A seller has no reason to claim that his
minimum acceptable price is $11.00 if it really is $9.00 and his lie will lose him the sale. This
economizing on information is a huge advantage of the marketplace, even more important than
the fact that it does not need to pay police to enforce its prices and quantities.

The logic of this is similar to the logic of why the single transaction between Smith and Jones
maximized value. Indeed, we can think of Smith and Jones as being just two of the thousands
of participants in the market. What is dierent in the case of the market is that anonymous

Hosted by The Berkeley Electronic Press
market forces determine that the price will be $10, whereas in the single transaction example I
said that Smith oered Jones a price of $10 without explaining where that price came from. But
in both the single transaction and the market, the essential idea is that if both buyer and seller
voluntarily agree to a transaction, it (a) benefits both of them, and (b) increases total value.

Thus, in ordinary economic transactions, the free market maximizes value without the need
for government intervention. That is why economists are generally against government regulation.
It is an important idea to understand, and much of an introductory economics class is devoted
to trying to explain it. But economists also recognize that there are situations in which some
premise underlying this reasoning fails, so that government regulation could help. These are called
situations of market failure, and will be our next topic.

2.3 Product Quality

Not all government regulation is of prices and quantities. As much or more concerns other
characteristics of a market product quality, safety, contract terms, and suchlike. Here, too, the
free market has its virtues, especially in a situation of limited information.

Take as a simple example the market for childrens carseats. Suppose carseats can be made
of varying strengths, ranging from carseats that will protect only 50% of children from serious
injury in a crash at 40 m.p.h., little better than no carseat at all, to carseats that will protect
99% of them. Which kind of carseat would you buy?

The answer depends on the price. If the cost of producing carseats ranges from $20 for a
50% safety rating to $40 for 70%, $80 for 80%, $160 for 90%, $320 for 98%, and $900 for 99%, as
shown by the total cost curve in Table 3 and Figure 2, then you probably would not choose the
safest carseat. You have a total benefit curve for your personal tradeo between cheapness and
quality that shows how much you would pay for a given quality. The total benefit curve in Figure
2 says that the consumer would pay

$100 for a 50% safety rating, $600 for 70%, $800 for 80%, $900 for 90%, $950 for 98%, and
$953 for 99%.

Figure 2: Costs and Benefits of Childseats

If the consumer were able to buy any of these carseats at cost, which one would he buy?
Notice that he is willing to buy any of them, because his benefit from any one of them exceeds
its cost. If he does have a choice, however, he will choose the carseat rated at 90%. His net
benefit his consumer surplus would be $80 for a 50% safety rating, $560 for 70%, $720 for 80%,
$740 for 90%, $630 for 98%, and $53 for 99%, as shown in Table 3. The consumer will not pick
the carseat he values highest, the one rated at 98% with the value of $953. Rather, he will pick
the carseat that gives him the best value for the money. Moreover, his choice maximizes value
according to our criterion, because the transaction adds $740 in value to society, the consumers
$740 in consumer surplus plus the sellers $0 (since he is selling at cost) of producer surplus.

Hosted by The Berkeley Electronic Press

Safety Rating Cost Benefit Price Consumer Surplus Producer Surplus Value Created
(= Cost) (= Benefit-Price) (= Price- Cost) (= CS + PS)

50% $20 $100 $20 $80 $0 $80

70% $40 $600 $40 $560 $0 $560
80% $80 $800 $80 $720 $0 $720
90% $160 $900 $160 $740 $0 $740
98% $320 $950 $320 $630 $0 $630
99% $900 $953 $900 $53 $0 $53

By the way, this is a good example of the unreliability of asking consumers about how much
they would pay for something or even yourself. Despite all the jokes made about it, shopping
truly can be a tiring activity, and a big reason is that the shopper actually thinks very hard
about the tradeo between buying good X or good Y or buying neither and saving the money
for another day. If the shopping is for a childseat, and the tradeo is between the childs safety
and saving money, the shopper bears the additional mental strain of having to admit to himself
that a little extra safety is not worth doubling the price. The baby is cute, but its safety is not
really priceless. If the situation is not real shopping, and is just telling some stranger what kind
of childseat you would buy, the embarassment of saying that you would pick a cheap one rises
even further. The mental strain and embarassment can all be avoided, though, by quickly saying
that you would of course buy the $900 carseat.

Returning to the main thread of argument, we have seen that the value-maximizing level of
safety is 90%. What level would be sold in the free market?

The free market would provide exactly the right level of safety. We can see this by an
argument similar to the one for why the market provides the right price, and thus the right
quantity. Suppose the market initially had all the sellers selling carseats rated at 50% at a price
of $90. That price would not be stable, because a seller who cut his price to $89 would have
almost the same profit margin, but would attract all the customers, vastly increasing his profits.
Other sellers would undercut him, and the price would be pushed down to the cost, $20. Producer
surplus would be driven down to $0 by competition, and consumer surplus would rise to $80.

$20 would be the equilibrium price if the only possible childseat were one rated at 50%.
That, however, is not the case in our example. One of the sellers would innovate by introducing a
higher quality, higher price childseat. Since the buyers are getting $80 in consumer surplus from
the childseat rated at 50%, he would have to give them just as good a deal from his innovation.
He can do that by oering a childseat rated at 90% for $500, which would yield $400 in consumer

surplus ($900-$500) and $340 in producer surplus ($500-$160). That deal would outcompete the
cheap low-quality childseat. Other sellers would imitate him, though, and in equilibrium they will
all oer the childseat rated at 90%, at a price of $160. This is a stable outcome, because no seller
could change his price or quantity and increase his producer surplus.

This argument does depend crucially on competition among sellers, and one might wonder
whether consumers in an uncompetitive market, with just a single monopoly seller of carseats,
would have to pay high prices for low quality. We will address that in more detail later, but the
bottom line is that although usually monopolies will charge prices too high to maximize value,
there is no reason to expect them to choose low quality instead of high quality. In the childseat
example, in fact, a monopolist would maximize value even in his choice of price.

Table 4 illustrate what happens. Now that we have a monopoly seller, he is free to choose
the price and safety to maximize his producer surplus without any threat from competitors un-
dercutting his price or introducing new products. The lowest quality possible is the childseat
rated at 50%, and the monopolist would sell that for $100, the maximum the consumers would
pay, generating a hefty $80 in producer surplus for him. But he can do better. If, instead, he
chose the value-maximizing safety level of 90%, he could raise his price to $900. Since his cost
would only rise to $160, his producer surplus would go up to $740, a clear gain. The monopolists
product will thus be just as good as the competitive markets.

The logic is simple. Since the monopolist is able to grab all the consumer benefit by charging
a price equal to that benefit, he has an incentive to make that benefit bigger. He will not choose
quality to be at the maximimum possible level, since he does have to pay more costs to producer
higher quality, but he will increase the quality so long as the extra quality is worth the extra cost
exactly the calculation that maximizes value.

Hosted by The Berkeley Electronic Press

Safety Rating Cost Benefit Price Consumer Surplus Producer Surplus Value Created
(= Cost) (= Benefit-Price) (= Price- Cost) (= CS + PS)

50% $20 $100 $400 $0 $80 $80

70% $40 $600 $600 $0 $560 $560
80% $80 $800 $80 $0 $720 $720
90% $160 $0 $160 $0 $740 $740
98% $320 $950 $320 $0 $630 $630
99% $900 $953 $900 $0 $53 $53

Government regulation of safety is therefore unnecessary in this example (though we will

later be examining some of the assumptions underlying it). In fact government regulation might
fall prey to the information problem I mentioned above, that people who are not seriously buying
a childseat would not provide accurate information about their valuations. If a state legislator
must vote publicly on whether he thinks children should be protected at the 99% level instead of
the 90% level, what will he do? Like the consumer, he may shut his eyes and sincerely choose 99%
to avoid admitting to himself that he trades o cheapness against child safety. Even if he knows
what maximizes value, though, he may still want to avoid showing in a publicly recorded vote
that he puts money ahead of children, especially if he hopes that the bill will fail even if he votes
for it. And he is right to fear that public vote, because the voters form their publicly announced
political positions in the same way as they answer surveys about how much they would pay for
new products without much thought, and knowing that whatever words they speak, it wont
take money out of their pockets. Thus, getting to the optimal quality by regulation is typically
more dicult than getting there by the free market.

2.3 Innovation

There is a common perception that regulation hurts innovation. This needs some analy-
sis, though, because we must answer why regulation would hurt innovation more than it hurts
established industries, and why, indeed, it should aect innovation at all. If a country imposes
onerous regulations that raise the cost of doing business, it is clear that this will reduce the level
of production in that country. If the government simply imposed a tax of 10% on all economic
activity, we would expect economic activity to decline. People would substitute, increasing the
the amount of their time spent on noneconomic activities, which are untaxed, instead of on labor

to generate taxable income. People would save and invest less, too, since the production in which
they invest would be subject to the tax.13 The eect on investment is secondary, though it is
not that investment is taxed, only that since production is taxed, activities such as investment
that are complements to production become less attractive. Nonetheless, investment would fall,
and so would economic output and growth of output.

Regulation is typically dierent from a tax of 10% of revenue, however. It creates two kinds
of costs: a one-time fixed cost, and a variable cost that must be paid each year and rises with
output. If the government requires that each factory submit an annual report listing the machinery
it uses and possible hazards to workers, the variable cost is the cost of measuring the amount of
machinery and filling out the form, a cost that must be borne each year and that rises with the
amount of machinery used. But there is also a one- time fixed cost: each factory must learn what
the regulation requires of it, where to obtain the forms, and how to figure out the hazards from
each type of machinery used.

These two costs have dierent impacts. The variable cost is one that will depress total
surplus each year after the regulation is imposed. The fixed cost is a one-time cost that will have
its biggest impact in the year the regulation is imposed. After that, factories that use old types of
machinery will not have to incur the cost again there will just be a momentary decline in surplus
from that source as far as they are concerned. New factories, however, or old factories that use
new kinds of machinery, still have to pay the fixed cost.

Joseph Schumpeter defined innovation as a combining of factors in new ways, as opposed to

simply an increase in the amount in capital, labor, and other inputs in old combinations. This can
be done in a variety of ways. What happens with all of them is that more value can be created
without any increase in the amount of resources. This is not the only source of economic growth,
or even the most important one. There are also increases in labor because of population growth or
increased working hours from the same size population, and increase in capital because of saving
and investment, but neither of those increase output without increasing the amount of inputs.
Moreover, much, perhaps most of the growth from increases in inputs is also innovation, because
the inputs are combined in new ways. A new worker enters the workforce and is employed on a
new software project, which is financed by bonds that increase the total amount of capital in the
economy. That is as opposed to an existing factory hiring the new worker and floating bonds to
add a duplicate of its existing building and machinery.
I am ignoring special cases here, which arise in theoretical models but are implausible in the real world.
For example, in theory a higher tax rate could induce people to spend more time at work, not less, because
they would want to restore their post-tax income to its old level and need higher pre-tax income to do
that. This could happen, at, say, a 99% tax rate, since somebody who did not move to a 14-hour work day
would starve to death on the 1% of an 8-hour days income left to him after that high a tax. The Laer
Curve fails at such high tax rates.

Hosted by The Berkeley Electronic Press
Schumpeter says this better than I can. In Chapter 2, Section II of his 1912 book, The
Theory of Economic Development, he says:

In so far as the new combination may in time grow out of the old by continuous
adjustment in small steps, there is certainly change, possibly growth, but neither a
new phenomenon nor development in our sense. In so far as this is not the case, and
the new combinations appear discontinuously, then the phenomenon characterising
development emerges. For reasons of expository convenience, henceforth, we shall
only mean the latter case when we speak of new combinations of productive means.
Development in our sense is then defined by the carrying at of new combinations.
This concept covers the following five cases: (i) The introduction of a new good
that is one with which consumers are not yet familiar or of a new quality of a
good. (2) The introduction of a new method of production, that is one not yet tested
by experience in the branch of manufacture concerned, which need by no means
be founded upon a discovery scientifically new, and can also exist in a new way of
handling a commodity commercially. (3) The opening of a new market, that is a
market into which the particular branch of manufacture of the country in question
has not previously entered, whether or not this market has existed before. (4) The
conquest of a new source of supply of raw materials or half-manufactured goods, again
irrespective of whether this source already exists or whether it has first to be created.
(5) The carrying out of the new organisation of any industry, like the creation of
a monopoly position (for example through trustification) or the breaking up of a

Notice that the new combination of inputs is not restricted to what we normally think
of as innovation new products, or new processes for producing old products. It also includes
introducing old products to new markets, new sources of supply, and new organizatonal forms.
All of these are ways to get more value out of the same old quantity of inputs.

Schumpeters concept of innovation (development in his terminology) is appropriate for

looking at the eect of regulation on growth, because it is the new combination of inputs that must
rethink how to satisfy government regulations, especially if it is done by a new firm, but even if it
is just an old firm doing things dierently. Bureaucratic procedures are inherently conservative;
the bureaucracy, no more than anyone else, wants to think about how to deal with new situations
unless they are paid for it. And unlike the private entrepreneur, the bureaucrat does not gain
wealth by encouraging innovation, unless it is by accepting bribes.14

Thus, new combinations of inputs are more aected than old combinations, because they
See Hernando de Soto . Shleifer, A. S. Djankov, R. La Porta, and F. Lopez-de-Silanes (2002), The
Regulation of Entry,Quarterly Journal of Economics.

face a one-time fixed cost that the old combinations have already paid, a fixed cost which amounts
to a tax on innovation. This fixed cost is all the more important, however, for two additional
reasons: the marginal profitability and the small size of the average innovation.

The average innovation expects, adjusting for its riskiness, to be barely more profitable than
the least profitable existing stable business, which is just covering its cost of capital. Some existing
firms are highly profitable, earning much more than their cost of capital. They, of course, will
continue in business. Indeed, since existing firms have already sunk many of their capital costs,
they will continue to operate even if they face an unforeseen regulatory burden. A few existing
firms are marginal, however, and will go out of business because of the new cost. We should
expect more innovations, however, to be of marginal profitability, ex ante. If an innovation were
to be exceptionally profitable now, it would have been undertaken already. We think of innovation
as being highly profitable, but that is only because we think only about successful innovations.
For every successful innovation, there are many unsuccessful ones, and it is hard to know which
is which in advance. But this means that a small increase in costs will have a disproportionate
eect on innovations. It will shift the balance for many of them from bare ex ante profitability
to a negative value. New factories are only marginally profitable projects. Otherwise, they would
already have been built. And so even a small extra cost will often kill the project.

The second problem is that innovations tend to be smaller than existing businesses. They
are new and risky, and that means it makes sense to start them small and expand their scale
only if they are successful. Also, it is hard to evaluate their profitability, especially for someone
outside the industry, and this means entrepreneurs have more diculty in obtaining capital than
well- established firms. Banks and stock owners prefer known risks. Thus, the entrepreneur will
start at a smaller scale.

If a firm is smaller, though, a fixed cost cannot be spread across so many dollars of sales
revenue. A million-dollar firm feels a fifty thousand dollar regulatory cost much more than a
hundred-million- dollar firm would. This means that the fixed cost of regulation will hit innovation
harder than it does existing production. Thus, regulation hurts innovation more than it hurts
existing production. It hurts both new small businesses, and new large ones.

I would like to discuss one fixed cost of regulation, a less tangible one, in more detail. This
is the the problem of regulatory risk. A major cost of regulation, and especially of regulation
that changes over time, is learning the rules. If you are operating a business in world without
regulation, and you know that it is without regulation, you can devote your energy to everyday
operations and you can plan without fear of running into some unknown government constraint.
If, however, the government regulates, and you know this but you do not know exactly how it
regulates suddenly you face a risk which may be much greater than the regulation itself. It is
like walking across a minefield. Almost every step you take in a minefield is safe. The mines are
scattered, and cover only a small fraction of the square footage. If you knew where each mine

Hosted by The Berkeley Electronic Press
was, you would have only a few extra steps to take to cross safely. But if you do not know where
they are, the fact that most steps are safe is little comfort. You know that sooner or later you
almost certainly will step on a mine.

Unknown regulation is like that. The business may know that most of its activities are legal
but which of them are not? And what will happen when it breaks a regulation? The trouble
spent dealing with potential regulations that turn out not to exist may be far greater than the
trouble spent dealing with the regulations that actually exist.15

My own university, with the aid of the Federal Government, provides a humorous example.
At various times in history, scholars have peformed experiments that endangered or hurt human
subjects. At worst, they tried out dangerous new medical treatments or withheld useful existing
ones to see what would happen. Less culpably, psychologists did experiments that tricked the
subjects or made them anxious, such as the famous Milgram experiments in which the subjects
were asked to administer dangerous levels of electric shock to other people or so they thought,
since actually the shocks were fake. The government response was regulation requiring universities
to check all experiments and only approve good ones.

Here is how Indiana University has responded:

1. If I am just talking with people about....Im not doing anything to themthere are
no experiments, no clinical trials, do I need human subjects approval?
Yes (and we are not kidding). Federal regulations define human subjects research
broadly to cover interactions as well as interventions with human subjects for research
purposes. So. . . surveys, interviews, questionnaires and oral history interviews,
etc. are all covered by the federal regulations. And, yes, you need prior committee
approval. Most of this type of research, however, qualifies as EXEMPT.

2. If my research qualifies as exempt under the federal regulations, must I submit

an application for approval?
Yes. The term exempt in terms of the review process really means exempt from
full committee review. It is the policy of Indiana University to review all research
in order to ensure that the research is, in fact, exempt. The application for exempt
research is less extensive, and under most, but not all, circumstances, written, signed
informed consent is not required. We do, as a general matter, require investigators
This is like the joke about the man who figured out a way to stop the thief who was stealing watermelons
from his garden. He put a sign out, saying, Thief: One of the watermelons in this garden is poisoned.
This, of course, made it it unprofitable for the thief to steal, even though almost all the watermelons were
perfectly harmless. (The punchline though, is what happened the next morning. The man looked out, and
found all his watermelons still there, untouched. He was happy till he saw his sign. The One had been
scratched out and replaced with Two.

give to their subjects a written statement containing information about the study
and their part in it.

3. If I am using data complied by other third parties, e.g. governmental agency or

another researcher, do I need approval?
Yes. Use of existing data is exempt research but still requires prior approval. Even
if you have worked with the original investigator in obtaining the data, your indepen-
dent use of the data requires approval. This issue comes up often with Ph.D. disser-
tation research. All Ph.D. candidates must get their own human subjects approval
for their research even if it involves working with data collected by a collaborator
with his or her own human subjects approval.16

Thus, Indiana University requires researchers to submit applications for approval to the
human subject research committee even if the research is exempt from the federal regulation,
so the university can decide if it is exempt. This is a substantial burden, since the University
requires a detailed description of the project, which must not begin before getting approval,
and new approval must be gotten if the research plan changes which research plans always do.
Moreover, the University requires approval even if the research never involves interaction with a
human subject, and merely uses data collected years before by some other researcher.

The apparent reason for this is that the University is afraid that even research that seems
to be exempt from the federal regulation might not be that the federal government is entirely
unreasonable and will cut o federal grant support to Indiana University. This seems unlikely,
but the potential cost is so great that the organization has responded to a relatively mild federal
regulation with much greater internal compliance procedures.

In private business, too, fear of unknown regulations is an important motivation in an atmo-

sphere in which firms know that numerous regulations do exist but do not know them exactly
or in which regulations are so numerous that nobody knows all of them. Such may reasonably
prevent someone from innovating at all, knowing that somewhere along the way, a regulatory
roadblock is likely to kill his project, or at least make the dierence between profit and loss.

Finally, a last problem that regulation can create for innovation is by directly freezing current
production methods in place. This is not a feature of ideal regulation, but it is a common feature
of regulation in practice. One way this can happen is if regulation prevents entry of new firms.
In the America of 1960, for example, it was dicult to start a new bank, or even a new branch
of an old bank. Naturally, this makes innovation more dicult, since innovation would have to
be carried out by existing firms. A second way it can happen is by prescriptive regulation:
FREQUENTLY ASKED QUESTIONS About the Human Subjects Approval Process, Blooming-
ton Campus Committee for the Protection of Human Subjects 5/21/2004,
resrisk/faq.html (viewed 8 January 2005).

Hosted by The Berkeley Electronic Press
regulation that prescribes what technology is to be used to solve some problem. A standard ex-
ample is the requirement that coal-burning power plants use scrubbers on their smokestacks to
remove sulphur. This certainly reduced pollution, but it froze the technology in place. Another
technology that existed at the time was simply to use low-sulphur coal from Wyoming instead
of the cheaper high-sulphur coal from West Virginia. Low-sulphur coal without scrubbers gen-
erated no more pollution than high-sulphur coal with scrubbers. But the prescriptive regulation
required scrubbers to be used regardless of the type of coal. Moreover, the regulation gave no
incentive to firms to come up with more eective scrubbers than currently existed, or to find other
technologies some new type of boiler, for example that might reduce sulphur emissions.17 Thus,
in a number of ways, government regulation of economic activity hurts innovation even more than
it raises the costs of operation using old techniques. The cost of regulation must be calculated to
include not just an increase in the cost of current methods, but what is much harder to measure
because it does not yet exist: the lost innovation and economic growth.

3 Market Failure and Economic Regulation

Markets usually work well. They do not always work well, however, even for maximizing
value. There are a number of separate justifications for economic regulation which we will look
at one by one.

Let us return to the single-transaction case as a running example. To summarize the case we
discussed earlier, Smith is willing to pay up to $15 for a bottle of whisky and Jones is willing to
accept as little as $8. Our conclusion earlier was that the government should allow the transaction
to take place, because it benefits both buyer and seller. The free market works without any need
for government intervention.18

This story can be challenged in a variety of ways. First, it may be that you do not care for
the goal of value maximization. That is fine, but we will return to that in a later section. Even
if you accept value maximization, however, things can go wrong in the story, generating what is
called market failure.

3.1 Enforcing Property and Contract Rights

A certain amount of government regulation is needed even to support free market transac-
tions. If the government has no rules forcibly constraining anyones actions, then Smith faces no
government penalties if he steals the bottle of whisky from Jones instead of paying for it. Laws
against theft are a form of regulation and a constrain on our liberty, though we usually take such
xxx Wildavsky book : Nuclear power. Scrubbers story too. Presciprtive regulatoin is a bad idea. Too
specific, hurts innovation.EcREp PRes 2003.
xxx Are transaction costs another reason? Coase Theory of the Firm. If an allocation is repeated, and
simple, command and control may be best.

laws for granted. But if Smith can steal the bottle at low cost, he will steal it even if his value is
less than Joness value and value will not be maximized. Moreover, Jones will use time, energy
and resources in trying to prevent theft; and Smith will use time, energy, and resources in trying
to overcome Joness precautions. Most of that investment is nonproductive from a social point of
view. It may, in the end, be important to making sure that the bottle goes to the highest valuer,
but a cheaper way to do that is to outlaw theft and require voluntary transactions.19

Put another way, it is value maximizing to have a government provide some protection of
property rights. This protection includes not just laws against theft and police to enforce them,
but laws defining property rights. Such laws are simple when it comes to bottles of whisky,
but they get more complicated when it comes to such things as ownership of corporations, wild
animals, stolen goods, and children. Often, the government goes a step further and provides for
clear markers of who owns what, as in the requirements that automobile ownership be registered
and land titles be recorded. These regulations make it much easier to buy cars and land, since
the buyer has more assurance that the seller really owns what is being sold.

Closely related to this role of government is governments role in enforcing contracts. Suppose
Jones would like for Smith to pay today for a bottle of whisky Jones will only be able to deliver
tomorrow (this might be because Jones needs the cash to get the whisky from his distributor).
Without a law against regulating breach of contract, Smith will be reluctant to hand over the
money, since Jones could just keep it and refuse to deliver the bottle. Some regulation must say
what happens if Jones refuses delivery. The law could be that Jones must refund the money, or
that he must in fact deliver the bottle, or that he must pay enough money to make Smith as
content as if the bottle were delivered. Scholars debate which of these rules maximizes value, but
all agree that any of them is better than no rule at all.

This, in fact, illustrates an important point about government role in property and contract
regulation: What matters most for value maximization is that there be clear rules about who
owns what, and that there be high enough penalties for broken promises. Getting the exact rules
right is much less important, because people can adapt to imperfect rules. But if nobody is quite
sure what the rules are, or who has the power to get what he wants, that is sure to hurt value.20
I should caution, though, that in some situations government precautions against theft are not the
value-maximizing policy, because it may be that Jones can protect himself more cheaply than the govern-
ment can. In practice, we use a mixture of personal precautions locks, alarms, and handguns and public
precautions police, courts, and prisons.
If the government provides clear rules, and is a strong and honest government, its overwhelming power
is behind whoever owns the property under the rules. If the government does not provide any rules, then
personal power is what matters. If personal power were clear as when one person could defeat anybody
else then lack of rules would not matter. But almost always it is unclear who would win in private battles
among coalitions of citizens if winning comes down to pure power.

Hosted by The Berkeley Electronic Press
3.2 Property Rights and Creation of Goods

I said in the previous section that the most important thing is to provide clear rules as to who
owns what, and the precise rules matter less. That is true, but mattering less does not mean
does not matter. Particularly important is that the property rules award enough ownership to
someone who creates something new.

This is most obvious if someone creates a new physical good. Suppose Jones has distilled the
whisky using his own labor and corn that he grew himself. It is natural to have a government rule
saying that Jones therefore owns the whisky, and has the right to transfer ownership to Smith.

Suppose, though, that the government had a rule that whatever Jones creates belongs to
Smith automatically, without any need for Smith to pay. That is avery clear rule, but not a
value-maximizing one. The problem is that Jones now lacks all incentive to create new whisky,
since only Smith will get any benefit. Even if we add another rule allowing Smith to whip Jones if
he so pleases, Joness incentives are still not right for producing whisky, even if the cost to Jones
is less than the benefit to Smith. The problem is an information one: it is hard for Smith to know
what Jones is capable of doing, since Jones has every incentive to claim that he is untalented and
weak, and that even the threat of whipping would not be enough for him to make whisky.

This is why slavery is not value maximizing. I have just described a situation in which Jones
is Smiths slave. We do not have to appeal to natural right to deduce that slavery is bad. Even
with our goal of value maximization, it has a problem: slavery does not give enough incentives to
create value. And, in fact, in time when slavery was legal, slaveowners found that even though
the law did not require them to give rewards to their slaves, carrots had to be used as well as
sticks if a slave were to be induced to do anything but the simplest tasks. Crude methods can
get people to put in the hours, but not to be careful or imaginative.

The government does not have to give Jones the right to 100% of the whisky he produces to
give him reasonable incentives, and in fact very few governments do. The usual rule is that Jones
has the right to keep most but not all of what he produces, giving some to the government. That
is the essence of an income tax.

Some kinds of creative activities are regulated in a dierent way. If someone comes up with
a new idea, rather than new physical object, the government faces a quandry. If the person has
complete and exclusive rights to his idea, he will overcharge others to use it the problem of
market power that we will discuss next. But if he has fewer rights to it, he has less incentive to
come up with new ideas.

The compromise response is for government to award a limited period of exclusive ownership
of the idea, in the form of copyrights or patents. The creator gets ownership for 20 years (for
patentable ideas), with the limitation that he must pay taxes on his profits, but after 20 years,

anybody can use his idea for free.

3.3 Market Power

The first two uses of government regulation I have discussed are easy to accept. We so take
it for granted that the person who has created something or paid someone else for it owns it
that we do not consider government rules against stealing it at all intrusive. These regulations
seem to be a foundation for the free market rather than an intervention in it. But we now come
to a somewhat dierent form of market failure, one where value maximization requires that the
government restrict someones right to freely decide whether to buy or sell his property.

Ordinarily, an exchange benefits both parties, resulting in both producer and consumer
surplus. The market transaction part of the whisky example showed that, because the price of
$10 per bottle resulted in the quantity being traded that maximized value.

In the individual transaction part of the whisky example, the price of $10 is between Smiths
value of $15 and Joness value of $8. The price could be anywhere between $8 and $15, and the
sale could still take place. But the price is not pinned down. It is the result of bargaining between
the two parties. Bargaining is strategic, and sometimes, especially when information is poor, it
reaches inecient results. If Smith were to believe that Joness value for the bottle is only $5,
for example, Smith would hold out for a price of no more than $5. No sale would take place, and
value would not be maximized, since Smith really does value the whisky more than Jones.

Monopolies are a particular example of the problem of poor information and market power.
If a company has a monopoly on a product, that does not by itself create market failure, but in
practice it always does. The problem is that the company does not know the exact valuation of
each consumer. If it did (and it could block resale) then it would set a dierent price for each
consumer exactly equal to that consumers valuation, and all the value-maximizing sales would
take place, as in Table 4 earlier. If, however, the company does not know which consumers have
high value and which have lower values, it will have to charge one price to everyone. This price
will be a compromise between the valuations of the most eager consumers and the least eager. As
a result, it will be higher than the valuation of some consumers, and they will not buy. Worse,
the price will be higher than the valuations of some consumers who would be willing to pay more
than the sellers cost. As a result, some value-maximizng trades will fail to occur.21

If one side of the transaction has a monopoly, it can capture more of the surplus, but usually
at the expense of diminishing the total surplus. Monopoly welfare losses arise from these inecient
attempts to capture surplus. The loss may arise either from the attempt to become a monopoly
(as in the expense of lobbying the government for a monopoly), from the attempt to exploit the
monopoly (as in the administrative costs of price discrimination or the familiar triangle losses
xxx A graph would help here.

Hosted by The Berkeley Electronic Press
discussed next).

Monopoly creates other problems too. One of them is that the profits of monopoly give
people an incentive to create monopolies. That is good if they are creating new goods, but bad
if they are just devoting eort to monopolizing existing goods. Thus, J.P. Morgan exerted his
considerable talents to creating monopolies by giant mergers of existing companies. U.S. Steel is
the biggest example, which was created by merging Andrew Carnegies price-cutting steel firm
with a number of other steel firms, with the result that after Carnegie left the scene, prices rose.
This was a profitable transaction for Morgan, but it reduced value rather than creating it, not
just because it eliminated some value- increasing steel transactions, but because it consumed the
attention of investment bankers who could have been doing something that created value instead.

A second problem with monopoly is that it seems to increase production costs. Firm that
are monopolies seem to have higher costs than firms that must compete for their existence. This
is not an obvious result. A monopoly, like a competitive firm, prefers high profits to low profits,
and so prefers low costs to high costs. We have seen earlier that a monopolist, like a competitive
firm, has a strong incentive to produce a high-quality product. Yet although the monopolys
desire for high profits may be just as high as the competitive firms its ability might be lower.
This is not thoroughly understood, but I can give one clear example: the problem of executive
eort. Shareholders of a competitive firm can compare the performance of its executives to the
performance of executives at other firms. A monopoly cannot do that. As a result, the competitive
firm can punish or reward its executives much more eectively than can a monopoly.22 The most
important kinds of regulation justified by monopoly are public utility regulation and anti-trust
laws. The technology of electricity distribution is such that it is best that one firm provide the
service, but important to prevent it from doing so at such a high price the people are discouraged
from using electricity. In other industries, competition would naturally arise, but anti-trust policy
is useful for preventing the competitors from making agreements not to compete.23

3.4 Imperfect Information

Poor information has already featured prominently as both a reason why the free market
is useful and as a reason for market failure. Indeed, economists have realized increasing over
the period from 1950 to the present that solving information problems is the key to a successful
xxx Here cite Joe Farrell.
A quite dierent monopoly-based reason for regulation is to create monopoly profits. When producers
compete with each other, they inflict negative pecuniary externalities on each other by reducing profits.
What producers lose, and more, consumers gain, which is why monopoly is ordinarily thought to be bad.
But if the government does not care about consumers, as might be the case if the producers were domestic
and the consumers foreign, it might want to create monopolies. Thus, it makes sense for the Japanese
government to encourage its automakers to organize voluntary export restraints, because that raises the
prices of Japanese cars in the United States and increases Japanese profits at the expense of U.S. consumers.


Our argument for the free market assumes that everyone in the economy is well- informed
about the value to everyone else of the goods being exchanged. Otherwise, ineciencies result
as the parties try to take advantage of their private information. The monopoly ineciencies
described above are one example; ineency arises if Smith misestimates how little Jones is willing
to accept.

Government regulation of advertising and fraud are justified by information problems. If

Jones claims he is selling whisky, but might actually be selling colored water, one of two problems

First, it may truly be colored water, in which case if Smith buys it, his value is not $15, but
$0, and value has not been maximized (it is true that Joness value might be $0 too, but they
have at least lost the cost of making the worthless transaction, and if Smith is thereby prevented
from making a genuinely value-increasing trade, there is a genuine value loss).

Second, it might be that Jones actually does have real whisky, even though there is no legal
penalty against selling colored water. If Smith knows the legal rule, though, he might refuse to
buy anyway. He does not know that the bottle is full of real whisky, and Joness promises lack
credibility. When there are no laws against fraud, honest merchants lose, as well as charlatans,
because consumers lack confidence in the market.

It is especially dicult to start a new business if reputation is the only basis for consumer
trust. In the absence of government regulation of fraudulent new products, innovation is stifled.
Or, it may be that the consumer does not know what he wants, exactly, and would like to
delegate this task to the government. Thus, the government requires technical safety features on
cars which it thinks almost all consumers would desire. The wide variety of problems arise from
imperfect information have been the subject of a vast literature since 1975, including much of my
own work.24 Most of these problems arise from the diculty of enforcing contracts: the payer
cannot costlessly force the performer to perform, whether that performance consists of product
quality, work eort, or provision of human talent. Many of these problems cannot be remedied by
government regulation, but some of them can, especially by regulations encouraging the truthful
provision of useful information.25

3.5 Externalities
See Eric Rasmusen Games and Information: An Introduction to Game Theory, 3rd edition,(1st edition,
1989), Blackwell Publishers, Oxford (2001), for references.
This kind of regulation is strewn with potential pitfalls for regulation, however. If, for example,
regulation requires extensive disclosure of unimportant information, as is perhaps the case with credit
agreements, the resulting information overload can actually leave the consumer more confused and apt to
buy from disreputable sellers than if the sellers had been allowed to disclose only useful information.

Hosted by The Berkeley Electronic Press
If someone takes action X which has an impact on someone else, but neither party can compel
an exchange of money, then it is said that X has a negative or positive externality, depending on
whether the impact is bad or good. Suppose that Smith, after buying whisky from Jones for $10
and drinking it, will throw the bottle onto the sidewalk in front of Browns house, where it will
shatter and cost Brown $20 to clean up. The transaction between Smith and Jones has indeed
created value of $7 as far as thos two are concerned Smiths value is $15 and Joness was only
$8. But that ignores Brown. External to the transaction is Brown, a third party, who suers a
value loss of $20. This more than wipes out the gain of $7.

The shattered whisky bottle is an example of a real externality: a spillover in which one
persons action aects the utility of someone else without any payment being made, and aects
the utility directly rather than just through the actions impact on prices.26

Pollution controls are an example of regulation based on externalities. When a tire factory
manufactures tires, the fumes created in the process reduce the utility of its neighbors. As a result,
the government commonly forbids factories to emit more than a certain amount of pollution.27

3.6 Coordination Problems

If the spillover results from the prices, it is called a pecuniary externality. If, for example, Jones has
been regularly selling whisky at $10/bottle to Smith, making a profit of $2 per bottle, and then Anderson
begins selling at a price of $7/bottle, Anderson has inflicted a negative pecuniary externality of $2/bottle
on Jones. This is only a pecuniary externality, however, because it arises through prices, not through
changes in the amount of goods that are ultimately consumed. Suppose Andersons cost is $6 per bottle.
Jones has indeed lost $2 per bottle, but Smith has gained $3, and Anderson has gained $1, so there is
no net gain or loss. A pecuniary externality is a spillover harm to someone that is exactly balanced by a
spillover gain to someone else.
Pecuniary externalities have been used as a justification for regulation, but not because they reduce eco-
nomic eciency. In a well- functioning marketplace, pecuniary externalities are inevitable and desirable.
But see Justice Brandeiss position in the text at note ?? infra for an example. On pecuniary externalities,
see R. Tresch (1981) Public Finance: A Normative Theory p. 91; J. Viner, Cost Curves and Supply
Curves, 3 Zeitschrift fur national-okonomie (1932), (reprinted in American Economic Association, Read-
ings in Price Theory, (1952) (the origin, so far as I know, of the term pecuniary externality);Allyn Young,
Pigous Wealth and Welfare, Quarterly Journal of Economics, 27:672-686 (August 1913). English law
early recognized that pecuniary externalities are not a social evil and do not create a cause of action. (The
opening of a new private school reduces the tuition that an old school is able to charge, but that does not
justify compensation, from a medieval case cited in Keeble v. Hickeringill, 103 Eng. Rep. 1127 (Queens
Bench, 1707).)
The perceptive reader will note that the externality justification for government regulation can be
reduced to the justification of setting property rights. The Coase Theorem tells us that what matters most
is that either the factory or the neighbors own the right to clean air, and that this right be alienable.
Ronald Coase, The Problem of Social Cost, Journal of Law and Economics, 3: 1 (1960). If the factory
owns it, it will sell the right to the neighbors; if the neighbors own it, they will keep it. The diculty is
that bargaining costs and information imperfections make it dicult to carry out trades in pollution rights.

The final type of market failure that I will mention is the coordination problem.

When expectations matter, there may exist a number of stable configurations of behavior,
each with its own set of expectations, and some of these equilibria may lead to better results than
others. Such multiple equilibria have been studied chiefly in macroeconomics.28 In the whisky
example, Jones will not expend the $8 to produce the whisky unless he expects Smith to buy
it. But Smith will not waste time visiting Jones unless he thinks Jones has whisky to sell. In
one equilibrium, the whisky is produced and traded; in a second equilibrium, with lower surplus,
Smith stays home and no whisky is produced. This has been suggested as a cause of business
cycles; nobody in the general economy will produce goods unless they expect other people to
produce goods for which to exchange them. One equilibrium has low production and welfare;
another has high production and welfare. Government jawboning might shift the economy from
one equilibrium to the other.29 In another example, nobody wants to be the last depositor
to withdraw their money in a bank run. In one equilibrium, nobody expects a bank run, so
nobody bothers to withdraw their money. In a second equilibrium, everybody expects a bank
run, so one occurs, and the banks fail. Government deposit insurance is intended to eliminate the
bad equilibrium although we have seen from the savings-and-loan mess that it introduces its
own problems.30 In each of the two examples, the government helps by changing expectations.
This may justify government macroeconomic intervention, and some forms of regulation such as
banking law, but it ordinarily does not apply to what we think of as regulation.31

The coordination function of government can be considered a background function, like

protecting property rights, rather than requiring active and continual intervention. Money solves
a coordination problem. People wish to trade with each other, but it is very inconvenient to
have to figure out that value of the goods a person produces in terms of the value of every other
good, and very inconvenient to carry around truckfuls of each thing produced for use in barter. I
xxx A dierent kind of multiple equilibria is the possibility that more than one set of prices might equate
supply and demand across the general equilibrium of the economy. This does not create any ineciency.
Also, microeconomists frequently find multiple equilibria in models of incomplete information such as
signalling models, but it is then dicult to know whether to attribute the market failure to information or
coordination problems. See Chapter 10 of Rasmusen (2001), and the references cited there.
See, for example, Peter Diamond(1982) Aggregate Demand Management in Search Equilibrium,
Journal of Political Economy, 90: 881-894 (October 1982).
Douglas Diamond & Philip Dybvig (1983) Bank Runs, Deposit Insurance, and Liquidity, Journal of
Political Economy, 91: 401-419 (June 1983).
A somewhat dierent multiple-equilibrium problem is created by network externalities: the exter-
nality that if two companies produce a product according to the same standard, sales of both can increase
because of the interchangeability. Thus, government may be helpful in setting a standard railway gauge
or a standard technology for high- definition television. On standards, see Michael Katz & Carl Shapiro
(1985) Network Externalities, Competition, and Compatibility, American Economic Review, 75: 424-40
(June 1985); Charles Kindleberger (1983) Standards as Public, Collective and Private Goods, Kyklos 36:
377-396 (1983).

Hosted by The Berkeley Electronic Press
teach students and write books; would I have to carry books and students to the grocery store to
exchange for pancake mix? Instead, the government provides money for use as a unit of account
and medium of exchange.

Less obviously, the government provides laws that help coordination. Henry Smith has
argued that land may only be sold with certain provisions so as to reduce the transaction costs
of land sales. Rather than checking the fine print on property deeds, the buyer can rest assured
that when he has bought land, he has bought the conventional package of rights.

In most economic transactions, market failure does not arise. Welfare losses are limited
because of the monetary character of economic transactions, because where there is imperfection,
there is profit. The making of the profit will ameliorate the imperfection, though at some real
cost. It may not matter, for example, whether consumers can themselves test the quality of
car bumpers, because either the competing sellers will themselves try to demonstrate quality,
to obtain competitive advantage, or businesses such as the many car magazines one can buy at
supermarkets will enter to provide information to consumers, at a small price. Similarly, the losses
from externalities are limited by bargaining costs. The problems create costs, but these costs are
limited by the cost of technological solutions.

4 Government Failure and Economic Regulation

Even when market failure is a problem, it does not follow that regulation is the proper
response, because one must also consider government failure, which arises from the unwillingness
or inability of the government to regulate eciently. When the government exerts force to restrict
behavior, it also redistributes wealth. This gives private parties an incentive to expend resources
to induce the government to shift the wealth to them, an activity known as rent- seeking.32
Journal of Political Economy, 83:807-27 (1975). Other parties, who want to prevent the
redistribution, will also expend resources. The expenditure might be for bribing and counter-
bribing government ocials, or it might be for political campaigns. In either case, since wealth
is just redistributed, not created, and people use resources to eect the redistribution, societys
total wealth diminishes.

Even if the government is motivated by the public interest rather than the pressures of rent-
seekers, it is not necessarily intelligently motivated. Since the costs and benefits of government
actions flow to third parties rather than the government decisionmakers, there is little incentive
for the decisionmakers to expend eort to make good decisions. And to the extent that those who
See Gordon Tullock, The Welfare Costs of Taris, Monopolies, and Theft, Western Economic
Journal, 5:224-32 (1967); George Stigler The Economic Theory of Regulation, Bell Journal of Economics,
2 (1971); Richard Posner, Taxation by Regulation, 2 Bell Journal of Economics, 22-50 (1971); Richard
Posner, The Social Costs of Monopoly and Regulation,

are hurt or helped have the ability to influence the government decisionmakers, the rentseeking
problem is exacerbated. The government falls between the Scylla of interested rentseeking and
the Charybdis of disintererested incompetence.33

Thus, economic regulation is severely limited in its optimality for two reasons: the Invisible
Hand leads most markets to optimality, and even in the rare cases when government regulation
could help, the helpful regulation may not be what the government implements. We must now
see whether this is as true for social regulation.

I could say much more about government failure, but I will not, for the moment.34

5. Some Actual Regulations of Dubious Value

In this section let us apply these ideas to some actual economic regulations in the United
States and Europe. The theory I have outlined above is uncontroversial. By this, I mean that
they are generally accepted by the scholarly community of economists. I do no know the extent to
which the general public understands or agrees with them, and this would vary among countries.
The situation is a little like evolution versus creationism. Among biologists, the theory of evolution
is generally accepted. Fine points of the theory are still unsettled, and application to particular
species and animal features generate disagreement, but evolution is the working paradigm of
scholars. Some biologists and scholars outside of biology do prefer creationism, and they may
even be right, but they are far from the mainstream, except where they engage the mainstream
in making specific criticism of evolution. Similarly, the economic theories I have laid out the
idea that markets typically maximize value, but market failure does occur, as does government
failure are generally accepted and are what are taught in all the leading universities, but that
does not mean there do not exist heterodox economists and scholars outside of economics who
reject them.

Where there is more controversy within economics is in how to apply the ideas. Does market
failure justify government provision of health insurance? Is this outweighed by the government
failure that occurs when governments do try to provide it? These questions can be answered
dierently by people operating within the standard paradigm, depending on how they view the
facts. The question of the degree of government failure, in particular, is very dicult to answer
with precision, because it requires making political predictions. Ones view of the desirability of
laws to regulate libelous newspaper articles may depend heavily on whether it is the Communist
Party that will win future elections or a democratic party.35

On the pros and cons of reliance on interested providers of information, see Eric Rasmusen, Lobbying
When the Decisionmaker Can Acquire Independent Information, Public Choice, 77: 899-913 (1993).
xxx Do write more here.
xxx to be finished at some later date. In this section, I can at least hope to show the reader what facts

Hosted by The Berkeley Electronic Press
he must believe are true to support certain government regulations currently used in the United States
and Europe. Also asdd a section 6 on alterntative evaluation rules such as Murrays creativity, religious
reasons, and so forth.

Faculty of Law, Monash University
Research Paper No 2012/05

Conceptualising Social and Economic

Regulation: Implications for Modern
Regulators and Regulatory Activity

Eric Windholz and Graeme Hodge

This paper can be downloaded without charge from the Social Science Research Network
electronic library at: 2215334

This paper was first published in (2012) 38(2) Monash University Law Review 212

Electronic copy available
available at:

+$ )+,/#$ (.$ 2(-%+1$ +,<$ #-(,(&%-$ (9A#-*%3#27$ 8,<#'#,<#,*$ )#/01+*()2$ +)#$
#-(,(&%-$ 3+10#27$ !"%2$ +)*%-1#$ #>+&%,#2$ *"#$ #>*#,*$ *($ :"%-"$ *)+<%*%(,+1$
-(,-#'*%(,2$ (.$ 2(-%+1$ +,<$ #-(,(&%-$ )#/01+*%(,$ -(,*%,0#$ *($ ')(3%<#$
+$ 02#.01$ .)+&#:()?$ :%*"%,$ :"%-"$ *($ +,+162#$ &(<#),$ )#/01+*()2$ +,<$
%,$ ')+-*%-#$ :%*"$ *"#$ )%2?$ *"+*$ *"#%)$ -(,*%,0#<$ 02#$ %2$ +'*$ *($ -(,.02#7$ !"#$
+)*%-1#$ *"#,$ '(2%*2$ +,$ +1*#),+*%3#$ :+6$ (.$ -(,-#'*0+1%2%,/$ #-(,(&%-$ +,<$
%,$ :"%-"$ +11$ )#/01+*%(,$ %2$ 0,<#)'%,,#<$ 96$ +$ &%>$ (.$ %,*#)-(,,#-*#<$ +,<$
%,$ 1#/%*%&%2%,/$ *"#$ )#/01+*()6$ #,<#+3(0)7$ !"%2$ 1+*#)$ '(%,*$ C$ :"%-"$ "+2$

!"#$% &'#% ()*&% +,% -#)$*% &'#$#% ')*% .##/% )% 0$)1)&23% 2/3$#)*#% 2/% .4&'% &'#% "4561#%
47% $#865)&24/9% )/0% &'#% /61.#$% 47% $#865)&4$-% )6&'4$2&2#*:% )3$4**% .4&'% #34/4123%
)/0% *432)5% *#3&4$*;<% =#865)&24/% 2/% &$)02&24/)5% #34/4123% )$#)*% 47% 3411#$3#:%
34$(4$)&24/*:% !%/)/3#% )/0% 341(#&2&24/:% )/0% &$)02&24/)5% *432)5% )$#)*% 47% '#)5&':%
*)7#&-:% 34/*61#$% ($4&#3&24/% )/0% &'#% #/"2$4/1#/&:% ')"#% 2/3$#)*#0% 2/% /61.#$:%
.$#)0&':% $#)3'% )/0% 341(5#>2&-;% ?#@% $#865)&4$-% )$#/)*% ')"#% #1#$8#0% 2/% )$#)*%
[% L**432)&#:%S4/)*'%C#/&$#%74$%=#865)&4$-%X&602#*:%P)365&-%47%O)@:%S4/)*'%\/2"#$*2&-;
[[% K$47#**4$% 47% O)@% )/0% E2$#3&4$:% S4/)*'% C#/&$#% 74$% =#865)&4$-% X&602#*:% P)365&-% 47% O)@:% S4/)*'%
9% !$8)/2*)&24/% 74$% B34/4123% C4D4(#$)&24/% )/0% E#"#54(1#/&:% HIJK$ L#3%#:2$ (.$ L#/01+*()6$ L#.()&M$
L#/01+*()6$ N(1%-%#2$ %,$ HIJK$ J(0,*)%#2M$ O)(&$ 8,*#)3#,*%(,%2&$ *($ L#/01+*()6$ @(3#),+,-#% F<,,<GH%
=#865)&24/% I)*J74$3#:% K)$52)1#/&% 47% L6*&$)52):% L#*"%,?%,/$ L#/01+*%(,M$ L#'()*$ (.$ *"#$ !+2?.()-#$ (,$
<% N)32/&%N4$0)/):%E)"20%O#"2DP)6$%)/0%Q)"2#$%P#$/)/0#R%2%S)$T/:%UI'#%V54.)5%E2776*24/%47%=#865)&4$-%

Electronic copy available

Electronic copyat:available
!"#$%&'()*+,+#-./"$+)*.)#0.1$"#"2+$.3%-(*)'+"#4.52&*+$)'+"#,.6"7.8"0%7#.3%-(*)'"7,.)#0. !"#

<8-6,.8*0 &46.,'6*0 4$260 966%0 18&0 7.,:$,)0 &.0 6=15$'%0 &4'*0 '%+,6$*60 '%0 $%0 6,$0
.9*6,26)/0 -.)6,%0 :6*&6,%0 >.26,%-6%&*0 4$260 8%)6,>.%60 78%)$-6%&$50 +4$%>6/0
-.2'%>0 7,.-0 $0 1.*'&'260 *&$&60 C0 '%0 :4'+40 >.26,%-6%&*0 '%&6,26%6)0 )',6+&530 '%0
.,)6,0 &.0 $+4'6260 $0 ,$%>60 .70 *.+'$50 $%)0 6+.%.-'+0 >.$5*0 C0 &.0 $0 ,6>85$&.,30 *&$&60
C0 '%0 :4'+40 )',6+&0 *6,2'+60 )65'26,30 '*0 '%+,6$*'%>530 .8&*.8,+6)0 &.0 &4',)0 1$,&'6*/0
:4.0 >.26,%-6%&*0 *66D0 &.0 +.%&,.50 $%)0 '%!086%+60 &4,.8>40 $0 -'=0 .70 +.%&,$+&8$50
$,,$%>6-6%&*/0 ,856*0 $%)0 ,6>85$&'.%*;#0 E,$'&4:$'&6/0 F.>5'$%6*60 $%)0 G62'(H$8,0
*'-'5$,530 .9*6,26)0 &4$&0 .70 &460 &4,660 78%+&'.%*0 .70 >.26,%-6%&*0 C0 1,.2')'%>/0
&460 1,.2'*'.%0 .70 >.26,%-6%&0 *6,2'+6*0 4$*0 966%0 $++.-1$%'6)0 930 $0 1,.5'76,$&'.%0
6776+&'26%6**K0 $%)0 &460 *.+'$50 ,6*1.%*'9'5'&30 .70 &460 1,'2$&60 *6+&.,0 .,>$%'*$&'.%*0
U&46,0 +.--6%&$&.,*0 $&&,'98&60 &460 '%+,6$*60 &.0 46'>4&6%6)0 1895'+0 6=16+&$&'.%*;0
?8%*&6'%/0 7.,0 '%*&$%+6/0 $,>86*0 &4$&0 &460 '%+,6$*60 '*0 $0 ,6!06+&'.%0 .70 $%0 6=&6%)6)0

#0 V'$%).-6%'+.0 A$B.%6/0 NW460 S'*60 .70 &460 S6>85$&.,30 ?&$&60 '%0 X8,.16O0 I"YYLJ0 "QI#J0 <%,'. 1(7"&%)#.
="*+'+$,0 QQK0 V'$%).-6%'+.0 A$B.%6/0 NH,.-0 &460 Z.*'&'260 ?&$&60 &.0 &460 S6>85$&.,30 ?&$&6[0 F$8*6*0 $%)0
U,>$%'*$&'.%0 7.,0 X+.%.-'+0 F.(.16,$&'.%0 $%)0 ]6265.1-6%&0 *'-'5$,530 .9*6,26)0 &4$&[0 NS6>85$&'.%0 4$*0
L0 A.,60,6+6%&53/0$0%8-96,0.70+.--6%&$&.,*04$260.9*6,26)0&4$&0&460>.26,%-6%&O*0&,$)'&'.%$50N:657$,60,.56O0
M0 `.4%0 E,$'&4:$'&6/0 F$,30 F.>5'$%6*60 $%)0 ]$2')0 G62'(H$8,/0 NF$%0 S6>85$&'.%0 $%)0 V.26,%$%+60 A$D60 $0
P0 A$,&'%0 A'%.>86/0 NV.26,%$%+6(E$*6)0 @%$53*'*0 .70 S6>85$&'.%O0 I!^^!J0 Q#ILJ0 :##)*,. "6. =(?*+$. )#0.
!""&%7)'+;%. 1$"#"2+$,0 PLYK0 A$,&'%0 A'%.>860 $%)0 G6)'2'%$0 R0 F$,'%./0 3%-(*)'"79. G";%7#)#$%. +#.
Q0 ]$2')0G62'(H$8,/0NW460V5.9$50]'778*'.%0.70S6>85$&.,30F$1'&$5'*-O0I!^^MJ0MYT0:##)*,."6.'E%.:2%7+$)#.
T0 ?&626%0c6%&0R.>65/0J7%%7.8)7D%',H.8"7%.3(*%,4.3%-(*)'"79.3%6"72.+#.:0;)#$%0.5#0(,'7+)*.!"(#'7+%,0
IF.,%6550 d%'26,*'&30 Z,6**/0 "YYPJ;0 E,$'&4:$'&6/0 $9.260 %0 Q/. *'-'5$,530 )6*+,'96*0 &460 ,6+'1,.+$50 I$%)0

Electronic copy available

Electronic copyat:available
*,- !"#$%&'(#)*+,%)-.'/$0'1+*)+0!"#$%!&'(!)$!*+

B9! 678429! 2$! ;0%3B60(! 0>1%$C>0/2(! 0@?.B27$/(! 3$$@(! 4$?97/8(! B@0D?B20! >0@7.B%!
M/$2406! 240$6C! 79! 24B2! ;0! %7=0! 7/! B! 5679I! 9$.702C:! ;4060! B@=B/.09! 7/! 9.70/.0!
B/@! 20.4/$%$8C! 4B=0! .60B20@! /0;! 679I9! 24B2! 60D?760! 910.7B%792! >B/B80>0/2! "08!
<7$20.4/$%$8CE! .40>7.B%! 0N1$9?60E! >$<7%0! 14$/0! 6B@7B27$/E! /?.%0B6! 1$;06E!
A! B.2?B67B%(! 9$.7B%! B/@! 1$%727.B%! A! B/@! .$>>0/2! 7/978423?%%C! 24B2! 608?%B2$6C!
S02! $2406! .$>>0/2B2$69! B68?0! 24B2! 240! 7/.60B90! 7/! 608?%B27$/! >BC! 7/! 1B62! <0!
0N1%B7/0@! <C! B! @0.60B90! 7/! 240! 1?<%7.:9! 26?92! 7/! 9$.7B%(! 0.$/$>7.! B/@! 1$%727.B%!

F! UB99!V!W?/9207/(!23-+,'-&+'1)4&-%'1+*"56-)"#7'1+8"#8+)*)#4'-&+'1+465$-",.'9-$-+!"QB6=B6@!X/7=06972C!
,J! WB>! G0%2H>B/(! 1+465$-)"#' $#:' -&+' ;$-6,$5' <,"4,+%%' "3' =>65+#8+! "MYZP[6$$I7/89! \$7/2! U0/206! 3$6!
,,! V$<062!K?22/06(!A*+,.-&)#4'3",'9$5+7'B&+'C),-6+%'$#:'/)@)-%'"3'!$,D+-%!"M%360@!M!K/$13(!,FF]+!*'*L!
,*! X%67.4! [0.I(! 1)%D' 9"8)+-.7' B"0$,:%' ;+0' !":+,#)-.! "WB80! G?<%7.B27$/9(! ,FF*+E! U46792$1406! Q$$@(!
Q0/6C! V$249207/! B/@! V$<062! [B%@;7/(! B&+' E"*+,#@+#-' "3' 1)%D7' (#:+,%-$#:)#4' 1)%D' 1+465$-)"#'
,&! V08?%B27$/!_B9I3$6.0(!B<$=0!/!,E!_$/C![%B76(!5U$>10/9B27$/!U?%2?60:!"W100.4!@0%7=060@!B2!240!Z/9272?20!
$3! G?<%7.! G$%7.C! V090B6.4(! X/7=06972C! U$%%080(! `$/@$/(! *T! aBC! *JJR+! b4221cdd;;;L8?B6@7B/L.$L?Id
,-! U46792$1406! Q$$@! B/@! aB627/! `$@80(! 5GB=%$=7B/! Z//$=B27$/(! G02! W$%?27$/9! B/@! Y.$/$>7H7/8! $/!
1+465$-",.' F##"*$-)"#7' 2' G"@>$,$-)*+' 2#$5.%)%! "Y@;B6@! Y%8B6(! *JJR+! ,&'E! Q$$@(! V$249207/! B/@!
[B%@;7/(! B<$=0! /! ,*(! -^RE! O8?9(! 1+465$-)"#7' /+4$5' ?",@' $#:' A8"#"@)8' B&+",.(! B<$=0! /! ,J(! &&'E!
O68B/79B27$/! 3$6! Y.$/$>7.! U$P$106B27$/! B/@! g0=0%$1>0/2(! =AGH' 1+*)+0%' "3' 1+465$-",.' 1+3",@(!
,R! h7$/B!QB7/09(!M@B>!W?22$/!B/@!U4679!G%B2B/7BPG4?/8(!5Z2:9!M%%!M<$?2!V79I(!Z9/:2!Z2f!W.70/.0(!G$%727.9(!
G?<%7.! O17/7$/! B/@! V08?%B2$6C! V03$6>:! "*JJ]^J'+! ,J! ?5)#:+,%' I"6,#$5' "3' /$0' 1+3",@! -&RE! h7$/B!
,T! \B.7/2!\$6@B/B!B/@!gB=7@!`0=7PhB?6(!5_40!G$%727.9!$3!V08?%B27$/!7/!240!M80!$3!i$=06/B/.0:!7/!\B.7/2!
!"#$%&'()*+,+#-./"$+)*.)#0.1$"#"2+$.3%-(*)'+"#4.52&*+$)'+"#,.6"7.8"0%7#.3%-(*)'"7,.)#0. !"#

5(3-.233+ &,*2.+ %23-3*3+ .28+ %2=(10*-&.+ &.+ 2)&.&7-)+ =%&(.'3+ E+ 0%=(-.=+ *60*+ -*+
70>23+ *627+ 1233+ )&7$2*-*-/2+ -.+ 0.+ -.)%203-.=1B+ =1&501+ 70%>2*+ E+ 0.'+ )011-.=+
G6(3+ =&/2%.72.*3+ !+.'+ *627321/23+ 3-7(1*0.2&(31B+ 52-.=+ 03>2'+ *&+ 30,2=(0%'+ 0.+
,0-%.233?+ L*+ -3+ .&+ 8&.'2%+ *60*+ 7&'2%.+ =&/2%.72.*3+ !+.'+ *627321/23+ 9)&.3*0.*1B+
03>2'+ *&+ '&+ -*+ %2K(-%2+ (3+ *&+ (.'2%3*0.'+ *62+ .&*-&.3+ &,+ 92)&.&7-);+ 0.'+ 93&)-01;+
&,+ 03+ )&.*%03*-.=+ $&1-)B+ $0-%3+ 8-*6+ 2)&.&7-)+ %2=(10*-&.+ '23-=.2'+ *&+ -7$%&/2+
'23-%0512+ &(*)&723+ 2-*62%+ 5B+ )&%%2)*-.=+ ,&%+ *62+ '070=-.=+ 2,,2)*3+ &,+ 2)&.&7-)+
5B+ .&.<70H&%-*0%-0.+ -.3*-*(*-&.3+ 3*0,,2'+ 5B+ *2)6.-)01+ 2:$2%*3+ 320%)6-.=+ ,&%+

"@+ O=(34+3%-(*)'+"#4.<%-)*.="72.)#0.1$"#"2+$.>?%"794+05&/2+.+"I4+#"PQJ+R05%-S-&+T-10%'-4+U0)-.*+U&%'0.0+
"C+ O,+ )&(%32+ 5(3-.233+ )0.+ )6&&32+ *&+ $033+ *62+ 0''-*-&.01+ )&3*+ *6%&(=6+ *&+ )&.3(72%3+ -.+ 6-=62%+ $%-)23+ -,+
"F+ M224+2=4+_(3-.233+`&(.)-1+&,+D(3*%01-04+9_(3-.233+X2=(10*-&.+D)*-&.+a10.+,&%+R(*(%2+a%&3$2%-*B;+[X2$&%*4+
!I+ _(3-.233+ `&(.)-1+ &,+D(3*%01-04+ 9X2360$-.=+D(3*%01-0;3+ R2'2%0*-&.Y+D+ c28+ `&.*%0)*+ ,&%+ R2'2%01<M*0*2+
X210*-&.3;+ [X2$&%*4+ O)*&52%+ !II]\J+ _(3-.233+ `&(.)-1+ &,+ D(3*%01-04+ 9G&80%'3+ 0+ M2071233+ N)&.&7BY+
!"+ M*2/2.+d0.+'2+A01124+9L.*2%.0*-&.01+`&7$0%-3&.3+&,+a(51-)+M2)*&%+a2%,&%70.)2;+[!IIF\+""["\+@(D*+$.
!!+ G62%2+ 0%2+ 70.B+ 80B3+ &,+ '-,,2%2.*-0*-.=+ 52*822.+ *B$23+ &,+ %2=(10*-&.?+G6-3+ 0%*-)12+ ,&)(323+ &.+ *8&+ E+
166%$'+;%. 3+,A. 3%-(*)'+"#+ [Z0%/0%'+ e.-/2%3-*B+ a%2334+ "FFQ\\+ 0.'+ 9-.*2=%-*B;+ %2=(10*-&.+ '23-=.2'+ *&+
*,- !"#$%&'(#)*+,%)-.'/$0'1+*)+0!"#$%!&'(!)$!*+

0=74703.418.>! @?! 67A$41374172! 128313931$28! 12! B=10=! 0$2"!10312<! 123.4.838! 74.!>!72>!5$%10?!$930$6.8!2.<$3173.>:*&!C2>.5.2>.23!4.<9%73$48!$D.48..12<!
.0$2$610! 4.<9%731$2! 0$23129.! 3$! 54$D1>.! 7! 98./9%! /476.B$4E! B13=12! B=10=! 3$!
67?! 2$3(! B=.3=.4! 3=.4.! 18! 72! 7%3.42731D.! B7?! $/! 0$20.5397%1812<! .0$2$610! 72>!
%7@.%8! 18! 753! 3$! 0$2/98.:! S2! 7%3.42731D.! 6$>.%! 18! 3=.2! 89<<.83.>! 12! B=10=! 7%%!
>.81<2.>! 3$! 70=1.D.(' B=1%.! 4.0$<21812<! 3=.! 165$43723! 4$%.! 3=.! 54.8.20.! $/! 3=.!

*&! P$2?! R4$88.4(! 2&+' 1+345$-",.' 6#-+,7,)%+! "JF/$4>! U21D.4813?! R4.88(! *V,V+G! W172>$6.210$! X7A$2.(!
*H! ;..!/$4!.F765%.!3=.!S98347%172!\2.4<?!Y.<9%73$4!"]3=+G!C2>.5.2>.23!R41012<!72>!Y.<9%73$4?!P41@927%!
$/! 3=.! P78672172! \0$2$610! Y.<9%73$4G! \88.2317%! ;.4D10.8! ]$661881$2! "#10+G! U31%131.8! ]$661881$2!
*M! Q13=! 4.85.03! 3$! JK;! 8..! /$4! .F765%.! ]$6074.! B=10=! 18! B13=12! 3=.! ]$66$2B.7%3=! _.57436.23! $/!
\>90731$2(! \65%$?6.23! 72>! Q$4E5%70.! Y.%731$28G! Q$4E;7/.!S]P! B=10=! 18! B13=12! 3=.! _.57436.23!
$/! `98310.! 72>! ]$669213?! ;7/.3?G! Q$4E5%70.! K.7%3=! 72>! ;7/.3?! ^9..28%72>! B=10=! 18! B13=12! 3=.!
*-! K7272!K7@.4(!IY.<9%7312<T/$4TQ.%/74.a!S!]$6574731D.!;39>?!$/!bY.<9%73$4?!Q.%/74.!Y.<16.8c!12!3=.!
*O! ;39743!f.%%8!72>!S41.!e4.1@.4<(!I\0$2$610!Y.<9%731$2L!12!S41.!e4.1@.4<!".>+(!2&+'2""5%'";'1+345$-)"#!
!"#$%&'()*+,+#-./"$+)*.)#0.1$"#"2+$.3%-(*)'+"#4.52&*+$)'+"#,.6"7.8"0%7#.3%-(*)'"7,.)#0. !"#

9'/,&)-:( )01( 9.,/0/;&,:( 2%.0( )<<-&.1( +/( +%.( +.*;( 9*.5=-)+&/0:>( $%&'( *)&'.'( +%.(
1&33.*.0+( <./<-.:( 2&+%( 1.!(0&+&/0'( 8)*4&05( ),,/*1&05( +/( <*/3.''&/0)-( 1&',&<-&0.F(
</-&+&,)-( &1./-/54( )01( .8.0( 5./5*)<%4>!G( H.!(0&+&/0'( 8)*4( 3*/;( +%.( -.5)-&'+&,(
+%.( 8)*&/='( ,/0,.<+&/0'( /3( *.5=-)+&/0>( L/*( &'( &+'( <=*</'.( +/( '..M( +/( <*/8&1.( )(
+%)+( 9&+( 2/=-1( 6.( 3=+&-.( )01( '/;.2%)+( 0/0'.0'&,)-( +/( /33.*( /0.( )=+%/*&+)+&8.(
'.0'.FO"( )01( +%)+( +%.( 9'<.,&!(,( ,/0+.I+( )01( 5/)-( S'%/=-1T( '%)<.( +%.( <)*+&,=-)*(
K0( +%&'( )*+&,-.( 2.( )*.( ,/0,.*0.1( +/( =01.*'+)01( +%.( &;<-&,)+&/0'( 3/*( ;/1.*0(
*.5=-)+/*'( )01( *.5=-)+/*4( ),+&8&+4( /3( +%.( ,%)05.'( +%)+( %)8.( 6..0( +)M&05( <-),.(
&0( 2%)+( 5/8.*0;.0+'( 1/F( )01( %/2( +%.4( 1/( &+>( U&+%( +%&'( '+)+.( ,.0+*.1( <=*</'.(
&0( ;&01F( 2.( %)8.( ,%/'.0( +/( 1.!(0.( *.5=-)+&/0( 64( 6=&-1&05( =</0( +%.( 1.!(0&+&/0(

!G( C.8&DE)=*F( 9Q.5=-)+&/0( V( Q.5=-)+/*4( W/8.*0)0,.:F( )6/8.( 0( !!F( OXY>( C.8&DE)=*( 0/+.'( +%)+( -.5)-(
!J( Q/6.*+( R)-12&0F( [/-&0( \,/++( )01( [%*&'+/<%.*( ]//1F( :. 3%)0%7. "#. 3%-(*)'+"#( ^_I3/*1( `0&8.*'&+4(
cZ( C.8&DE)=*F( 9Q.5=-)+&/0( V( Q.5=-)+/*4( W/8.*0)0,.:F( )6/8.( 0( !!F( OXYZ(B*&.( E*.&6.*5F( =>%. =""*,. "6.
OP( N/*1)0)()01(C.8&DE)=*F(9$%.(a/-&+&,'(/3(Q.5=-)+&/0:F()6/8.(0("YF(O>
O"( R-),MF(9[*&+&,)-(Q."(.,+&/0'(/0(Q.5=-)+&/0:F()6/8.(0(!JF(!e>(\..()-'/(N=-&)(R-),MF(9U%)+(&'(Q.5=-)+/*4(
K00/8)+&/0A:( &0( N=-&)( R-),MF( d)*+&0( C/15.( )01( d)*M( $%)+,%.*( ^.1'bF( 3%-(*)'"79. 5##";)'+"#4. :.
O!( N/*1)0)()01(C.8&DE)=*F(9$%.(a/-&+&,'(/3(Q.5=-)+&/0:F()6/8.(0("YF(c>
OO( R-),MF(9[*&+&,)-(Q."(.,+&/0'(/0(Q.5=-)+&/0:F()6/8.(0(!JF(!Y>
*,' !"#$%&'(#)*+,%)-.'/$0'1+*)+0!"#$%!&'(!)$!*+

-.! /%0$! /11! 23.! 45/%6!!.7! 23/2! 23.! 87$9.00! 60! $:.! 5:1.72/;.:! <=! $7! 5:1.7! 23.!
/50869.0! $7! /523$762=! $>! ?$@.7:A.:2(! 2350! 7.2/6:6:?! 23.! 02/2.! /0! 23.! 0$579.! $>!
7.?5%/2$7=! 2$$%0! /2! 620! 1608$0/%! /0! D.%%! /0! 9$E! /:1! 0.%>E7.?5%/2$7=! 7.?6A.0! 23/2!
C360! 1.!!:626$:! 3/0! /! :5A<.7! $>! 6A8$72/:2! >./257.0! /:1! /1@/:2/?.0B! I6702(! /0!
C3.! %/D! >/96%62/2.0! 23.! .:267.! 7.?5%/2$7=! .:1./@$57! <=! 97./26:?! 23.! >7/A.D$7;!
D6236:! D3693! 62! 9/:! .>>.926@.%=! 2/;.! 8%/9.N! >7$A! 9$:0275926:?! /:1! 9$:027/6:6:?!
.:>$79.A.:2! /:1! 160852.! 7.0$%526$:B&P! J$7.$@.7(! 23.! %/D! 7.A/6:0! /! 16026:926@.!
06R! 9/2.?$76.0! $>! 7.?5%/2$7=! 2$$%0N! ",+! .9$:$A69! 2$$%0! "0593! /0! 237$5?3! A/;6:?!
"0593! /0! 7.?6027/26$:(! %69.:06:?! $7! /997.162/26$:+O! "H+! 6:>$7A/26$:/%! 2$$%0! "0593!
/0! 87$1592! %/<.%%6:?! $7! 9$:26:5$50! 1609%$057.! 7.?6A.0+O! "M+! 02759257/%! 2$$%0! "$>!
83=069/%! 1.06?:(! $7! 87$9.00.0! 0593! /0! $57! STKU! 2/R! /77/:?.A.:20+O! /:1! "P+!
%.?/%! 2$$%0! "0593! /0! %/D0(! 75%.0! /:1! 7.?5%/26$:0+B&'! #6.D.1! 2360! D/=(! 7.?5%/26$:!

&H! W/:!X/72%.!/:1!S.2.7!#/00(!FL.%>EY.?5%/26$:!/:1!23.!Y.?5%/2$7=!L2/2.N!T!L57@.=!$>!S$%69=!/:1!S7/9269.G!
"Y.0./793! Y.8$72! )$! ,Q(! Z.:27.! >$7! 23.! L251=! $>! Y.?5%/2.1! W:150276.0(! [:6@.7062=! $>! X/23(! \92$<.7'
&M! J$7?/:!/:1!K.5:?(!/<$@.!:!*V(!HB!
&P! I7.6<.7?(!/<$@.!:!*V(!,Q'O!J$7?/:!/:1!K.5:?(!/<$@.!:!*V(!&_QB!J$7?/:!/:1!K.5:?!/%0$!61.:26>=(!6:!
&Q! J$7?/:!/:1!K.5:?(!/<$@.!:!*V(!MO!I7.6<.7?(!/<$@.!:!*V(!,Q'O!).6%!U5::6:?3/A!/:1!S.2.7!U7/<$0;=(!
&'! I7.6<.7?(!/<$@.!:!*VB!L..!/%0$!S/7;.7!/:1!X7/623D/62.(!/<$@.!:!*VB!
&V! X/%1D6:(!L9$22!/:1!`$$1(!/<$@.!:!*V(!HO!S.2.7!a!J/=(!FL$96/%!Y.?5%/26$:G!6:!b.02.7!J!L/%/A$:!".1+(!?&+'
!"#$%&'()*+,+#-./"$+)*.)#0.1$"#"2+$.3%-(*)'+"#4.52&*+$)'+"#,.6"7.8"0%7#.3%-(*)'"7,.)#0. !"#

$69' 9.42*&+.6,' 4)*5.//0' $69' ,-$,' ).:12$,3*6' 5$6' ?.' /-$4.9' $69' @5*;4)*915.9A'
C*1),-0' 169.)' ,-3/' 9.!'63,3*6' ).:12$,3*6' 3/' 41)4*/3<.0' /1/,$36.9' $69' (*51//.9B'
$69' 9./3:6.9' ,*' /*2<.' $' 4$),3512$)' 4)*?2.+' *)' 4)*915.' $' 4$),3512$)' *1,5*+.B7!'
F*).*<.)0' 3,' 3/' 6*,' $5-3.<.9' ?&' /3+42&' 4$//36:' $' 2$%B' E,' ).G13)./' *6:*36:'
+*63,*)36:' *(' ,-.' ).2.<$6,' $5,3<3,3./0' 5*6,361$2' =).>$//.//+.6,' *(' <$21./' $69'
L69' !'(,-0' ,-.' 9.!'63,3*6' 42$5./' /*+.' ).$/*6$?2&' 52.$)' ?*169$)3./' $)*169' ,-.'
5*65.4,' *(' ).:12$,3*6' ,-$,' 93((.).6,3$,.' 3,' ()*+' ,-.' *,-.)' 4)36534$2' $5,3<3,3./' *('
:*<.)6+.6,' =4)*<3936:' $69' 93/,)3?1,36:>0' $69' 4).<.6,' 3,' ?.5*+36:' ,**' ?)*$9'
=?&' ?.5*+36:' 5*++.6/1)$,.' %3,-' ,-.' .6,3).' 2.:$2' /&/,.+>' *)' ,**' $?/,)$5,' *)'
$+*)4-*1/' =3(' 36521/3<.' *(' $22' +.5-$63/+/' *(' /*53$2' 5*6,)*2' *)' 36"'1.6536:'


M$<36:' $9*4,.9' ,-3/' 9.!'63,3*6' *(' ).:12$,3*60' ,-.' 6.K,' G1./,3*6' 3/' -*%' ,*' ?./,'
">' ,-.'$5,*)'*)'$5,3<3,3./'?.36:').:12$,.907Q
!>' ,-.'6$,1).'*(',-.').:12$,*)&'36/,)1+.6,/'*)',**2/'.+42*&.907R'*)

78' S123$'T2$5U0'@V.5.6,)36:'D.:12$,3*6P'W69.)/,$6936:',-.'D*2.'*('D.:12$,3*6'$69'O.2(;D.:12$,3*6'36'$'
XY*/,;D.:12$,*)&Z'[*)29A' =!88">' Q7=">' !(77%#'. <%-)*. =7">*%2,' "8I\' T2$5U0' @])3,35$2' D."'.5,3*6/' *6'
7"' ^-.'36<*2<.+.6,'*('6*6;/,$,.'$5,*)/').36(*)5./',-.'3+4*),$65.'*('$6'.((.5,3<.'/,$,.'4)./.65.'36'.6/1)36:'
,-$,' ).:12$,3*6' (1),-.)/' 41?235' 41)4*/./' $69' 6*,' ,-.' /4.53$2' 36,.)./,/' *(' ,-.' 6*6;/,$,.' $5,*)/' ,-)*1:-'
7!' c*,.',-$,'$'41)4*/3<.'9.!'63,3*6'$2/*'-$/'3,/'93/$9<$6,$:./B'C3)/,0'3,'5$6'?.'93(!'512,',*'52.$)2&'93/5.)6',-.'
D.:12$,*)&' O4$5.A' 36' `.3:-' M$65-.)' $69' F35-$.2' F*)$6' =.9/>0' !)&+')*+,2@. !(*'(7%. )#0. 1$"#"2+$.
3%-(*)'+"#' =]2$).69*6' Y).//0' "#e#>' !f"0' !#Ibf\' g)35' [369-*2d0' @g<$21$,36:' ,-.' M$)+*63/$,3*6' *('
L1/,)$23$A/' aMO' `$%/P' ]-$22.6:./' $69' a44*),163,3./A' =!8"8>' I!=!>' :,+). =)$+!.$. A"(7#)*. "6. =(>*+$.
7I' ^*6&'Y)*//.)0'@D.:12$,3*6'$69'O*53$2'O*239$)3,&A'=!88R>'II=I>'A"(7#)*."6.<)?.B./"$+%'9'IR70'IfQ\'Y-3234'
77' C).3?.):0'$?*<.'6'!#0'I\'_36:/(*)9'O+3,-0'$?*<.'6'7"0'77QbRB
7Q' T2$5U0'@])3,35$2'D."'.5,3*6/'*6'D.:12$,3*6A0'$?*<.'6'!#0'"#\'O16/,.360'$?*<.'6'#0'!7IB
7R' C).3?.):0'$?*<.'6'!#B
**, !"#$%&'(#)*+,%)-.'/$0'1+*)+0!"#$%!&'(!)$!*+

&+! -./!0120$3/!$4!-./!2/51%6-$27!68-9:9-7!9;!-/2<3!$4!-./!02$=%/<3!9-!93!>/395;/>!
B;! -.93! 62-98%/! C/! .6:/! 8.$3/;! -$! >93-9;5193.! =/-C//;! D3$896%E! 6;>! D/8$;$<98E!
=/3-! 2/!!/8-3! -./! 8$;-/F-16%! 6;>! 0120$39:/! ;6-12/! $4! 2/51%6-9$;! <6G9;5! H! -.6-!
68-9:9-7! =63/>! >/"!;9-9$;3! 6%3$! 469%! 9;! -.6-! -./7! >$! ;$-! 68G;$C%/>5/! -.6-! 3$896%!
P6:9;5! 6>$0-/>! 0120$39:/! 829-/296! 4$2! >944/2/;-96-9;5! 3$896%! 2/51%6-9$;! 42$<!

A Concept of Economic Regulation

M;! /F6<9;6-9$;! $4! -./! %9-/26-12/! 2/:/6%3! 6! 5/;/26%%7! C/%%L>/:/%$0/>! 6;>!
8$<0/-9-9$;?@R! K18.! 6;6%7393! .63! 5/;/26%%7! =//;! 8$;393-/;-! 6=$1-! 9-3! 0120$3/?!
S70986%! 93! -./! 4$%%$C9;5! >/"!;9-9$;! $4! D/8$;$<98! 2/51%6-9$;E! 13/>! =7! -./!
U8$;$<98! 2/51%6-9$;3! 9;-/2:/;/! >92/8-%7! 9;! <62G/-! >/8939$;3! 318.! 63!
02989;5(! 8$<0/-9-9$;(! <62G/-! /;-27(! $2! /F9-?! Y/4$2<! 69<3! -$! 9;82/63/!
$4-/;! -.2$15.! >/2/51%6-9$;! 6;>! 13/! $4! /4"!89/;87L02$<$-9;5! 2/51%6-9$;(!
6;>! =7! 9<02$:9;5! 2/51%6-$27! 426</C$2G3! 4$2! <62G/-! 41;8-9$;9;5! 6;>!

@A! I6%>C9;(!K8$--!6;>!P$$>(!6=$:/!;!*R(!@[\!]67(!6=$:/!;!&R(![ZA?
@'! I%68G(!DV29-986%!Y/!!/8-9$;3!$;!Y/51%6-9$;E(!6=$:/!;!*R(!*^?!K//!6%3$!_2/9=/25(!6=$:/!;!*R(!@?
@R! I218/!M! `9%%96<3! 6;>!M%=/2-! Y! ]6-./;7(! 2+3"4,$4.5' 2)$6"78+5' $#9' :#*),"#3+#-$6' 2)%;8-+%<' =&+'
Z,! T256;936-9$;!4$2!U8$;$<98!V$L$0/26-9$;!6;>!W/:/%$0</;-(!=&+'A:>2'1+;",-'"#'1+786$-",.'1+?",3!
"[RRA+! ^?! K//! 6%3$X! T513(! 1+786$-)"#<' /+7$6' B",3' $#9' :4"#"3)4' =&+",.(! 6=$:/! ;! [,\! I6%>C9;! 6;>!
!"#$%&'()*+,+#-./"$+)*.)#0.1$"#"2+$.3%-(*)'+"#4.52&*+$)'+"#,.6"7.8"0%7#.3%-(*)'"7,.)#0. !!"

(&4%.4+1) $%**+$0(&,) 5%*) '/*8+0) 5/(.-*+1) %*) ('7+*5+$0(%&1) 09/0) *+3-$+) +$%&%'($)
(&) '/*8+0) 2+9/4(%-*>?!) A0) 1++1) 09+) '/*8+0) /1) 09+) 2+10) /4/(./2.+) '+$9/&(1') 5%*)
09+) +5!)$(+&0) 7*%3-$0(%&) %5) ,%%31) /&3) 1+*4($+1) /&3) 5%*) 09+(*) +5!)$(+&0) /..%$/0(%&)
2+0:++&) '+'2+*1) %5) 09+) $%''-&(06) 1%) /1) 0%) '/=('(1+) 1%$(+06B1) :+/.09>?C)

B Concept of Social Regulation


?") K++;)+,;)L,-1;)3%-(*)'+"#4.<%-)*.="72.)#0.1$"#"2+$.>?%"79;)/2%4+)&)"MJ)N/.3:(&)/&3)O/4+;)/2%4+)&)
?!) L*,/&(1/0(%&) 5%*) #$%&%'($) O%T%7+*/0(%&) /&3) U+4+.%7'+&0;) @1!A. 3%;+%B,. "6. 3%-(*)'"79. 3%6"72;)
?C) R%4+*&'+&0)%5)S($0%*(/;)/2%4+)&)?M;)1)!>">
?H) #-,+&+) N/*3/$9) /&3) D%2+*0) W) X/,/&;) YA&0*%3-$0(%&B) (&) #) N/*3/$9) /&3) D) W) X/,/&) F+31G;) /"$+)*.
3%-(*)'+"#4. /'7)'%-+%,. 6"7. 3%6"72) FA&10(0-0+) 5%*) O%&0+'7%*/*6) K0-3(+1;) "PZ!G! CJ) W&09%&6) L,-1;)
??) @9+).(0+*/0-*+)*+4+/.1)/)7%0+&0(/.)09(*3)$%&$+70-/.(1/0(%&)%5)1%$(/.)*+,-./0(%&;)&/'+.6)*+,-./0(%&)3+1(,&+3)
X) K7/**%:;) >?%. 3%-(*)'"79. !7)6'4. !"#'7"**+#-. 3+,E,F. /"*;+#-. C7"G*%2,F. )#0. 8)#)-+#-. !"2&*+)#$%)
b*2/&) A&10(0-0+) a*+11;) "PZPG! "Pc>) Q%:+4+*;) :+) 3%) &%0) $%&1(3+*) 09(1) 0%) 2+) /) 1+7/*/0+) $/0+,%*6>) _%*)
Y7*%0+$0(4+T*+,-./0(%&) /,+&$(+1B) (&0+*$9/&,+/2.6;) ^+4(T_/-*;) YD+,-./0(%&) d) D+,-./0%*6) R%4+*&/&$+B;)
/2%4+) &) !!;) "C>) a*%0+$0(4+) *+,-./0(%&) $/&) 2+) $/0+,%*(1+3) /1) +(09+*I) F/G) $%**+$0(&,) 5%*) 09+) 3/'/,(&,)
*** !"#$%&'(#)*+,%)-.'/$0'1+*)+0!"#$%!&'(!)$!*+

1 Correcting for the Damaging Effects of Economic Activity

(Market Failures)
,-./! 0$112.3-3$45! 62!!.2! 5$07-%! 4289%-37$.! 7.! 32415! $:! 0$442037.8! :$4! 3;2!
E5F$07-%! 4289%-37$.! 32.65! 3$! $@24-32! -04$55! -%%! 5203$45! $:! 3;2! 20$.$1/!
3;-3! %-7552PQ:-742! :$4025! B$9%6! $3;24B752! @4$6902R=IS! T;252! 7.0%962! @$%%937$.(!
1-4H23! :-7%9425! 3;-3! 87<2! 4752! 3$! 3;252! 6-1-87.8! 2::2035(! 590;! -5! 2?324.-%73725(!
8$$6(! 0$%%2037<2! -037$.! -.6! $3;24! 0$Q$467.-37$.! @4$U%215=JV! W3! -%5$! 0$442035! :$4!

IJ! A-%6B7.(!C0$33!-.6!D$$6(!-U$<2!.!*S(!YX=
IL! Z2324! [%2-4/! K2-824(! 2&+' /)3)-%' "4' /$05' 2&+' 6789):' 1+;79$-)"#' "4' 6,)*$-+' 6"997-)"#! "[-1U47682!
I'! N273;!D-BH7.5!-.6!A476823!,!D93324(!OT;2!]25@$.52!$:!A957.255!3$!C$07-%!]289%-37$.!7.!^.8%-.6!-.6!
IS! N933.24(!-U$<2!.!XX(!*'X=!
JV! b895(!1+;79$-)"#5'/+;$9'=",3'$#>'?:"#"3):'2&+",.(!-U$<2!.!XV(!YcId!b895(!O_";+73;24!3;2!^0$.$170!
T;2$4/! $:! ]289%-37$.eR(! -U$<2! .! IY(! &*d! N933.24(! -U$<2! .! XX(! **Lc'd! f7-.6$12.70$! ,-g$.2(! OT;2!
JX! A-46-0;(!-U$<2!.!II(!XS'=!C22!-%5$!b895(!O_";+73;24!3;2!^0$.$170!T;2$4/!$:!]289%-37$.eR(!-U$<2!.!IY(!
J*! Z4$5524(!2&+'1+;79$-",.'?#-+,D,)%+(!-U$<2!.!*&(!X=
J&! ,-/! 7.! :-03! 62!!.25! 20$.$170! 4289%-37$.! 3$! 7.0%962! 12-59425! 62578.26! 3$! -<$76! 0$.59124! -.6! $3;24!
!"#$%&'()*+,+#-./"$+)*.)#0.1$"#"2+$.3%-(*)'+"#4.52&*+$)'+"#,.6"7.8"0%7#.3%-(*)'"7,.)#0. !!"



Social Regulation

#$%&' ()*(+,-' -.+/-&' &)(%/0' .+120/-%)*' /&' /' -+($*%(/0' %&&2+' %*&20/-+3' 4.)5'
3%&-.%62-%7+' )68+(-%7+&9' :-' .+8+(-&' -$+' *)-%)*' -$/-' %*-+.7+*-%)*&' 6;' ,+.&)*&' /*3'

2 Attaining Socially Desirable Outcomes

#$+' &+()*3' /*3' ),,)&%*1' ()*(+,-' )4' &)(%/0' .+120/-%)*' 3+!'*+&' %-' %*' -+.5&' )4'
/--/%*%*1' (+.-/%*' &)(%/00;' 3+&%./60+' )2-()5+&F' )4' 5++-%*1' ()00+(-%7+' 3+&%.+&' ).'
,.)32(+3' 6;' /*' +4!'(%+*-0;' ),+./-%*1' 5/.<+-' +()*)5;9DD' #$+' &)(%/00;' 3+&%./60+'
)2-()5+&' .+"'+(-' 6.)/3+.' &)(%+-/0' 7/02+&' &2($' /&' 82&-%(+=' +>2%-;' /*3' 4/%.*+&&=DH'

DE' M.)&&+.=' <=%. 3%-(*)'"79. 1#'%7&7+,%=' /6)7+' *' !"=' N9' O++' /0&)' P+7%@Q/2.=' BR+120/-%)*' S' R+120/-).;'
DG' O2*&-+%*='/6)7+'*'J='GHUDV9
DD' M.)&&+.='BR+120/-%)*'/*3'O)(%/0'O)0%3/.%-;C='/6)7+'*'E"='"HGF'W).1/*'/*3'X+2*1='/6)7+'*'!J='!J9
DH' Y)$*' R/?0&=' :. <=%"79. "6. >(,'+$%' KZ[4).3' \*%7+.&%-;' M.+&&=' NJHNLF' ].-$2.' W' Z<2*=' 1?()*+'9. )#0.
DI' M.)&&+.='BR+120/-%)*'/*3'O)(%/0'O)0%3/.%-;C='/6)7+'*'E"9'
DJ' T2**%*1$/5'/*3'O%*(0/%.='/6)7+'*'ND='IHVU!F'Q.+%6+.1='/6)7+'*'!J='N"UND9
**, !"#$%&'(#)*+,%)-.'/$0'1+*)+0!"#$%!&'(!)$!*+

-.!!/0-12! 3456-2! -/$1$30-789:! ;$/04%! 5-<=%420$1(! 542>-5! 2>41! ?-01<! 4! 7=?7-2!

$.! -/$1$30/! 5-<=%420$1(! 41@! @-A-1@-12! $1! 7$3-! .$53! $.! 3456-2! .40%=5-! .$5! 027!


Social Regulation


$1! 4! A=5-%G! =20%0245041! ?4707(! 41@! 2>42! A501/0A%-7! $.! 01@0C0@=4%! B=720/-(! -J=02G!

9:! )-0%!O=1101<>43(!2$3+45$,6)#4'-&+'7",8+,9':";'<$=$,6%'$#6'-&+'1">+'"3'-&+'/$0!"P4E!Q$$6(!MR',+!
9M! W4E%7(!4?$C-!1!T9(!MXU!Y!P!F!Y452(!ZQ-2E--1![20%02G!41@!W0<>27\!"MR9R+!9R!?">5@;)$'/$0'1+*)+0!'*'8
9*! O=1101<>43(!4?$C-!1!9:(!*R&U!;=172-01(!4?$C-!1!R(!&RU!F12>$1G!]<=7(!ZW-<=%42$5G!N17202=20$17!41@!
9&! Q5402>E402-(!4?$C-!1!9(!MR'8
!"#$%&'()*+,+#-./"$+)*.)#0.1$"#"2+$.3%-(*)'+"#4.52&*+$)'+"#,.6"7.8"0%7#.3%-(*)'"7,.)#0. !!"



#$%&%' ()&*)+,*)-+&' .%*/%%+' %,-+-0),' 1+(' &-,)12' 3%4521*)-+' 6' 1+(' .%*/%%+' *$%'
%+1.2%<' 8-3' )+&*1+,%<' -+%' *-' 8-,5&' -+' *$%' 81,%*' -8' *$%' ,-+,%7*' *$%9' (%&,3).%' 1+('
81)253%' )4+-3%&' *$1*' *$%3%' ,1+' .%' ,-07%22)+4' &-,)12' 3%1&-+&' 8-3' 4-:%3+0%+*'
73-,%&&' @5+(%3*1B%+' .9' *%,$+-,31*&A' 31*$%3' *$1+' 1&' *$%' 0%%*)+4' -8' ,-07%*)+4'
&-,)12' 1+(' %,-+-0),' :125%&' (%.1*%(' )+' *$%' 7-2)*9;CC' J+(' 1&' >21,B' +-*%&<' %:%+'
,-00%+*1*-3&' 83-0' &*3-+4' %,-+-0),' *31()*)-+&' $1:%' 2-+4' 1,B+-/2%(4%(' *$%'
%=)&*%+,%' -8' 1' 31+4%' -8' +-+I%,-+-0),' 3%4521*-39' 4-12&' &5,$' 1&' ()&*3).5*)-+12'
N)0)21329<' *-' :)%/' &-,)12' 3%4521*)-+' -+29' )+' *%30&' -8' 1**1)+)+4' &-,)1229' (%&)31.2%'
-5*,-0%&<' .%**%3' *$1+' *$-&%' 73-(5,%(' .9' 1+' %8!',)%+*29' -7%31*)+4' 013B%*<' 3)&B&'
)4+-3)+4' -3' 4):)+4' )+&58!',)%+*' /%)4$*' *-' 013B%*' 8-3,%&' 1+(' *$%' )07-3*1+,%' -8'

CD' O1,$' -8' *$%' ,-+,%7*&' $1&' )*&' 8-5+(1*)-+&' )+' )(%-2-49' 1.-5*' *$%' 3%&7%,*):%' 3-2%&' -8' 4-:%3+0%+*&' 1+('
013B%*' %,-+-0)%&<' 1+(' *$%' 0-312' 73%,%7*&' /$),$' &$-52(' 45)(%' -53' &-,)%*9;' P)B%' &-' 01+9' )&&5%&' -+'
C"' N5+&*%)+'01B%&'*$%'7-)+*'*$1*'/$1*',-+&*)*5*%&'1+'%=*%3+12)*9'G5&*)89)+4'3%4521*)-+')&<')+'*$%'!'3&*'721,%<'
CH' N*5()%&' -+' *$%' 42-.12' ()885&)-+' -8' 3%4521*-39' 15*$-3)*)%&' 1+(' -8' 3%4521*-39' ,17)*12)&0' 12&-' %:)(%+,%'
*$1*' 4-:%3+0%+*&' 13%' )+,3%1&)+429' ,$--&)+4' *-' %072-9' 013B%*&' 1&' 3%4521*-39' 0%,$1+)&0&<' N%%<' %4<'
CC' K3-&&%3<'Q_%4521*)-+'1+('N-,)12'N-2)(13)*9R<'1.-:%'+'Dc<'cCcT";'N%%'12&-'V19<'1.-:%'+'cS<'WCW;'
CM' >21,B<'Q`3)*),12'_%"'%,*)-+&'-+'_%4521*)-+R<'1.-:%'+'!S<'S;
**, !"#$%&'(#)*+,%)-.'/$0'1+*)+0!"#$%!&'(!)$!*+

>34-=! $:! -/$1$?0/! :-A6%3;$:2! ;$$%4! /31! <%32! 01! 3/90-501A! 4$/03%! $>B-/;05-4! /31!
3/;050;2! ;$! 3! ?$:-! ;$%-:31;! :-A6%3;$:2! :-A0?-(! $:! ;9-! /:-3;0$1! $.! /$?<-143;$:2!
31=! 01;-:=-<-1=-1/-! $.! -/$1$?0/! 31=! 4$/03%! 53%6-4J! ;93;! 3%%! :-A6%3;0$1! 04! 3!
$;9-:! 931=(! 04! -.!!/0-1/2(! 31=! 0;4! <:0?3:2! <6:<$4-! 04! ;$! 0?<:$5-! ;9-! -/$1$?0/!
:-A6%3;0$1J! 31=! 4$/03%! /$140=-:3;0$14! ";9-! 1--=! .$:! :-A6%3;0$1! ;$! >-! %-A0;0?3;-!
N! ;$! >-! 4--1! 34! =-40:3>%-(! <:$<-:! 31=! 3<<:$<:03;-+! <%3/-! %0?0;4! $1! -/$1$?0/!
E0;9! -/$1$?0/! 53%6-4(! 04! 3;! :04@! $.! >-/$?01A! >-/$?-! 61464;3013>%-(! ;$;;-:01A!
O6/9! 3! <-:4<-/;05-! $.! 4$/03%! 31=! -/$1$?0/! :-A6%3;0$1! 04! :-<:-4-1;-=!
=03A:3??3;0/3%%2! 01! R0A6:-4! &3! 31=! &>(! E0;9! -C3?<%-4! $.! ;9-! <:0?3:2! 53%6-4!
31=! 46<<$:;01A! "0-! 4-/$1=3:2+! 53%6-4! .$:! >$;9! -/$1$?0/! 31=! 4$/03%! :-A6%3;0$1!

FG! S-;-:!T:3>$4@2(!UV$61;-:<:$=6/;05-!W-A6%3;0$1X!"QGGY+!*&"Z+!2#-+,#$-)"#$3'4"5,#$3'"6'-&+'7"8)"3"9.'"6'
'P! R$:!-C3?<%-(!01![64;:3%03(!=-4<0;-!$<010$1!<$%%4!49$E01A!?3B$:0;2!<6>%0/!46<<$:;!.$:!A$5-:1?-1;!3/;0$1!
'Q! R$:! -C3?<%-(! 01! [64;:3%03(! ;9-! K$E3:=! V$3%0;0$1! T$5-:1?-1;X4! 01=64;:03%! :-%3;0$14! :-.$:?4!
"UH$:@V9$0/-4X+! E-:-! =-40A1-=! ;$! ?3@-! ;9-! [64;:3%031! %3>$6:! ?3:@-;! ?$:-! "!-C0>%-! 31=! -.!!/0-1;7!
!"#$%&'()*+,+#-./"$+)*.)#0.1$"#"2+$.3%-(*)'+"#4.52&*+$)'+"#,.6"7.8"0%7#.3%-(*)'"7,.)#0. !!"

!"#$%&'()'*'+,-").'/&#$.)0",1' !"#$%&'(:'*';-,1,4"-'/&#$.)0",1'
23%"4)%5'6,-").'7).$&68'6$33,%0"7&' 23%"4)%5'&-,1,4"-'7).$&68'
&-,1,4"-'7).$&69 6$33,%0"7&'6,-").'7).$&69

Social Values Economic Values

Economic Values


1$"#"2+$.3%-(*)'+"# /"$+)*.3%-(*)'+"#
#$%&'$()*'+,-. /0!)1%-21( ;,.6%1-
34&5-6%6%42 <'%$2-..
72248'6%42 /=,%6(
729%8%9,'+%.& >41%'+)34:-.%42
3:4%1- ?$,.6
>,554$6%2@)A%-) ;,.6%1- /0!)1%-21(
>-1429'$(B) <'%$2-.. 34&5-6%6%42
*'+,-. /=,%6( 72248'6%42
>41%'+)34:-.%42 729%8%9,'+%.&
?$,.6 3:4%1-

6:-$-) %.) 24) 1+-'$) 9%8%9%2@) +%2-) %2) $-@,+'64$() 5$'16%1-) C-6H--2) -1424&%1)
9-1%.%42.) H:%1:) 1'2) C-) $-.4+8-9) 6:$4,@:) -M5-$6%.-) '29) .41%'+) 9-1%.%42.)

G!) >--F) -@F) PL,2F) 'C48-) 2) Q"O) R'N42-F) I?:-) /,$45-'2) 34&&,2%6() S-6H--2) >41%'+) #4+%1() '29) >41%'+)
**' !"#$%&'(#)*+,%)-.'/$0'1+*)+0!"#$%!&'(!)$!*+

,-,%./01,%! 2342$565(! 73/! 0/! 05! 809!!13%/! /$! ,22%.! /$! /:6! ;,<$4! 46=3%,/$4.!
0-! 86!!-0-=! /:6! 7$3-8,4065(! ,-8! 24$@080-=! /:6! 9$3-8,/0$-5! 9$4! /:6! 5/,70%0/.!
,-8! %6=0/0;,1.(! $9! /:6! 46=3%,/$4.! 46=0;6! B! 730%85! 32$-! ,-8! 6E/6-85! 6E05/0-=!
$9! 0;2%01,/0$-5! 9$4! ;$864-! 46=3%,/$45! ,-8! 46=3%,/$4.! ,1/0@0/.>! H/! 05! /$! /:656!
0;2%01,/0$-5! /:,/! /:6! ,4/01%6! -$?! /34-5(! 76=0--0-=! ?0/:! ,! 74069! 6E,;0-,/0$-! $9!


G:6! ;$86%! $3/%0-68! ,7$@6! 24$@0865! ,! @,%3,7%6! ,%/64-,/0@6! 1$-162/3,%! 94,;6!

A Utility Regulation Traditional Economic Regulation

K! 3/0%0/.! 46=3%,/$4! 05! ,! 1%,5501! 6E,;2%6! $9! ,! 46=3%,/$4! /4,80/0$-,%%.! %,76%%68! ,5!
C61$-$;01D>! G:6.! ,46! 8650=-68! /$! 24$/61/! ,=,0-5/! ,73565! $9! ;$-$2$%.! 2$?64!
7.! 468310-=! 7,440645! /$! 1$;26/0/0$-! ,-8! 0--$@,/0$-! ,-8! 0-146,50-=! 1$-53;64!
1:$016! $4(! 0-! 50/3,/0$-5! ?:646! 0/! 05! -$/! 2$5507%6! /$! 0-/4$8316! 1$;26/0/0$-(! 7.!
0-/64@6-0-=! /$! ;0;01! /:6! $3/1$;65! $9! ,! 1$;26/0/0@6! ;,4F6/>'L! G:6! M556-/0,%!
N64@0165! O$;;0550$-! 0-! #01/$40,(! K35/4,%0,(! 05! /.201,%! $9! 531:! 46=3%,/$45>! H-!
0/5! $?-! ?$485(! 0/! 05! C#01/$40,D5! 0-8626-86-/! 61$-$;01! 46=3%,/$4! $9! 2465140768!
6556-/0,%! 3/0%0/.! 564@0165! 5322%068! 7.! /:6! 6%61/4010/.(! =,5(! 2$4/5(! ,-8! 4,0%! 9460=:/!
,22%.0-=! /$! /:6! 0-835/4.D! ,-8! C%$?! 0-1$;6! ,-8! @3%-64,7%6! 1$-53;645D>'R! G:05!

'&! S4$5564(!2&+'1+345$-",.'6#-+,7,)%+(!,7$@6!-!*&(!R>!N66!,%5$!S4$5564(!CT6=3%,/0$-!,-8!N$10,%!N$%08,40/.D(!
'L! K%%,-! K5:64(! CG:6! N1$26! ,-8! T,-=6! $9! M1$-$;01! T6=3%,/0$-I! S645261/0@6! 94$;! /:6! K35/4,%0,-!
'P! N66!M556-/0,%!N64@0165!O$;;0550$-(!>9"4-'-&+'6A8!"**!U,.!*XZ*+![://2I\\???>651>@01>=$@>,3\K7$3/]
T6=3%,/$4.! G4073-,%! $9! )6?! N$3/:! `,%65! [://2I\\???>02,4/>-5?>=$@>,3_! ,-8! /:6! M556-/0,%! N64@0165!
'R! 6%%+#-)$5'A+,*)=+%'8"@@)%%)"#'>=-!BCCD!"#01+!5!'K"Z+"8+(!"0+>
!"#$%&'()*+,+#-./"$+)*.)#0.1$"#"2+$.3%-(*)'+"#4.52&*+$)'+"#,.6"7.8"0%7#.3%-(*)'"7,.)#0. !!"

#$%& '(#()& *+,)-$.)$(/& 0$& 12-03& 4""56& 78+/)& 2-0$9023)/& 0$93:%);& 0$& #%%0(0+$& (+&
>)9+3+*09#33=& /:/(#0$#B3)& %),)3+2.)$(A;& #$%& >+99:2#(0+$#3& 8)#3(8& #$%& /#<)(=A6CD&
78:/;& ?803)& #& :(030(=& -)*:3#(+-A/& 2-0.#-=& 2:-2+/);& #$%& (8)& 2-0.#-=& ,#3:)/&
78)& 0.2+-(#$9)& +<& B#3#$90$*& )9+$+.09& #$%& /+90#3& ,#3:)/& B)9#.)& 93)#-& %:-0$*&
)#-3=& !EEE/6& 1.0%/(& (8)& 9-)#(0+$& +<& $)?& )3)9(-090(=& .#-F)(/& #(& (8)& ?8+3)/#3)&
:/)-/;& #$%& B)(?))$& %+.)/(09& #$%& 0$%:/(-0#3& :/)-/6& ':98& 9-+//& /:B/0%0)/& 9-)#()&
+<& )<!&90)$9=6CC& L)(& (8)0-& 9+.23)()& -).+,#3& ?+:3%& 8#,)& +<<)$%)%& /+90#3& ,#3:)/&
.#$=& 3+$*/(#$%0$*& B0330$*& #$%& %0/9+$$)9(0+$& 2-#9(09)/& Q& (8:/& 8+$+:-0$*& (8)&
:(030(=& 2-+,0%)-/& (+& 98#-*)& ,:3$)-#B3);& 3+?& 0$9+.)& #$%& -)*0+$#3& 9:/(+.)-/&
#$& >)<!&90)$(& .#-F)(& 2-09)AR& H(& 0/& /:B.0(()%& (8#(& 0(& ?+:3%& $+(& (#F)& 3+$*& B)<+-)&
9+..:$0(=;& .)%0#& #$%& 2+30(09#3& 2-)//:-)& B)9#.)& /:98& (8#(& (8)& <:$%#.)$(#3&

B Occupational Health and Safety Regulation Traditional

Social Regulation


CD& K+:$903&+<&1:/(-#30#$&U+,)-$.)$(/;&!"2&%'+'+"#.<7+#$+&*%,.:-7%%2%#';&44&12-03&4""5;&/&4N9O6
CC& V&V&S<!&9)-;&>W-0,#(0X#(0+$&+<&W:B309&1//)(/A&0$&K+..0(())&<+-&J9+$+.09&Y),)3+2.)$(&+<&1:/(-#30#&N)%O;&
C"& 7+$=&Z#-%&#$%&U-#).)&T+%*);&>J3)9(-090(=&W-0,#(0/#(0+$[&78)&I09(+-0#$&\+%)3A&0$&U-#).)&T+%*)&)(&
*&, !"#$%&'(#)*+,%)-.'/$0'1+*)+0!"#$%!&'(!)$!*+

89:! -267%./0$1(! 25$1$?05! B.%724! .%$12! <$! 1$/! <2!!12! /=2! 2L/21/! .1<! 1./7-2! $3!
M! $12! <240612<! /$! .5=02B2! 4$502/.%! $7/5$?24! <0332-21/! /$! .1<! >2//2-! /=.1! /=$42!
A-$<752<! >D! .1! 23!!5021/%D! $A2-./016! ?.-@2/F! 89:! -267%./0$1! 04! A-2B21/./0B2! 01!
1./7-2(! <240612<! /$! A-$/25/! E$-@2-4! T3-$?! .! B.-02/D! $3! .44.7%/4! /=./! %.0442UV3.0-2!
3$-524! E$7%<! $/=2-E042! A-$<752NFGW! R/! 04! >.42<! $1! B.%724! $3! 3.0-1244(! 2C70/D! .1<!
$3! 89:! -267%./$-4F! H=2! 23345$-)"#$6' 7+$6-&' $#8' 9$:+-.' ;3-' <==>' "#05+! ?.@24!

G,! S4! $12! .1$1D?$74! -232-22! 5$??21/2<(! ?.-@2/! 3.0%7-24! .-2! A-$>.>%D! /=2! ?.;$-! -2.4$1! 3$-! 89:!
GO! :22(! 26(! ^-$<75/0B0/D! Z$??0440$1(! ?$-)"#$6' @",A+,%B' C"D5+#%$-)"#' $#8' 23345$-)"#$6' 7+$6-&' $#8'
G*! b711016=.?(!.>$B2!1!_,(!*__F
G&! ^-$442-(!TP267%./0$1!.1<!:$50.%!:$%0<.-0/DN(!.>$B2!1!W&(!&_,K!b711016=.?(!.>$B2!1!_,(!*G[`\F!:22!.%4$!
GW! Y7//12-(!.>$B2!1!OO(!*'OF
G[! 92.%/=!.1<!:.32/D!Z$??0440$1(!;'9-,$-+F.':",'@",A56$3+'7+$6-&'$#8'9$:+-.')#'H,+$-'I,)-$)#'-"'<=<='
G\! 23345$-)"#$6'7+$6-&'$#8'9$:+-.';3-!<==>!"#05+!4!*"O+".+F
!"#$%&'()*+,+#-./"$+)*.)#0.1$"#"2+$.3%-(*)'+"#4.52&*+$)'+"#,.6"7.8"0%7#.3%-(*)'"7,.)#0. !"#

7C05'2* -2* 'D/+-6'0(G<H* B'* $3,'* (''2* )$'* 0'&C+3)-06* /'25C+CD* (B%2&* .0-D* 3*
)-* 31$%','* &-3+(* -.* FC()%1'9* (3.')6* 325* ('1C0%)6* N* 7C)* B%)$%2* 3* (6()'D* ($3/'5*

<=* <$$(&)'+"#)*.=%)*'>.)#0./)6%'9.:$'*?@@A*OP%1Q*((*R9*!J;*A''*3+(-*M-04A3.'*P%1)-0%39*="B.C"7D/)6%.
:&&*+%,. '>%. E)B. +#. 3%*)'+"#. '". 3%),"#)F*9. G7)$'+$)F*%4.:. H(+0%*+#%. 8)0%. I#0%7. /%$'+"#. J?. "6. '>%.
<$$(&)'+"#)*. =%)*'>. )#0. /)6%'9. :$'. ?@@A* O"J* S-,'D7'0* !JJ=Q* T$))/EUUBBB;B-04(3.';,%1;&-,;3CV;*
W$'* 'XC%,3+'2)* /0-,%(%-2(* %2* )$'* 2'B* $30D-2%('5* D-5'+* 30'* .-C25* %2* C"7D. =%)*'>. )#0. /)6%'9. :$'.
?@JJ. OY)$Q* /)* !* 5%,* #9* ((* "O!Q* 325* #H;*W$'* D-5'+*L1)* %(* 3,3%+37+'* -2* A3.'*M-04*LC()03+%38(* B'7(%)'*
<H* P30%37+6*5'(10%7'5*3(*:5'0'&C+3)%-289*:7'))'0*0'&C+3)%-289*:0'5*)3/'*0'5C1)%-289*:0'5C1%2&*)$'*0'&C+3)-06*
7C05'28* 325* :0')$%24%2&* 0'&C+3)%-289* )$'('* %2%)%3)%,'(* %2,-+,'* (C7F'1)%2&* 0'&C+3)-06* /0-/-(3+(* )-*
<<* W-* 7-00-B* )$'* +32&C3&'* -.* `3&230* b* c-.()'5)9* :W$'* AB%2&* -.* )$'* `'&C+3)-06* a'25C+CD* %2* bC0-/'E*
]0-D* a0'13C)%-2306* a0%21%/+'* )-* O`'&C+3)-06Q* >D/31)* L23+6(%(8* O!JJRQ* !HO"Q* L"(7#)*. "6. 3+,D. )#0.
I#$%7')+#'9* !"=;* A''* 3+(-* KC22%2&$3D* 325* K037-(469* 37-,'* 2* "=9* H9* B$-* 0'.'0* )-* )$'* %2"*C'21'* -.*
2'-\+%7'03+* )$%24%2&* 3)* )$'* /-+%)%13+* +','+G* )-* '1-2-D%1* 03)%-23+%(D* 5-D%23)%2&* (-1%3+* /-+%16* 5'73)'(G*
#JJ* L*1-+-C0.C+*'I3D/+'*%(*)$'*'+'1)%-2*D32%.'()-*B0%))'2*.-0*d'0'D6*Y+304(-29*$-()*-.*)$'*fg*)'+',%(%-2*
#J#* Y-C21%+*-.*LC()03+%32*K-,'02D'2)(8*^'')%2&9*!"22(#+N(O*O!j*^301$*!JJHQ*T$))/EUUBBB;1-3&;&-,;3CU
*&* !"#$%&'(#)*+,%)-.'/$0'1+*)+0!"#$%!&'(!)$!*+


C Australian Competition and Consumer Commission A

More Economically and Socially Centred Regulator

03.4B! 0129/1924! 56! ;2$>35313-D! ,-13K/$7;4131384! /$-.9/1! ,-.! 24D9%,13-D! ,//400!
1$! 7$-$;$%6! 04283/40(EF&! ,-.! ,! C4%%K3-:$274.! ,-.! /$-!!.4-1! A.47,-.K03.4B! 56!
;2$14/13$-! 1>30! 7,-.,14! 7,6! C,22,-1! ?@@@! 3-14284-13$-! -$1! $-%6! 1$! /$224/1!
,-.! 0$/3,%%6! /4-124.! 24D9%,1$2! 3-! 1>,1! 3-! 7,-6! 3-01,-/40! 31! 30! 24Q9324.! 1$! 1,J4!
S43-D! ,! 7$24! 4/$-$73/,%%6! ,-.! 0$/3,%%6! /4-124.! 24D9%,1$2! 30! ,2D9,5%6! 7$24!
/$7;%4H! 1>,-! 543-D! ,! /%4,2! 4/$-$73/! 24D9%,1$2! 095P4/1! 1$! 09;;$213-D! "$2!
04/$-.,26+! 0$/3,%! 8,%940(! $2! ,! /%4,2! 0$/3,%! 24D9%,1$2! 095P4/1! 1$! 04/$-.,26! "$2!
09;;$213-D+! 4/$-$73/! 8,%940<! =>4! /$7;%4H316! ,-.! .3:!!/9%16! $:! 1>30! 1,0J! 30!
31! /$-12$%%4.! ,5$91! 1C$K1>32.0! $:! 1>4! D2$/426! 04D74-1<! G-313,%%6! 1>4! ?@@@! C,0!

EF*! T$2! 4H,7;%4(! 3-! 1>4! U1,1474-1! $:! VH;4/1,13$-0! ;2$83.4.! 1$! 1>4! ?@@@! 56! 1>4! @$77$-C4,%1>!
,-.! @$-09742! @$773003$-(! *F! T4529,26! *FFX! Z>11;[\\,2/>384<124,0926<D$8<,9\.$/974-10\E*'X\O]T\
EF&! =>404!,24!1>4!$5P4/13840!$:!;10!G#!,-.!`GS!,-.!;10!GGG?!,-.!`G@!240;4/1384%6!$:!1>4!2"34+-)-)"#'$#5'
EFL! =>30! 30! 1>4! $5P4/1384! $:! ;10! G#S(! `G(! `G??! ,-.! U/>! *! $:! 1>4! 2"34+-)-)"#' $#5' 2"#%63+,' 78-! 9:;:!
EFN! 2"34+-)-)"#'$#5'2"#%63+,'78-!9:;:!"@1>+!0!*<
EFR! O2$.9/138316!@$773003$-(!1+*)+0'">'76%-,$<)$?%'2"#%63+,'@"<)8.'A,$3+0",B(!G-Q9326!c4;$21!)$!LN!
!"#$%&'()*+,+#-./"$+)*.)#0.1$"#"2+$.3%-(*)'+"#4.52&*+$)'+"#,.6"7.8"0%7#.3%-(*)'"7,.)#0. !""

#$%&'() $') (*&) +,-&) .+) (*&#&) /&0&%.12&'(#3) 14&#52,6%7) (,8$'9) (*&) 0$&:) (*,() (*$#)
:,#) 14&/.2$','(%7) ,') &-.'.2$-) $##5&) ;) *&,%(*7) -.21&($($.') ,() $(#) 6&#() ;) ,)
,'/) #2,%%) 65#$'&##) 4&(,$%&4#3) ,'/) .') (*&) #.-$,%) +,64$-) .+) (*&) %.-,%) ,'/) 4&9$.',%)
-.'-&4') ,6.5() (*.#&) $21,-(#B3) $21,-(&/) '&9,($0&%7) .') (*&) >???@#) -4&/$6$%$(7)


,%#.) .1&4,(&#) :$(*$') ,) #7#(&2) -.'#(4,$'&/) 67) (*&) .(*&4) 0,%5&#<) D() ,%#.) 4&0&,%&/)
;) :$(*) /$++&4&'() 6,%,'-&#) 6&$'9) #(45-8) :$(*) 4&#1&-() (.) /$++&4&'() 1.%$-7) ,4&,#3)
6.(*) (.) &'*,'-&) &+!)-$&'-7) ,'/) -.'#52&4) -*.$-&) ,'/) ,((,$') -&4(,$') #.-$,%%7)
/&#$4,6%&) .5(-.2&#) A&9) &##&'($,%) #&40$-&#) ,'/) 5($%$($&#) 4&95%,($.') /&#$9'&/) (.)
.5(-.2&#3) ,'/) 14.(&-() (*&) 05%'&4,6%&) ,'/) %&##) 1.:&4+5%B<) D'/&&/3) $') 2./&4')

FGH) I&&3)&93)J%$T,6&(*)U'$9*(3)EI$%&'()P,(-*/.9)V'%$8&%7)(.)W,48):*$%&)?.'#52&4#),4&)?,(#)(*,()9&()(*&)
_$6&4,%) I&',(.4) W$%%) =&++&4','3) d,($.',%#) I&',(.4) ^.*') P$%%$,2#) ,'/) >5#(4,%$,') f4&&'#) I&',(.4)
FGN) J0$/&'-&)(.)I&',(&)J-.'.2$-#)K&+&4&'-&)?.22$((&&3)`,4%$,2&'().+)>5#(4,%$,3)g)X,4-*)!GFF3)FgbcH<)
*&, !"#$%&'(#)*+,%)-.'/$0'1+*)+0!"#$%!&'(!)$!*+

-./0.1! 21! 34$1$5.4! 4$1/36/7! 89! :;<1! $=93>?39@! A42B./2%.95! 21C! C35$4>24D! 777!
2%%! <1C3>%D.1M! 299<5B/.$19! 21C! -$>;! $1%D! -./0! ?2>.2=%39! 9<40! 29! F<21/./D! 21C!
244$>C.1M! /$! /0.9! B3>9B34/.?3(! .9! 1$/! /03! 2=93143! $G! $13! $>! 5$>3! ?2%<39(! $>! /03!
?.4/$>D! $G! $13! 93/! $G! ?2%<39! $?3>! 21$/03>! "29! .5B%.3C! =D! 5<40! $G! /03! %./3>2/<>3!
/02/! B$9./9! /0393! ?2%<39! .1! 4$5B3/./.$1! -./0! $13! 21$/03>+(! =</! /03! B>.524D! 21C!
9<BB$>/.1M! >$%39! B%2D3C! =D! 9$53/.539! 4$1/>2C.4/$>D! ?2%<397! L03! .5B%.42/.$1!
81C! -0.%9/! /03! 521C2/3! $G! 2! >3M<%2/$>D! =$CD! 52D! -3%%! 02?3! =331! 93/! /0>$<M0!
?2%<39! 9<BB$>/! 21C! 902B3! /03! /3>>2.1! 4$?3>3C! =D! /03! >3M<%2/$>7! L0<9(! /03! .99<3!
36B%.4./! 21C! 4%32>7! P1! 5$9/! .19/21439! /03! B>3G3>3143! 90$<%C! =3! G$>! />219B2>314D!
9$4.2%! ?2%<39! 21C! />2C3Q$GG9! =3/-331! /035! ?.9.=%37! L$! .M1$>3! $>! B>3/31C! /0393!
B>.$>./D! $>! 0.3>2>40D! 52D! 1$/! =3! 9$! $=?.$<9! "-0.40! 2>M<2=%D! .9! /03! 4293! -./0!
52D! 1$/! !!/! -3%%! -./0.1! /03! <9<2%%D! G2>! 9.5B%3>! 3?3>DC2D! B<=%.4! C.94$<>93! 21C!
B$%.4D! C3=2/39(! 29! -3%%! 29! /03! 242C35.4! 5$C3%9! 2C$B/3C! G$>! >3M<%2/.$17! P/! 2%9$!

IJK! :;<1(!2=$?3!1!UV(!I*J7
IIJ! W3?.QX2<>(!AY3M<%2/.$1!21C!Y3M<%2/$>D!Z$?3>12143H(!2=$?3!1!**(!I,[IU7!O33!2%9$!Z>2353!8!\$CM3(!
Z>3?3! 21C! 81/0$1D! S! _$2>C521! "3C9+(! 2#-+,#$-)"#$3' 4$#56""7' "#' 8963):;8,)*$-+' 8$,-#+,%&)<%!
!"#$%&'()*+,+#-./"$+)*.)#0.1$"#"2+$.3%-(*)'+"#4.52&*+$)'+"#,.6"7.8"0%7#.3%-(*)'"7,.)#0. !"#

$%&'(%)* +* &%+,-&.* /%01%%* '2* 3'44567-.* )3%8-737)4* '2* 1%05,+-'1)* -&+-* 81%)%6-*
:'51-&;* <&7,%* 7-* 4+.* )%%4* )'4%<&+-* '$(7'5)* -'* '$)%1(%* -&+-* +,,* 1%05,+-7'6*
76(',(%)* +* $+,+63760* '2* )'4%-74%)* 3'6-1+/73-'1.* (+,5%);* <&+-* -&7)* 4'/%,*
%48&+)7)%)* 7)* -&%* 748'1-+63%* '2* 1%05,+-'1)* +3=6'<,%/0760* +6/* )5$)-+6-7(%,.*
$%* ,%07),+-7(%,.* $%)-'<%/9CCC* D%05,+-'1)* +74760* -'* 4+76-+76* +6/* %6&+63%* -&%71*
:72-&;* 76* 4+=760* -&%)%* '$)%1(+-7'6)* <%* <'5,/* 6'-* <+6-* -'* $%* +335)%/* '2*
'(%1)748,7!*3+-7'69* E&%* 81'3%))* '2* $1760760* %3'6'473* +6/* )'37+,* (+,5%)* 76-'*
(+,5%)H;* 4%%-760* -&7)* /72!*35,-* 3&+,,%60%* +6/* $1760760* -&%* /722%1%6-* /74%6)7'6)*
1%05,+-'1)* -'* +3&7%(%* -&7)* $+,+63%;* )+(%* -'* 6'-%* -&+-* 0''/* 1%05,+-'1.* 81'3%))%)*
+1%* 0%6%1+,,.* 4'1%* 3'6/537(%* -'* 81'/53760* >)4+1-?* 1%05,+-7'6* +6/* 0''/*
+6/* 8+1-7378+-'1.* 4'/%,)* )53&* +)* P,+3=?)* >-&73=* 81'3%/51+,7A+-7'6?* $+)%/* '6*

CCC* Q+1=* @* B53&4+6;* >Q+6+0760* R%07-74+3.S* B-1+-%073* +6/* J6)-7-5-7'6+,* O881'+3&%)?* GCTT#H* !UG"H*
CC!* B%%;*%0;*E*E.,%1;*?@9.A%"&*%.BC%9.'@%.D)<4.A7"$%0(7)*.E(,'+$%F.D%-+'+2)$9.)#0.!"2&*+)#$%*GL1763%-'6*
Q'-7(+-7'6+,* L')-51%);* +6/* D%)8'6)7(%* D%05,+-7'6?* G!UUVH* !T* D)<. )#0. A"*+$9* C"VZ* W5,7+* P,+3=;*
>@'6)-153-760* +6/* @'6-%)-760* R%07-74+3.* +6/* O33'56-+$7,7-.* 76* L',.3%6-173* D%05,+-'1.* D%074%)?;*
CC"* K+76%);*B5--'6*+6/*L,+-+67+ML&560;*+$'(%*6*C#;*N#C9*B%%*+,)'*K+76%);*+$'(%*6*C#;*NX]#C9
CCN* ^566760&+4*+6/*^1+$')=.;*+$'(%*6*"V9*
CC#* P,+3=;*>L1'3%/51+,7A760*D%05,+-7'6S*L+1-*J?;*+$'(%*6*TT9
*&, !"#$%&'(#)*+,%)-.'/$0'1+*)+0!"#$%!&'(!)$!*+

.36! 9G<$0-34-! 946-9-2-9$43%! 9G<%9=3-9$46H! K./! -0359-9$43%! 59=.$-$G;! :/-8//4!
345! $-./0! 6$=93%%;! 5/6903:%/! $2-=$G/6! :/941! =$452=-/5! -.0$21.! G3N$09-30934!


0/12%3-9$4! 345! /=$4$G9=! 0/12%3-9$4H! K.96! 59>!!=2%-;! .36! ://4! =$G<$245/5! :;!
$>! -.96! =.341/! >$0! -0359-9$43%! 59=.$-$G9/6! 345! 343%;-9=3%! >03G/8$076H! R4! -.96!
96! 9G<$0-34-! -.3-! 0/12%3-$0;! =$4=/<-6! 30/! =$4-9423%%;! 0/@969-/5! 345! =%309!!/5! -$!
0/@9/8! 95/4-9!!/5! 3! :0$35%;! =$4696-/4-! =$4=/<-! $>! /=$4$G9=! 0/12%3-9$4! =/4-0/5!
$4! 9G<0$@941! /=$4$G9=! />!!=9/4=;! 345! =$G</-9-9$4H! B$8/@/0(! 89-.! 0/6</=-! -$!
6$=93%! 0/12%3-9$4! -8$! :0$35! 345! =$4"!9=-941! =$4=/<-23%963-9$46! 8/0/! 95/4-9!!/5H!
K./! !!06-! @9/86! 6$=93%! 0/12%3-9$4! 36! 3! G/=.3496G! -$! =$00/=-! >$0! -./! 53G31941!
/>>/=-6! $>! /=$4$G9=! 3=-9@9-;S! -./! 6/=$45! 36! 3! G/=.3496G! -$! 3--394! 6$=93%%;!
3==203-/%;! 0/"!/=-6! 8.3-! .3<</46! 94! <03=-9=/H! Q$@/04G/4-6! 30/! 69G2%-34/$26%;!
345! -$! <0$52=/! 6$=93%%;! 5/6903:%/! $2-=$G/6! 59>>/0/4-! -$(! 345! :/--/0! -.34(! -.$6/!
-.3-! 35@34=/6! :0$35/0! 6$=9/-3%! @3%2/6! 62=.! 36! >3904/66(! N26-9=/! 345! /P29-;(! 345!
-0359-9$43%%;! =$4=/9@/5! $>! 6$=93%! 0/12%3-$06! 30/! 94=0/36941%;! :/941! 0/P290/5! -$!

AA,! U0$66/0(!2&+'1+345$-",.'6#-+,7,)%+(!3:$@/!4!*&H!
AAI! B$$5(!C$-.6-/94!345!D3%5894(!3:$@/!4!A*(!A'VH
AA'! L//!596=2669$4!3-!3:$@/!4!*&H!
!"#$%&'()*+,+#-./"$+)*.)#0.1$"#"2+$.3%-(*)'+"#4.52&*+$)'+"#,.6"7.8"0%7#.3%-(*)'"7,.)#0. !"#

,+(6%(-%(&) 7.(0) *+1%&-1%*) -0%+/+3-,.//8) 02-6%(9) 4*%) +:) &5%*%) &2.0-&-+(./) /.$%/*)
>5-*) .2&-,/%) 5.*) *433%*&%0) &5.&) .) $%&&%2) '.8) +:) ,+(,%<&4./-*-(3) *+,-./) .(0)
0-*&-(,&-+() $%&'%%() *+,-./) .(0) %,+(+1-,) 2%34/.&-+() /-%*) -() &5%) <2-1.,8) +:) &5%)
4*) &5.&) <42%) %,+(+1-,) 2%34/.&-+() +2) <42%) *+,-./) 2%34/.&-+() +(/8) %?-*&*) -() &5%)
European Financial Management, Vol. 11, No. 4, 2005, 439451

Understanding Regulation
Andrei Shleifer
Whipple V. N. Jones Professor of Economics, Harvard University

1. The Puzzle of Regulation

The American and European societies are much richer today than they were 100 years
ago, yet they are also vastly more regulated. Today, we live in houses and apartment
buildings whose construction from zoning, to use of materials, to fire codes is
heavily regulated. We eat food grown with heavily regulated fertilizers and hormones,
processed in heavily regulated factories with publicly monitored technologies, and
sold in heavily regulated outlets with elaborate labels and warnings. Our means of
transport, including cars, buses, and airplanes, are made, sold, driven, and maintained
under heavy government regulation. Our children attend schools that teach heavily
regulated curriculae, visit doctors following heavily regulated procedures and paid
government-controlled prices, and play on play-grounds using government-mandated
safety standards.
The extraordinary pervasiveness of government regulation in our lives raises a
number of questions. Is regulation generally a good idea, as the positive correlation
between its growth and the growth of income seems to indicate, or has it been an
obstacle to economic and social progress? Have the USA and Western Europe grown
in spite of it? How much regulation of a particular activity is appropriate? Does the
nature of the activity being regulated, or the characteristics of a country, influence the
optimal choice? Is the level of regulation we observe in fact an outcome of efficient
social choice, or are other factors as or more important?
Over the twentieth century, economists have come up with a number of ways of
thinking about government regulation. In this paper, I review some of the key theories
of economic regulation, and assess their relevance, paying particular attention to the
regulation of securities markets. The three theories I focus on are the welfare-theoretic
or public interest theory of regulation associated with Pigou (1938), the contracting
theory associated with Coase (1960), and the capture theory of Stigler (1971). I then
describe an alternative way of thinking about regulation and social control of business
more generally, developed in a series of papers with Simeon Djankov, Edward
Glaeser, Rafael La Porta, and Florencio Lopez-de-Silanes. Finally, I use this theory
to shed light on some differences in regulatory patterns around the world.

Keynote Address at the 2004 European Financial Management Association (EFMA) Meetings,
Basel, Switzerland, 2 July 2004.

# Blackwell Publishing Ltd. 2005, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
440 Andrei Shleifer

2. Theories of Regulation

The standard public interest or helping hand theory of regulation is based on two
assumptions. First, unhindered markets often fail because of the problems of mono-
poly or externalities. Second, governments are benign and capable of correcting these
market failures through regulation. This theory of regulation has been used both as a
prescription of what governments should do, and as a description of what they
actually do, at least in democratic countries. According to this theory, governments
control prices so that natural monopolies do not overcharge, impose safety standards
to prevent accidents such as fires or mass poisonings, regulate jobs to counter the
employers monopsony power over the employee, regulate security issuances so
investors are not cheated, and so on. The public interest theory of regulation has
become the cornerstone of modern public economics, as well as the bible of socialist
and other left-leaning politicians. It has been used to justify much of the growth of
public ownership and regulation over the twentieth century (Allais, 1947; Meade,
1948; Lewis, 1949).
Public interest theory of regulation has been subjected to a number of criticisms,
associated mostly with the Chicago School of Law and Economics. These criticisms
proceed in three intellectual steps. First, markets and private orderings can take care
of most market failures without any government intervention at all, let alone regula-
tion. Second, in the few cases where markets might not work perfectly, private
litigation can address whatever conflicts market participants might have. And third,
even if markets and courts cannot solve all problems perfectly, government regulators
are incompetent, corrupt, and captured, so regulation would make things even worse.
Consider these three lines of argument in order.
The first line of attack criticises the public interest theory for exaggerating the
extent of market failure, and for failing to recognise the ability of competition and
private orderings to address many of the alleged problems. Competition for labour,
the argument goes, itself assures that employers provide safety and good working
conditions for employees. If an employer failed to do so, his competitors would offer
the more efficient packages, and thereby attract better workers at lower wages.
Likewise, private markets assure the efficient safety levels in a variety of products
and services, such as trains, houses or cars. Sellers who fail to deliver such levels of
safety lose market share to competitors who run safer trains, build safer houses, or
produce safer cars. The competition criticism also maintains that what looks like a
monopoly to would-be regulators is subject to potential entry and competition.
Moreover, cartels typically break up after a short time because their participants
cheat to make windfall profits.
Even when competitive forces are not strong enough, private orderings work to
address potential market failures. Neighbours resolve disputes among themselves,
without any government intervention, because they need to get along with each
other over long stretches of time (Ellickson, 1991). Industries form associations that
guarantee quality, and penalise cheaters among themselves to assure that, in the long
run, customers continue their patronage (Greif, 1989; Bernstein, 1992). Families,
cities, and ethnic groups establish reputations in the marketplace, and thereby control
any possible misconduct by their members.
The thrust of these arguments is that the domain of market failure or socially
harmful conduct that is not automatically controlled by impersonal forces of compe-
tition is extremely limited, and therefore so is the scope for any desirable intervention

# Blackwell Publishing Ltd, 2005

Understanding Regulation 441

by the state. But this, of course, is only the first step in a much broader assault on
The second step, originating in the work of Coase (1960), maintains further that, in
the few cases where competition and private orderings do not successfully address
market failures, impartial courts can do so by enforcing contracts and common law
rules for torts. Employers can offer workers employment contracts that specify what
happens in the event of an accident, security issuers can voluntarily disclose informa-
tion to potential investors and guarantee its accuracy, and so on. As long as courts
enforce these contracts, equilibrium outcomes are efficient. Indeed, even when there
are no contracts, efficient adjudication by courts restores efficiency through appro-
priate tort rules. When courts award damages to harmed plaintiffs correctly, potential
tort-feasers face exactly the right incentives to take the efficient level of precaution
(Posner, 1972). With well functioning courts enforcing property rights and contracts,
the scope for desirable regulation even by a helping hand government is minimal.
Coases logic has proved extremely powerful, both as a technical critique of regula-
tion and as a libertarian manifesto. Following Coase, the Chicago school has gone
much further. The third, and crucial step in its critique of regulation is to question the
assumptions of a benevolent and competent government. This is the essence of
Stiglers capture theory (Stigler, 1971; Posner, 1974). As forcefully summarised by
Peltzman (1989), this theory consists of two basic propositions. First, the political
process of regulation is typically captured by the industry. Regulation not only fails to
counter monopoly pricing, but is to the contrary used to sustain it through state
intervention. Second, even in the cases where, under the influence of organised
consumer groups, regulators try to promote social welfare, they are incompetent
and rarely succeed. Thus the scope for government regulation is minimal at best,
and such intervention is futile and dangerous even in the rare cases where there is
The Chicago critique of public interest regulatory theory is one of the finest
moments of twentieth century economics. The pioneers of this critique not only
provided new theories for thinking about the role of government, but also delivered
predictions which in many cases have been supported by the evidence particularly
the evidence of pervasive regulatory failure. Yet the Chicago critique cannot be the
final answer, for two crucial reasons.
At the theoretical level, the Chicago Schools confidence in private orderings and in
courts is excessive. Private orderings indeed work extremely well in some situations,
but they also degenerate into the anarchy of private enforcement, where the strong
and not the just win the day. Moreover, Coase and his followers have given far too
much leeway to courts, relying on them as unbiased, informed, and incorruptible
promoters of social welfare. Much evidence, however, shows that courts around the
world are more often than not highly inefficient, politically motivated, slow, and even
corrupt (Johnson et al., 2002; Djankov et al., 2003a). The lopsided belief in the
benevolence of courts and the malevolence of regulators has neither a conceptual
foundation, nor a solid grounding in reality. After all, both judges and regulators are
government agents, subject to political pressures, incentives, and constraints.
At the empirical level, the Chicago tradition has failed to come to grips with the
basic facts described in the first paragraph of this paper, namely that today we live in
a much richer, more benign, but also more regulated society, and that as consumers
we are generally happy with most of the regulations that protect us. We are happier
knowing that trains and airplanes are safe than savouring the thought of a fortune
# Blackwell Publishing Ltd, 2005
442 Andrei Shleifer

which our loved ones would collect in a trial should we die in a fiery crash. In
securities markets, investors prefer a level-playing regulated field to the prospect of
loss recovery through litigation. Indeed, as I discuss below, there is strong evidence
that regulation is beneficial for the development of financial markets and to public
participation in them (Glaeser et al., 2001; La Porta et al., 2004). A more nuanced
theory, which incorporates the powerful Chicago critiques of the public interest
approach to government, but also recognises the benefits of public involvement in
at least some activities, is clearly needed to keep the logic and the facts together.

3. The Enforcement Theory of Regulation

Suppose that the society wishes to control business to pursue some socially desirable
end: marginal cost pricing, safe food and water, or safety precautions by firms. As
Djankov et al. (2003b) argue, there are four distinct strategies of such control, invol-
ving ever growing powers of the state vis-a-vis private individuals: market discipline,
private litigation, public enforcement through regulation, and state ownership. These
four strategies for social control of business are not mutually exclusive: competition
and regulation often operate in the same market, as do private litigation and public
regulation. In addition, there are common intermediate strategies of social control of
business, such as private litigation to enforce public rules, which lies between pure
regulation and pure litigation. Nonetheless, these four categories provide a useful
analytical classification, which also has the advantage of following closely the his-
torical discussions of the proper role of government.
To illustrate these categories, take the example of social control of securities issues.
Promoters of such issues, be they entrepreneurs or underwriters, have a strong
incentive to cheat investors by selling them worthless or overvalued securities, taking
the money, and running off. When this so-called promoters problem is severe, people
stop buying new issues, with the result that financial markets fail to grow or even
Suppose that the society has an interest in having broad and liquid securities
markets and, to this end, deems it desirable that firms issuing equity disclose accurate
information about their circumstances. The society has four choices. First, it can rely
on the reputational incentives of the issuers themselves, or of their underwriters, to
disclose the truth about the securities this is the market discipline solution. Second,
the society can rely on private suits by buyers of securities who feel that they had been
cheated, under the general doctrines of contract or tort. In this scenario, security
issuance would be treated as any other instance of a contract or a tort. The question
for the court is whether the issuer or the underwriter provided inaccurate information
or, alternatively, failed to provide information that a plausible standard of care would
require. Third, the society can create a regulatory agency which mandates what
should be disclosed by security issuers, inspects these disclosures, and penalises issuers
and underwriters who fail to confirm to the regulations.1 Finally the government can
nationalise all security issuance, so its own agents make representations and sell
stocks. These are the four basic strategies of enforcement of good conduct.

The regulatory agency can also establish the rules for security issuance, but leave the enforce-
ment of these rules to private litigation by the wronged investors.

# Blackwell Publishing Ltd, 2005

Understanding Regulation 443

These four strategies are ranked by the growing degree of public control over
economic activity. With competition and private orderings, there is basically no public
involvement at all. With courts, there is a role for impartial judges enforcing rules of
good behaviour. These rules do not need to come from legislation, but may instead
derive from custom or from judge-made common law and precedents. Even so, there
is a public agent, the judge, who has at least some decision-making authority. With
regulators, control by the state increases sharply. The state now writes the rules,
inspects the product before it is sold, and possibly penalises sellers for delivering a
bad product. Both the scope of government activity, and its centralisation, are greatly
increased relative to private litigation. Finally, with state ownership, the government
takes complete control over an activity.
The basic premise of the enforcement theory of regulation is that all of these
strategies for social control of business are imperfect, and that optimal institutional
design involves a choice among these imperfect alternatives. The enforcement theory
specifically recognises a basic trade-off between two social costs of each institution:
disorder and dictatorship. Disorder refers to the ability of private agents to harm
others to steal, overcharge, injure, cheat, impose external costs, etc. Dictatorship
refers to the ability of the government and its officials to impose such costs on private
agents. As we move from private orderings to private litigation to regulation to public
ownership, the powers of the government rise, and those of private agents fall. The
social losses from disorder decline as those from dictatorship increase. This tradeoff,
which Djankov et al. (2003b) call the Institutional Possibility Frontier, is shown in
Figure 1.
Consider the four strategies of social control of business from the perspective of this
trade-off. The principal strength of market discipline as the method of enforcing good
conduct is that it is free of public enforcers. There is no possibility of politicisation of

Social losses due to

private expropriation

Private orderings

Independent judges

Total loss Institutional possibility frontier (IPF)


Regulatory state

Social losses due to state

expropriation (Dictatorship)

Fig. 1. Institutional possibilities.

Source: Djankov et al., 2003b.

# Blackwell Publishing Ltd, 2005

444 Andrei Shleifer

rules of conduct, of corruption, of costly and delayed enforcement of rules, of random

or compromised choice of one competitor over another. But market discipline may be
very weak at controlling disorder. Market participants can use their economic,
political, or social resources to damage their rivals using methods ranging from
predation to monopoly pricing to social exclusion to outright theft or violence. One
mans peaceful private orderings become another mans death in the hands of the
mafia. When market discipline can successfully control disorder and avoid Hobbesian
anarchy, it is the best approach because it has the lowest social costs of dictatorship.
Any case for public intervention relies crucially on the presumptive failure of market
discipline to control disorder.
In many instances, this case for the effectiveness of market discipline is powerful.
Consider the regulation of entry: the restriction on entry by new entrepreneurs
through licensing and permits. Since entering firms are small, and since any failure
to deliver quality products would be almost immediately recognised and penalised by
customers, it is not clear why the quality of entrepreneurs or of their firms should be
regulated at the entry level. To the extent that market discipline can control disorder,
regulation or even courts, are unnecessary.
But market discipline may not be sufficient. Employers may under-invest in safety
and then blame accidents on an injured workers own carelessness. In security issu-
ance, a fraudulent scheme can separate investors from their money very quickly, and
undermine confidence in markets. The frequent Ponzi schemes in emerging markets,
in which hundreds of thousands of investors lose their lifetime savings, are colourful
evidence that market discipline does not eliminate fraud. In such instances, to control
disorder, societies may efficiently accept a higher level of government intervention.
The traditional libertarian response to such market failure is to move one notch
toward more dictatorship and less disorder by relying on the enforcement of good
conduct through private litigation. Injured employees can sue their employers for
damages. Investors can sue issuers and underwriters for damages when they believe
that representations about the companys prospects were false or incomplete. Ideally,
a judge would recognise quickly whether investors have been misled, and award
damages to compensate them for their losses.
Private litigation has many advantages. At least in principle, such litigation is of no
special interest to the government, and hence disputes can be resolved apolitically,
with no favours to influential parties. Judges may also acquire experience and expert-
ise in contract enforcement (as well as in handling tort cases), and hence address
problems efficiently and expeditiously. This, indeed, is what Coase (1960) and later
Posner (1995) had in mind in making the case for courts.
The reality of litigation is not, unfortunately, so perfect, and the trade-off between
dictatorship and disorder is helpful for thinking about courts as well. To begin, the
same forces that undermine the effectiveness of private orderings influence courts as
well. As a result, courts are often subverted and the strong not the just win the case
(Galanter, 1974). Some of the mechanisms of influencing courts are entirely legal.
Hiring superior legal talent or using legal delay tactics are among them. Individual
investors in Ponzi schemes or worthless security issues may stand little chance in court
against wealthy and well-represented promoters.
Still other mechanisms, such as political influence on judges, are perhaps less
appropriate, but still common, especially in countries where judges are not politically
insulated (Ramseyer and Rasmusen, 1997). Politicians can then influence judges to
help themselves and their financial backers, as recent evidence from Italy illustrates
# Blackwell Publishing Ltd, 2005
Understanding Regulation 445

only too clearly. In still other countries, judges are bribed with cash, benefits, or
promises of promotion, as well as threatened if they do not rule for the strong (Dal Bo
et al., 2003). Because the rich and the politically connected have more resources to
influence the path of justice, private litigation cannot be always counted on as an
effective mechanism of enforcing socially desirable conduct.
A common mechanism for protecting courts from influence is to formalise legal
procedures through codes, so as to minimise judicial discretion and the potential for
subversion. Most countries, especially those in the civil law tradition, have heavily
formalised their legal procedures to assure accuracy, and to prevent the subversion of
justice. But such formalism is associated with serious delays, as well as unpredictable
outcomes (Djankov et al., 2003a). The Coasian ideal of cheap and efficient justice
through private litigation is a far cry from reality.
A related mechanism for controlling the subversion of judges is to make them
employees of the state, whose career concerns protect them from succumbing to
outside influence. Truly independent judges are more vulnerable to private subversion
than the state employed ones. But as judges become more dependent on the state, the
risk of politicisation of their decisions rises.
As with market discipline, the enforcement theory points to the circumstances
where private litigation is likely to be relatively effective. It is likely to work better
where judges are better insulated from political pressure, which is probably the case in
the more advanced economies. It is also likely to be more effective in the cases where
the problem of inequality of weapons between the litigants is smaller. For example,
in relatively advanced economies, tenant landlord disputes or employment contract
disputes may well be most efficiently resolved in specialised courts. Yet in countries
and in the types of conflicts where judges are vulnerable to subversion, and the
inequality of weapons is considerable pure private litigation is unlikely to be the
efficient method of enforcing socially desirable conduct. Securities markets are one
example illustrating this point: it is simply not plausible that defrauded investors can
prevail in court against the richer, better connected, and better represented promoters
and underwriters. Mechanisms for social control of business that are more effective at
controlling disorder may be needed even if they are more vulnerable to dictatorship.
This brings us to the third strategy of enforcing rules, government regulation.
Before turning to full-fledged public enforcement, note an extremely important inter-
mediate strategy, namely private litigation using public rules. A government can
create a set of rules governing private conduct and then leave the enforcement of
these rules to private parties. The reason for doing so is that the enforcement of
specific statutes through litigation might be considerably cheaper than that of broad
contractual principles. It may be efficient, for example, for the government to specify
the appropriate safety standards but to leave their enforcement to workers through
private litigation. In securities markets, the government can mandate specific disclo-
sures by an issuer, but then let dissatisfied investors sue. It may be cheaper for
investors to establish in a trial that the company has failed to reveal specific informa-
tion whose disclosure was mandated by law, than to prove issuer negligence in the
absence of a statute.
Private enforcement of public statutes solves a number of problems with pure
litigation. First, as the examples above suggest, the burdens on the courts and the
litigants of establishing liability fall considerably when the statutes describe precisely
what facts are needed to do so. Second, subversion of judges becomes more difficult
and expensive when they lose discretion. It may be relatively easy to convince a
# Blackwell Publishing Ltd, 2005
446 Andrei Shleifer

judge by persuasion or bribery that a security issuer who concealed information

from investors is not liable when there are no specific rules as to what needs to be
disclosed. It is much harder for the issuer to convince the same judge when the law
states specifically what must be disclosed. Perhaps for these reasons, private enforce-
ment of public rules is a highly efficient strategy of enforcing good conduct in many
situations (Hay and Shleifer, 1998; Hay et al., 1996). La Porta et al. (2004) show
empirically that this is a crucially important strategy for enforcing good conduct in
security issuance. Barth et al. (2003) similarly point to the importance of private
enforcement of public disclosure rules in bank regulation.
At the same time, the creation of public rules even those enforced privately
raises the scope for public abuse. Such rules can be used to expropriate the politically
weak and to favour the politically strong. Mandatory safety precautions in factories,
mines, and meat-packing plants during the progressive era in the USA at the begin-
ning of the twentieth century, for example, are sometimes interpreted as an attempt by
large established firms to prevent entry by smaller rivals by raising their costs from
regulatory compliance (Libecap, 1992; Coppin and High, 1999).
Compared to the enforcement strategies described above, public regulation has a
number of advantages in controlling disorder. First, unlike judges, public regulators
can be expert and motivated to pursue social objectives in specific areas. This, indeed,
has been the principal argument for public regulation of securities markets (Landis,
1938; Glaeser et al., 2001; Pistor and Xu, 2002). A regulator can establish some
expertise, for example, as to what constitutes a material omission from a prospectus,
present market participants with specific rules, and then use its resources to make sure
that these rules are followed by imposing its own sanctions or by convincing courts to
rely on its rules. Second, because regulators can be provided with incentives to enforce
social policy, they can in principle be more difficult to subvert than the disinterested
judges. This combination of expertise and incentives is presumably what makes public
enforcement in some circumstances more efficacious than private enforcement.2
Alas, public regulation is not without its problems, and the key problem is the risk
of public abuse of market participants by an official who is either pursuing his own
political interests or is captured by a particular group, including the regulated indus-
try itself. Politicisation and over-enforcement are a particular problem in societies
with few checks and balances: the executive can selectively turn its regulators against
its enemies rather than violators of rules. Moreover, as emphasised by Stigler (1971),
regulation can be subverted by competitors who want to use it to deter entry or to
maintain cartels. Although motivated regulators might be more difficult to subvert
than judges, regulated industries have developed a range of techniques to turn
regulation into a mechanism of protecting their rents rather than public welfare.
The costs of dictatorship rise as those of disorder decline.
So what are the circumstances where regulation is the appropriate strategy of
enforcing good conduct? The basic implication of the theory that the resort to
regulation is only necessary when the level of disorder is too high for private orderings
and even courts to deal with successfully. This case is most compelling in situations
where the problem of inequality of weapons between private parties involved in a

Along these lines, Glaeser and Shleifer (2003) argue that The rise of the regulatory state in
the USA during the Progressive Era at the beginning of the twentieth century was a response to
the growing problems of subversion of courts by robber barons.

# Blackwell Publishing Ltd, 2005

Understanding Regulation 447

transaction is too severe. As indicated earlier, securities regulation is one instance

where this case for regulation based on the enforcement considerations is compelling.
Workplace safety is yet another. In contrast, entry of new firms is unlikely to require
regulatory control, and most labour regulations in environments that have competi-
tion for labour are difficult to justify on efficiency grounds.
The second implication is that the case for regulation is stronger when public abuse
of the private sector can be restrained. This prediction suggests that regulation
relative to doing nothing is a more attractive option in richer countries, where the
checks on the government are stronger. In contrast, regulation is a particularly poor
idea in undemocratic countries and in countries with extremely powerful executives,
where the risks of abuse are the greatest.
Finally, in some situations, nothing short of government ownership can address the
problem of disorder. If monopolies cannot be restrained through regulation, if quality
cannot be assured except with full state control, if public safety is jeopardised then
one can make a case for state ownership. For example, Hart et al. (1997) argue that
prisons might be properly publicly owned because the risk that private jailers mistreat
inmates, who after all have few legal rights and cannot count on the market or even
on regulation to protect them, is too high. Likewise, the military and the police tend to
be state-controlled because the risk of disorder from private control is unacceptable.
Although in some instances the case for government ownership as a means of
dealing with disorder is compelling, state ownership has the obvious problems of
public abuse and dictatorship. Because the government uses its control to pursue
political ends, the performance record of state enterprises around the world has been
dismal, and the benefits of privatisation large (La Porta and Lopez-de-Silanes, 1999;
La Porta et al., 2002; Megginson and Netter, 2001; Djankov and Murrell, 2002). The
dramatic failure of state socialism as an economic system is only the most remarkable
illustration of the problems of dictatorship taken to an extreme, where all economic
problems are solved to maintain political control of the communist party (Kornai,
1992). Although an efficiency case for state ownership as a means of controlling
disorder can be made, the range of activities where this case is compelling is modest.
This discussion completes a brief overview of alternative strategies of enforcement
of socially desirable conduct. I listed some of the ingredients of the efficient choice. In
the next section, I try to compare these implications to regulatory practice.

4. Regulatory Practice

In my presentation of the enforcement theory, I have focused on the costs and benefits of
alternative means of social control of business, and thereby pointed to what might be,
under different circumstances, the efficient choice. This focus on efficient institutional
choices has considerable descriptive and prescriptive power. Moreover, even efficient
institutional arrangements may exhibit significant levels of both disorder and dictator-
ship. As Coase (1960) argued long ago, the fact that a society is doing the best it can with
its institutional resources does not mean that all transaction costs are eliminated.
Still, in thinking about actual institutional choices, it is crucial to recognise that not
all we see is efficient. The first source on inefficiency is the bread-and-butter of public
choice theory (Buchanan and Tullock, 1962), namely the idea that politicians, once in
power, make economic policies and institutional choices to keep themselves in power
and, to the extent possible, to become rich. These theories generally point to a
tendency toward excessive dictatorship. We expect to see excessive centralisation,
# Blackwell Publishing Ltd, 2005
448 Andrei Shleifer

regulation, state ownership and so on even conditional on the trade-off a country

faces. Even in securities markets, we might see an excessive tendency toward state
control of stock exchanges and banks, and state restrictions on competition. The curse
of state socialism during the twentieth century illustrates this political tendency
toward dictatorship.
The second source of possible inefficiency of institutional choice is colonial trans-
plantation (La Porta et al., 1997, 1998). The legal and regulatory regimes of most
countries are not indigenous, but rather shaped by their colonial heritage. When the
English, the French, the Spaniards, the Dutch, the Germans, and the Portuguese
colonised much of the world (including the USA), they brought with them some of
their institutions. As in turns out, there is systematic variation among these institu-
tions of origin countries, shaped by their history over the last millennium (Glaeser and
Shleifer, 2002). England developed a common law tradition, characterised by the
independent judges and juries, relatively lower importance of statutory laws, and
the preference for private litigation as a means of addressing social problems.
France, in contrast, following the Romans, developed a civil law tradition, charac-
terised by state-employed judges, great importance of legal and procedural codes, and
a preference for state regulation over private litigation. Germany developed its own
civil law tradition, also based in Roman law. Finally, and crucially for the twentieth
century, the USSR developed its own system of socialist law.
Napoleon exported the French legal system through his conquests to Spain,
Portugal, and Holland, and through his and their own colonial conquests, it was
transplanted to all of Latin America, large parts of Europe, North and West Africa,
parts of the Caribbean, and parts of Asia. The common law tradition was trans-
planted by England to the USA, Canada, Australia, New Zealand, East Africa, large
parts of Asia (including India), and parts of the Caribbean. The German legal system
was voluntarily adopted in Japan, and through it Japan influenced the legal systems
of Korea, Taiwan, and China. Finally, the USSR transplanted its legal system to
socialist countries. These channels of transplantation suggest that there might be
systematic variation in property rights institutions among countries that is not a
consequence either of the pressures toward efficiency or of domestic political choice.
In the last several years, my collaborators and I completed several studies of state
ownership and regulation around the world. The areas we covered are the regulation
of entry by new firms, the regulation of judicial procedures in courts, and the
regulation of labour markets (Djankov et al., 2002, 2003a; Botero et al., 2004).
Although the data for each of these studies were collected using different procedures
covering somewhat different samples of countries, some systematic results emerge
from the analysis. Countries have pronounced styles of social control of business
intimately related to the origin of their laws. In all three areas of regulation entry,
courts, and labour socialist and French legal origin countries regulate activity more
heavily than do the common law countries. On average, the very same countries that
regulate entry also regulate courts and labour markets, and these correlations are at
least in part driven by legal origin. Moreover, the same rankings appear in the
measures of state ownership. Socialist and French legal origin countries have higher
levels of government ownership of banks (La Porta et al., 2002) and a greater role of
state-owned enterprises in the economy (La Porta et al., 1999) than do the common
law countries. This evidence suggests that colonial transplantation, rather than local
conditions, exerts a profound influence on national modes of social control of busi-
ness, including both state ownership and regulation.
# Blackwell Publishing Ltd, 2005
Understanding Regulation 449

This finding suggests that the observed institutional choices may well be inefficient. A
legal and regulatory system perfectly suitable for France might yield inefficiently high
levels of regulation and state ownership when transplanted to countries with fewer
checks on the government. Likewise, a system of independent courts that works in
Australia or the USA might fail in Malaysia or Zimbabwe. Indeed, the evidence on the
consequences of regulation shows that it is often excessive, especially in poor countries.
Higher levels of regulation of entry are associated with larger unofficial economies and
no measurable benefits for the quality of products (Djankov et al., 2002). Higher levels
of regulation of judicial procedures yield no benefits in simple disputes. In contrast,
more regulated legal systems appear to cost more and to produce higher delay, without
offsetting benefits in terms of perceived justice (Djankov et al., 2003a). Higher levels of
labour regulation are associated with larger unofficial economies, higher unemploy-
ment, and lower labour force participation (Botero et al., 2004).
The evidence on the importance of legal origin points to some tangible ways in
which the existing institutions fall short of their potential, as well as to some possible
directions of reform. In particular, the evidence suggests that deregulation particu-
larly in the areas such as entry and labour markets where the forces of competition are
potentially effective is a high level priority for poor countries. In these countries,
regulation is nearly universally associated with poor outcomes because public officials
abuse their power. Deregulation is likely to diminish dictatorship without a significant
increase in disorder.
But the evidence also points to some difficulties of reform. One cannot assume that,
in civil law countries, general jurisdiction courts could efficiently resolve disputes
these courts are too cumbersome to meet this goal. The most attractive areas for
deregulation in developing economies are those where competition and market dis-
cipline, rather than courts, can assure socially desirable outcomes and control dis-
order. In contrast, in the developed countries, courts especially specialised courts
are becoming an increasingly attractive alternative to regulation.

5. Conclusion

The framework presented here allows for a comparative analysis of institutions from
the perspective of the trade-off between dictatorship and disorder. This trade-off
looks different for different countries, and even for different activities within a
country. This trade-off can help organise the analysis of efficient institutional choice,
which recognises both the needs of a particular environment, and the constraints
imposed by a countrys political structure and institutional tradition. I apply this
framework to the example of regulation of securities markets, and argue that private
enforcement of public rules may emerge as an efficient strategy of social control of
these markets. Some empirical evidence assembled by La Porta et al. (2004) is broadly
consistent with this point of view. With more data about particular countries and
activities, one can use the framework described here to examine alternative strategies
of social control of business.

Allais, M., Le probleme de la planification economique dans une economie collectiviste,
Kyklos, Vol. II, 1947, pp. 4871.

# Blackwell Publishing Ltd, 2005

450 Andrei Shleifer

Barth, J., Caprio, G. and Levine, R., Bank supervision and regulation: what works best?
Journal of Financial Intermediation, Vol. 13, 2003, pp. 20548.
Bernstein, L., Opting out of the legal system: extralegal contractual relations in the diamond
industry, Journal of Legal Studies, Vol. 21, 1992, pp. 11557.
Botero, F., Djankov, S., La Porta, R., Lopez-de-Silanes, F. and Shleifer, A., The regulation of
labor, Quarterly Journal of Economics, Vol. 119, no. 4, 2004.
Buchanan, J. and Tullock, G., The Calculus of Consent, Logical Foundations of Constitutional
Democracy (Ann Arbor MI: University of Michigan Press, 1962).
Coase, R., The problem of social cost, Journal of Law and Economics, Vol. 3, 1960, pp. 144.
Coppin, C. and High, J., The Politics of Purity (Ann Arbor MI: University of Michigan Press, 1999).
Dal Bo, E., Dal Bo, P. and Di Tella, R., Plata o plomo? Bribe and punishment in a theory of
political influence, Working Paper (Harvard Business School, WP 03060, 2003).
Djankov, S., La Porta, R., Lopez-de-Silanes, F. and Shleifer, A., The regulation of entry,
Quarterly Journal of Economics, Vol. 117, no. 1, 2002, pp. 137.
Djankov, S., La Porta, R., Lopez-de-Silanes, F. and Shleifer, A., Courts, Quarterly Journal of
Economics, Vol. 118, no. 2, 2003a, pp. 453517.
Djankov, S., Glaeser, E., La Porta, R., Lopez-de-Silanes, F. and Shleifer, A., The new
comparative economics, Journal of Comparative Economics, Vol. 31, 2003b, pp. 595619.
Djankov, S. and Murrell, P., Enterprise restructuring: a quantitative survey, Journal of
Economic Literature, Vol. 40, no. 3, 2002, pp. 73992.
Ellickson, R., Order without Law (Cambridge: Harvard University Press, 1991).
Galanter, M., Why the haves come out ahead: speculations on the limits of legal change, Law
and Society Review, Vol. 9, 1974, pp. 95169.
Glaeser, E., Johnson, S. and Shleifer, A., Coase versus the Coasians, Quarterly Journal of
Economics, Vol. 116, no. 3, 2001, pp. 85399.
Glaeser, E., Scheinkman, J. and Shleifer, A. The injustice of inequality, Journal of Monetary
Economics, Vol. 50, 2003, pp. 199222.
Glaeser, E. and Shleifer, A., Legal origins, Quarterly Journal of Economics, Vol. 117, no. 4,
2002, pp. 11931229.
Glaeser, E. and Shleifer, A., The rise of the regulatory state, Journal of Economic Literature,
Vol. 41, no. 2, 2003, pp. 40125.
Greif, A., Reputation and coalitions in medieval trade: evidence on the Maghribi traders,
Journal of Economic History, Vol. 4, no. 94, 1989, pp. 85782.
Hart, O., Shleifer, A. and Vishny, R., The proper scope of government: theory and an
application to prisons, Quarterly Journal of Economics, Vol. 112, no. 4, 1997, 127161.
Hay, J. and Shleifer, A., Private enforcement of public laws: a theory of legal reform,
American Economic Review Papers and Proceedings, Vol. 88, no. 2, 1998, pp. 398403.
Hay, J., Shleifer, A. and Vishny, R., Toward a theory of legal reform, European Economic
Review, Vol. 40, nos. 35, 1996, pp. 55967.
Johnson, S., McMillan, J. and Woodruff, C., Property rights and finance, American Economic
Review, Vol. 92, no. 5, 2002, pp. 133556.
Kornai, J., The Socialist System: the Political Economy of Communism (Princeton NJ: Princeton
University Press, 1992).
Landis, J., The Administrative Process (Yale University Press, 1938).
La Porta, R. and Lopez-de-Silanes, F., The benefits of privatization: evidence from Mexico,
Quarterly Journal of Economics, Vol. 114, no. 4, 1999, pp. 11931242.
La Porta, R., Lopez-de-Silanes, F. and Shleifer, A., Government ownership of banks, Journal
of Finance, Vol. 57, no. 1, 2002, pp. 265301.
La Porta, R., Lopez-de-Silanes, F. and Shleifer, A., What works in securities laws (Harvard
University, Mimeo, 2004).
La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R., Legal determinants of external
finance, Journal of Finance, Vol. 52, no. 3, 1997, pp. 113150.

# Blackwell Publishing Ltd, 2005

Understanding Regulation 451

La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R., Law and finance, Journal of
Political Economy, Vol. 106, no. 6, 1998, pp. 111355.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R., The quality of government,
Journal of Law, Economics and Organization, Vol. 15, no. 1, 1999, pp. 22279.
Lewis, A., The Principles of Economic Planning (London: George Allen and Unwin, 1949).
Libecap, G., The rise of Chicago Packers and the origins of meat inspection and antitrust,
Economic Inquiry, Vol. 30, no. 2, 1992, pp. 24262.
Meade, J., Planning and the Price Mechanism: the Liberal Socialist Solution (London: George
Allen and Unwin, 1948).
Megginson, M. and Netter, J. M., From state to market: a survey of empirical studies on
privatization, Journal of Economic Literature, Vol. 39, no. 2, 2001, pp. 32189.
Peltzman, S., The economic theory of regulation after a decade of deregulation Brookings
Papers on Economic Activity, Special Issue, 1989, pp. 141.
Pigou, A., The Economics of Welfare, 4th ed. (London: Macmillan, 1938).
Pistor, K. and Xu, C., Law enforcement under incomplete law: theory and evidence from
financial market regulation, Mimeo (Columbia Law School, 2002).
Posner, R., A theory of negligence, Journal of Legal Studies, Vol. 1, no. 1, 1972, pp. 2996.
Posner, R., Theories of economic regulation, Bell Journal of Economics, Vol. 5, no. 2, 1974,
pp. 33558.
Posner, R., What do judges maximize? in Overcoming Law (Cambridge MA: Harvard
University Press, 1995, chapter 3).
Ramseyer, M. and Rasmusen, E., Judicial independence in the civil law regime: the evidence
from Japan, Journal of Law, Economics and Organization, Vol. 13, 1997, pp. 25986.
Shleifer, A., State versus private ownership, Journal of Economic Perspectives, Vol. 12, no. 4,
1998, pp. 13350.
Shleifer, A. and Vishny, R., The Grabbing Hand: Government Pathologies and their Cures
(Cambridge MA: Harvard University Press, 1998).
Stigler, G. J., The theory of economic regulation, Bell Journal of Economics, Vol. 2, 1971,
pp. 321.

# Blackwell Publishing Ltd, 2005

Arend Lijphart

Arend Lijphart is Research Professor Emeritus of Political Science at

the University of California, San Diego. He is the author of Patterns of
Democracy: Government Forms and Performance in Thirty-Six Coun-
tries (1999) and many other studies of democratic institutions, the
governance of deeply divided societies, and electoral systems.

Over the past half-century, democratic constitutional design has un-

dergone a sea change. After the Second World War, newly independent
countries tended simply to copy the basic constitutional rules of their
former colonial masters, without seriously considering alternatives.
Today, constitution writers choose more deliberately among a wide
array of constitutional models, with various advantages and disadvan-
tages. While at first glance this appears to be a beneficial development,
it has actually been a mixed blessing: Since they now have to deal with
more alternatives than they can readily handle, constitution writers risk
making ill-advised decisions. In my opinion, scholarly experts can be
more helpful to constitution writers by formulating specific recommen-
dations and guidelines than by overwhelming those who must make the
decision with a barrage of possibilities and options.
This essay presents a set of such recommendations, focusing in par-
ticular on the constitutional needs of countries with deep ethnic and
other cleavages. In such deeply divided societies the interests and de-
mands of communal groups can be accommodated only by the
establishment of power sharing, and my recommendations will indicate
as precisely as possible which particular power-sharing rules and insti-
tutions are optimal and why. (Such rules and institutions may be useful
in less intense forms in many other societies as well.)
Most experts on divided societies and constitutional engineering
broadly agree that deep societal divisions pose a grave problem for
democracy, and that it is therefore generally more difficult to establish
and maintain democratic government in divided than in homogeneous

Journal of Democracy Volume 15, Number 2 April 2004

Arend Lijphart 97

countries. The experts also agree that the problem of ethnic and other
deep divisions is greater in countries that are not yet democratic or
fully democratic than in well-established democracies, and that such
divisions present a major obstacle to democratization in the twenty-
first century. On these two points, scholarly agreement appears to be
A third point of broad, if not absolute, agreement is that the success-
ful establishment of democratic government in divided societies
requires two key elements: power sharing and group autonomy. Power
sharing denotes the participation of representatives of all significant
communal groups in political decision making, especially at the ex-
ecutive level; group autonomy means that these groups have authority
to run their own internal affairs, especially in the areas of education
and culture. These two characteristics are the primary attributes of the
kind of democratic system that is often referred to as power-sharing
democracy or, to use a technical political-science term, consocia-
tional democracy. 1 A host of scholars have analyzed the central role
of these two features and are sympathetic to their adoption by divided
societies.2 But agreement extends far beyond the consociational school.
A good example is Ted Robert Gurr, who in Minorities at Risk: A Glo-
bal View of Ethnopolitical Conflicts clearly does not take his
inspiration from consociational theory (in fact, he barely mentions it),
but based on massive empirical analysis reaches the conclusion that
the interests and demands of communal groups can usually be accom-
modated by some combination of the policies and institutions of
autonomy and power sharing. 3
The consensus on the importance of power sharing has recently been
exemplified by commentators reactions to the creation of the Govern-
ing Council in Iraq: The Council has been criticized on a variety of
grounds, but no one has questioned its broadly representative composi-
tion. The strength of the power-sharing model has also been confirmed
by its frequent practical applications. Long before scholars began ana-
lyzing the phenomenon of power-sharing democracy in the 1960s,
politicians and constitution writers had designed power-sharing solu-
tions for the problems of their divided societies (for example, in Austria,
Canada, Colombia, Cyprus, India, Lebanon, Malaysia, the Netherlands,
and Switzerland). Political scientists merely discovered what political
practitioners had repeatedlyand independently of both academic ex-
perts and one anotherinvented years earlier.

Critics of Power Sharing

The power-sharing model has received a great deal of criticism since
it became a topic of scholarly discourse three decades ago. Some crit-
ics have argued that power-sharing democracy is not ideally democratic
98 Journal of Democracy

or effective; others have focused on methodological and measurement

issues.4 But it is important to note that very few critics have presented
serious alternatives to the power-sharing model. One exception can be
found in the early critique by Brian Barry, who in the case of Northern
Ireland recommended cooperation without cooptationstraightfor-
ward majority rule in which both majority and minority would simply
promise to behave moderately. 5 Barrys proposal would have meant
that Northern Irelands Protestant majority, however moderate, would
be in power permanently, and that the Catholic minority would always
play the role of the loyal opposition. Applied to the case of the Iraqi
Governing Council, Barrys alternative to power sharing would call
for a Council composed mainly or exclusively of moderate members of
the Shiite majority, with the excluded Sunnis and Kurds in opposi-
tion. This is a primitive solution to ethnic tensions and extremism, and
it is nave to expect minorities condemned to permanent opposition to
remain loyal, moderate, and constructive. Barrys suggestion therefore
cannot beand, in practice, has not beena serious alternative to
power sharing.
The only other approach that has attracted considerable attention
is Donald L. Horowitzs proposal to design various electoral mecha-
nisms (especially the use of the alternative vote or instant runoff)
that would encourage the election of moderate representatives. 6 It re-
sembles Barrys proposal in that it aims for moderation rather than
broad representation in the legislature and the executive, except that
Horowitz tries to devise a method to induce the moderation that Barry
simply hopes for. If applied to the Iraqi Governing Council, Horowitzs
model would generate a body consisting mainly of members of the
Shiite majority, with the proviso that most of these representatives
would be chosen in such a way that they would be sympathetic to the
interests of the Sunni and Kurdish minorities. It is hard to imagine
that, in the long run, the two minorities would be satisfied with this
kind of moderate Shiite representation, instead of representation by
members of their own communities. And it is equally hard to imagine
that Kurdish and Sunni members of a broadly representative constitu-
ent assembly would ever agree to a constitution that would set up
such a system.
Horowitzs alternative-vote proposal suffers from several other weak-
nesses, but it is not necessary to analyze them in this article. 7 The
main point that is relevant here is that it has found almost no support
from either academic experts or constitution writers. Its sole, and only
partial, practical application to legislative elections in an ethnically
divided society was the short-lived and ill-fated Fijian constitutional
system, which tried to combine the alternative vote with power-shar-
ing; it was adopted in 1999 and collapsed in 2000. 8 With all due
respect to the originality of his ideas and the enthusiasm with which
Arend Lijphart 99

he has defended them, Horowitzs arguments do not seem to have

sparked a great deal of assent or emulation. 9

One Size Fits All?

In sum, power sharing has proven to be the only democratic model that
appears to have much chance of being adopted in divided societies, which
in turn makes it unhelpful to ask constitution writers to contemplate
alternatives to it. More than enough potential confusion and distraction
are already inherent in the consideration of the many alternatives within
power sharing. Contrary to Horowitzs claim that power-sharing democ-
racy is a crude one size fits all model,10 the power-sharing systems
adopted prior to 1960 (cited earlier), as well as more recent cases (such as
Belgium, Bosnia, Czechoslovakia, Northern Ireland, and South Africa),
show enormous variation. For example, broad representation in the ex-
ecutive has been achieved by a constitutional requirement that it be
composed of equal numbers of the two major ethnolinguistic groups (Bel-
gium); by granting all parties with a minimum of 5 percent of the legislative
seats the right to be represented in the cabinet (South Africa, 199499);
by the equal representation of the two main parties in the cabinet and an
alternation between the two parties in the presidency (Colombia, 1958
64); and by permanently earmarking the presidency for one group and
the prime ministership for another (Lebanon).
All of these options are not equally advantageous, however, and do
not work equally well in practice, because the relative success of a
power-sharing system is contingent upon the specific mechanisms de-
vised to yield the broad representation that constitutes its core. In fact,
the biggest failures of power-sharing systems, as in Cyprus in 1963 and
Lebanon in 1975, must be attributed not to the lack of sufficient power
sharing but to constitution writers choice of unsatisfactory rules and
These failures highlight the way in which scholarly experts can help
constitution writers by developing recommendations regarding power-
sharing rules and institutions. In this sense, Horowitzs one size fits
all charge should serve as an inspiration to try to specify the optimal
form of power sharing. While the power-sharing model should be adapted
according to the particular features of the country at hand, it is not true
that everything depends on these individual characteristics. In the fol-
lowing sections I outline nine areas of constitutional choice and provide
my recommendations in each area. These constitute a one size power-
sharing model that offers the best fit for most divided societies regardless
of their individual circumstances and characteristics.

1) The legislative electoral system. The most important choice facing

constitution writers is that of a legislative electoral system, for which
100 Journal of Democracy

the three broad categories are proportional representation (PR),

majoritarian systems, and intermediate systems. For divided societies,
ensuring the election of a broadly representative legislature should be
the crucial consideration, and PR is undoubtedly the optimal way of
doing so.
Within the category of majoritarian systems, a good case could be
made for Horowitzs alternative-vote proposal, which I agree is superior
to both the plurality method and the two-ballot majority runoff. 11 Nev-
ertheless, there is a scholarly consensus against majoritarian systems in
divided societies. As Larry Diamond explains:

If any generalization about institutional design is sustainable . . . it is that

majoritarian systems are ill-advised for countries with deep ethnic, re-
gional, religious, or other emotional and polarizing divisions. Where
cleavage groups are sharply defined and group identities (and intergroup
insecurities and suspicions) deeply felt, the overriding imperative is to
avoid broad and indefinite exclusion from power of any significant group.12

The intermediate category can be subdivided further into semi-pro-

portional systems, mixed systems, and finally, majoritarian systems
that offer guaranteed representation to particular minorities. Semi-pro-
portional systemslike the cumulative and limited vote (which have
been primarily used at the state and local levels in the United States) and
the single nontransferable vote (used in Japan until 1993)13may be
able to yield minority representation, but never as accurately and con-
sistently as PR. Unlike these rare semi-proportional systems, mixed
systems have become quite popular since the early 1990s.14 In some of
the mixed systems (such as Germanys and New Zealands) the PR com-
ponent overrides the plurality component, and these should therefore be
regarded not as mixed but as PR systems. To the extent that the PR
component is not, or is only partly, compensatory (as in Japan, Hungary,
and Italy), the results will necessarily be less than fully proportional
and minority representation less accurate and secure. Plurality combined
with guaranteed representation for specified minorities (as in India and
Lebanon) necessarily entails the potentially invidious determination of
which groups are entitled to guaranteed representation and which are
not. In contrast, the beauty of PR is that in addition to producing propor-
tionality and minority representation, it treats all groupsethnic, racial,
religious, or even noncommunal groupsin a completely equal and
evenhanded fashion. Why deviate from full PR at all?

2) Guidelines within PR. Once the choice is narrowed down to PR,

constitution writers need to settle on a particular type within that sys-
tem. PR is still a very broad category, which spans a vast spectrum of
complex possibilities and alternatives. How can the options be nar-
rowed further? I recommend that highest priority be given to the
Arend Lijphart 101

selection of a PR system that is simple to understand and operatea

criterion that is especially important for new democracies. From that
simplicity criterion, several desiderata can be derived: a high, but not
necessarily perfect, degree of proportionality; multimember districts
that are not too large, in order to avoid creating too much distance
between voters and their representatives; list PR, in which parties present
lists of candidates to the voters, instead of the rarely used single trans-
ferable vote, in which voters have to rank order individual candidates;
and closed or almost closed lists, in which voters mainly choose parties
instead of individual candidates within the list. List PR with closed
lists can encourage the formation and maintenance of strong and cohe-
sive political parties.
One attractive model along these lines is the list-PR system used in
Denmark, which has 17 districts that elect an average of eight represen-
tatives each from partly open lists. The districts are small enough for
minority parties with more than 8 percent of the vote to stand a good
chance of being elected.15 In addition to the 135 representatives elected
in these districts, there are 40 national compensatory seats that are ap-
portioned to parties (with a minimum of 2 percent of the national vote)
in a way that aims to maximize overall national proportionality.16 The
Danish model is advantageous for divided societies, because the com-
pensatory seats plus the low 2 percent threshold give small minorities
that are not geographically concentrated a reasonable chance to be rep-
resented in the national legislature. While I favor the idea of maximizing
proportionality, however, this system does to some extent detract from
the goal of keeping the electoral system as simple and transparent as
possible. Moreover, national compensatory seats obviously make little
sense in those divided societies where nationwide parties have not yet

3) Parliamentary or presidential government. The next important deci-

sion facing constitution writers is whether to set up a parliamentary,
presidential, or semi-presidential form of government. In countries with
deep ethnic and other cleavages, the choice should be based on the
different systems relative potential for power sharing in the executive.
As the cabinet in a parliamentary system is a collegial decision-making
bodyas opposed to the presidential one-person executive with a purely
advisory cabinetit offers the optimal setting for forming a broad power-
sharing executive. A second advantage of parliamentary systems is that
there is no need for presidential elections, which are necessarily
majoritarian in nature. As Juan Linz states in his well-known critique of
presidential government, perhaps the most important implication of
presidentialism is that it introduces a strong element of zero-sum game
into democratic politics with rules that tend toward a winner-take-all
outcome.17 Presidential election campaigns also encourage the poli-
102 Journal of Democracy

tics of personality and overshadow the politics of competing parties

and party programs. In representative democracy, parties provide the
vital link between voters and the government, and in divided societies
they are crucial in voicing the interests of communal groups. Seymour
Martin Lipset has recently emphasized this point again by calling
political parties indispensable in democracies and by recalling E.E.
Schattschneiders famous pronouncement that modern democracy is
unthinkable save in terms of parties.18
Two further problems of presidentialism emphasized by Linz are fre-
quent executive-legislative stalemates and the rigidity of presidential
terms of office. Stalemates are likely to occur because president and
legislature can both claim the democratic legitimacy of being popu-
larly elected, but the president and the majority of the legislature may
belong to different parties or may have divergent preferences even if
they belong to the same party. The rigidity inherent in presidentialism
is that presidents are elected for fixed periods that often cannot be ex-
tended because of term limits, and that cannot easily be shortened even
if the president proves to be incompetent, becomes seriously ill, or is
beset by scandals of various kinds. Parliamentary systems, with their
provisions for votes of confidence, snap elections, and so on, do not
suffer from this problem.
Semi-presidential systems represent only a slight improvement over
pure presidentialism. Although there can be considerable power shar-
ing among president, prime minister, and cabinet, the zero-sum nature
of presidential elections remains. Semi-presidential systems actually
make it possible for the president to be even more powerful than in most
pure presidential systems. In France, the best-known example of semi-
presidentialism, the president usually exercises predominant power; the
196274 and 198186 periods have even been called hyperpresiden-
tial phases.19 The stalemate problem is partly solved in semi-presidential
systems by making it possible for the system to shift from a mainly
presidential to a mainly parliamentary mode if the president loses the
support of his party or governing coalition in the legislature. In the
Latin American presidential democracies, constitutional reformers have
often advocated semi-presidential instead of parliamentary government,
but only for reasons of convenience: A change to parliamentarism seems
too big a step in countries with strong presidentialist traditions. While
such traditional and sentimental constraints may have to be taken into
account in constitutional negotiations, parliamentary government should
be the general guideline for constitution writers in divided societies.
There is a strong scholarly consensus in favor of parliamentary gov-
ernment. In the extensive literature on this subject, the relatively few
critics have questioned only parts of the pro-parliamentary consensus.
Pointing to the case of U.S. presidentialism, for instance, they have
noted that the stalemate problem has not been as serious as Linz and
Arend Lijphart 103

others have allegedwithout, however, challenging the validity of the

other charges against presidential government.20

4) Power sharing in the executive. The collegial cabinets in parliamen-

tary systems facilitate the formation of power-sharing executives, but
they do not by themselves guarantee that power sharing will be insti-
tuted. Belgium and South Africa exemplify the two principal methods of
doing so. In Belgium, the constitution stipulates that the cabinet must
comprise equal numbers of Dutch-speakers and French-speakers. The
disadvantage of this approach is that it requires specifying the groups
entitled to a share in power, and hence the same discriminatory choices
inherent in electoral systems with guaranteed representation for particu-
lar minorities. In South Africa there was so much disagreement and
controversy about racial and ethnic classifications that these could not
be used as a basis for arranging executive power sharing in the 1994
interim constitution. Instead, power sharing was mandated in terms of
political parties: Any party, ethnic or not, with a minimum of 5 percent
of the seats in parliament was granted the right to participate in the
cabinet on a proportional basis.21 For similar situations in other coun-
tries, the South African solution provides an attractive model. But when
there are no fundamental disagreements about specifying the ethnic
groups entitled to a share of cabinet power, the Belgian model has two
important advantages. First, it allows for power sharing without mandat-
ing a grand coalition of all significant parties and therefore without
eliminating significant partisan opposition in parliament. Second, it al-
lows for slight deviation from strictly proportional power sharing by
giving some overrepresentation to the smaller groups, which may be
desirable in countries where an ethnic majority faces one or more ethnic
minority groups.

5) Cabinet stability. Constitution writers may worry about one potential

problem of parliamentary systems: The fact that cabinets depend on ma-
jority support in parliament and can be dismissed by parliamentary votes
of no-confidence may lead to cabinet instabilityand, as a result, re-
gime instability. The weight of this problem should not be overestimated;
the vast majority of stable democracies have parliamentary rather than
presidential or semi-presidential forms of government.22 Moreover, the
position of cabinets vis-`a-vis legislatures can be strengthened by consti-
tutional provisions designed to this effect. One such provision is the
constructive vote of no confidence, adopted in the 1949 constitution of
West Germany, which stipulates that the prime minister (chancellor) can
be dismissed by parliament only if a new prime minister is elected simul-
taneously. This eliminates the risk of a cabinet being voted out of office
by a negative legislative majority that is unable to form an alternative
cabinet. Spain and Papua New Guinea have adopted similar requirements
104 Journal of Democracy

for a constructive vote of no confidence. The disadvantage of this provi-

sion is that it may create an executive that cannot be dismissed by
parliament but does not have a parliamentary majority to pass its legisla-
tive programthe same kind of stalemate that plagues presidential
systems. A suggested solution to this potential problem was included in
the 1958 constitution of the French Fifth Republic in the form of a provi-
sion that the cabinet has the right to make its legislative proposals matters
of confidence, and these proposals are adopted automatically unless an
absolute majority of the legislature votes to dismiss the cabinet. No con-
stitution has yet tried to combine the German and French rules, but such
a combination could undoubtedly give strong protection to cabinets and
their legislative effectivenesswithout depriving the parliamentary
majority of its fundamental right to dismiss the cabinet and replace it
with a new one in which parliament has greater confidence.

6) Selecting the head of state. In parliamentary systems, the prime min-

ister usually serves only as head of government, while a constitutional
monarch or a mainly ceremonial president occupies the position of head
of state. Assuming that no monarch is available, constitution writers
need to decide how the president should be chosen. My advice is two-
fold: to make sure that the presidency will be a primarily ceremonial
office with very limited political power, and not to elect the president by
popular vote. Popular election provides democratic legitimacy and, es-
pecially in combination with more than minimal powers specified in the
constitution, can tempt presidents to become active political partici-
pantspotentially transforming the parliamentary system into a semi-
presidential one. The preferable alternative is election by parliament.
A particularly attractive model was the constitutional amendment
proposed as part of changing the Australian parliamentary system from
a monarchy to a republic, which specified that the new president would
be appointed on the joint nomination of the prime minister and the
leader of the opposition, and confirmed by a two-thirds majority of a
joint session of the two houses of parliament. The idea behind the two-
thirds rule was to encourage the selection of a president who would be
nonpartisan and nonpolitical. (Australian voters defeated the entire pro-
posal in a 1999 referendum mainly because a majority of the
pro-republicans stronglyand unwiselypreferred the popular elec-
tion of the president.) In my opinion, the best solution is the South
African system of not having a separate head of state at all: There the
president is in fact mainly a prime minister, subject to parliamentary
confidence, who simultaneously serves as head of state.

7) Federalism and decentralization. For divided societies with geo-

graphically concentrated communal groups, a federal system is
undoubtedly an excellent way to provide autonomy for these groups.
Arend Lijphart 105

My specific recommendation regards the second (federal) legislative

chamber that is usually provided for in federal systems. This is often a
politically powerful chamber in which less populous units of the federa-
tion are overrepresented (consider, for example, the United States Senate,
which gives two seats to tiny Wyoming as well as gigantic California).
For parliamentary systems, two legislative chambers with equal, or sub-
stantially equal, powers and different compositions is not a workable
arrangement: It makes too difficult the forming of cabinets that have the
confidence of both chambers, as the 1975 Australian constitutional cri-
sis showed: The opposition-controlled Senate refused to pass the budget
in an attempt to force the cabinets resignation, although the cabinet
continued to have the solid backing of the House of Representatives.
Moreover, a high degree of smaller-unit overrepresentation in the fed-
eral chamber violates the democratic principle of one person, one vote.
In this respect, the German and Indian federal models are more attractive
than the American, Swiss, and Australian ones.
Generally, it is advisable that the federation be relatively decentral-
ized and that its component units (states or provinces) be relatively
smallboth to increase the prospects that each unit will be relatively
homogeneous and to avoid dominance by large states on the federal
level. Beyond this, a great many decisions need to be made regarding
details that will vary from country to country (such as exactly where the
state boundaries should be drawn). Experts have no clear advice to offer
on how much decentralization is desirable within the federation, and
there is no consensus among them as to whether the American, Cana-
dian, Indian, Australian, German, Swiss, or Austrian model is most worthy
of being emulated.

8) Nonterritorial autonomy. In divided societies where the communal

groups are not geographically concentrated, autonomy can also be ar-
ranged on a nonterritorial basis. Where there are significant religious
divisions, for example, the different religious groups are often intent on
maintaining control of their own schools. A solution that has worked
well in India, Belgium, and the Netherlands is to provide educational
autonomy by giving equal state financial support to all schools, public
and private, as long as basic educational standards are met. While this
goes against the principle of separating church and state, it allows for
the state to be completely neutral in matters of education.

9) Power sharing beyond the cabinet and parliament. In divided soci-

eties, broad representation of all communal groups is essential not only
in cabinets and parliaments, but also in the civil service, judiciary,
police, and military. This aim can be achieved by instituting ethnic or
religious quotas, but these do not necessarily have to be rigid. For ex-
ample, instead of mandating that a particular group be given exactly 20
106 Journal of Democracy

percent representation, a more flexible rule could specify a target of 15

to 25 percent. I have found, however, that such quotas are often unnec-
essary; it is sufficient to have an explicit constitutional provision in
favor of the general objective of broad representation and to rely on the
power-sharing cabinet and the proportionally constituted parliament
for the practical implementation of this goal.

Other Issues
As far as several other potentially contentious issues are concerned,
my advice would be to start out with the modal patterns found in the
worlds established democracies, such as a two-thirds majority require-
ment for amending the constitution (with possibly a higher threshold
for amending minority rights and autonomy), a size of the lower house
of the legislature that is approximately the cube root of the countrys
population size23 (which means that a country with about 25 million
inhabitants, such as Iraq, should have a lower house of about 140
representatives), and legislative terms of four years.
While approval by referendum can provide the necessary democratic
legitimacy for a newly drafted constitution, I recommend a constitu-
tional provision to limit the number of referenda. One main form of
referendum entails the right to draft legislation and constitutional amend-
ments by popular initiative and to force a direct popular vote on such
propositions. This is a blunt majoritarian instrument that may well be
used against minorities. On the other hand, the Swiss example has shown
that a referendum called by a small minority of voters to challenge a law
passed by the majority of the elected representatives may have the desir-
able effect of boosting power sharing. Even if the effort fails, it forces the
majority to pay the cost of a referendum campaign; hence the potential
calling of a referendum by a minority is a strong stimulus for the major-
ity to be heedful of minority views. Nevertheless, my recommendation is
for extreme caution with regard to referenda, and the fact that frequent
referenda occur in only three democraciesthe United States, Switzer-
land, and, especially since about 1980, Italyunderscores this guideline.
Constitution writers will have to resolve many other issues that I
have not mentioned, and on which I do not have specific recommenda-
tions: for example, the protection of civil rights, whether to set up a
special constitutional court, and how to make a constitutional or su-
preme court a forceful protector of the constitution and of civil rights
without making it too interventionist and intrusive. And as constitu-
tion writers face the difficult and time-consuming task of resolving
these issues, it is all the more important that experts not burden or
distract them with lengthy discussions on the relative advantages and
disadvantages of flawed alternatives like presidentialism and non-PR
Arend Lijphart 107

I am not arguing that constitution writers should adopt all my recom-

mendations without any examination of various alternatives. I recognize
that the interests and agendas of particular parties and politicians may
make them consider other alternatives, that a countrys history and tradi-
tions will influence those who must draft its basic law, and that professional
advice is almost alwaysand very wiselysought from more than one
constitutional expert. Even so, I would contend that my recommenda-
tions are not merely based on my own preferences, but on a strong
scholarly consensus and solid empirical evidence, and that at the very
least they should form a starting point in constitutional negotiations.


I am grateful to the Bellagio Study and Conference Center of the Rockefeller

Foundation for offering me the opportunity to work on this project while I was a
resident of the Center in MayJune 2003, and to Roberto Belloni, Torbjrn Bergman,
Joseph H. Brooks, Florian Bieber, Jorgen Elklit, Svante Ersson, John McGarry,
Brendan OLeary, Mogens N. Pedersen, Hugh B. Price, and Timothy D. Sisk for
their valuable advice. Some of the ideas presented in this article were first pub-
lished in my chapter The Wave of Power-Sharing Democracy, in Andrew
Reynolds, ed., The Architecture of Democracy: Constitutional Design, Conflict
Management, and Democracy (Oxford: Oxford University Press, 2002), 3754;
and in Democracy in the Twenty-First Century: Can We Be Optimistic? Uhlenbeck
Lecture No. 18 (Wassenaar: Netherlands Institute for Advanced Study, 2000).

1. The secondary characteristics are proportionality, especially in legislative

elections (in order to ensure a broadly representative legislaturesimilar to the
aim of effecting a broadly constituted executive) and a minority veto on the most
vital issues that affect the rights and autonomy of minorities.

2. Some of these scholars are Dirk Berg-Schlosser, William T. Bluhm, Laurence

J. Boulle, Hans Daalder, Edward Dew, Robert H. Dix, Alan Dowty, Jonathan
Fraenkel, Hermann Giliomee, Theodor Hanf, Jonathan Hartlyn, Martin O. Heisler,
Luc Huyse, Thomas A. Koelble, Gerhard Lehmbruch, Franz Lehner, W. Arthur
Lewis, Val R. Lorwin, Diane K. Mauzy, John McGarry, Kenneth D. McRae, Antoine
N. Messarra, R.S. Milne, S.J.R. Noel, Eric A. Nordlinger, Brendan OLeary, G.
Bingham Powell, Jr., Andrew Reynolds, F. van Zyl Slabbert, Jrg Steiner, Albert J.
Venter, Karl von Vorys, David Welsh, and Steven B. Wolinetz. Their most impor-
tant writings on the subject (if published before the mid-1980s) can be found in the
bibliography of Arend Lijphart, Power-Sharing in South Africa (Berkeley, Calif.:
Institute of International Studies, University of California, 1985), 13771.

3. Ted Robert Gurr, Minorities at Risk: A Global View of Ethnopolitical Con-

flicts (Washington, D.C.: U.S. Institute of Peace Press, 1993), 292, italics added.

4. I have responded to these criticisms at length elsewhere. See especially Lijphart,

The Wave of Power-Sharing Democracy, in Andrew Reynolds, ed., The Archi-
tecture of Democracy: Constitutional Design, Conflict Management, and Democracy
(Oxford: Oxford University Press, 2002), 4047; and Lijphart, Power-Sharing in
South Africa, 83117.

5. Brian Barry, The Consociational Model and Its Dangers, European Jour-
nal of Political Research 3 (December 1975): 406.

6. Donald L. Horowitz, A Democratic South Africa? Constitutional Engineer-

108 Journal of Democracy

ing in a Divided Society (Berkeley, Calif.: University of California Press, 1991),

188203; and Electoral Systems: A Primer for Decision Makers, Journal of
Democracy 14 (October 2003): 12223. In alternative-vote systems, voters are
asked to rank order the candidates. If a candidate receives an absolute majority of
first preferences, he or she is elected; if not, the weakest candidate is eliminated,
and the ballots are redistributed according to second preferences. This process
continues until one of the candidates receives a majority of the votes.

7. For a detailed critique, see Lijphart, The Alternative Vote: A Realistic Alter-
native for South Africa? Politikon 18 (June 1991): 9101; and Lijphart,
Multiethnic Democracy, in Seymour Martin Lipset, ed., The Encyclopedia of
Democracy (Washington, D.C.: Congressional Quarterly, 1995), 86364.

8. The alternative vote was also used for the 1982 and 1988 presidential elec-
tions in Sri Lanka and for the 2000 presidential elections in the Republika Srpska
in Bosnia. Nigeria has used a similar system favored by Horowitz (requiring a
plurality plus at least 25 percent of the votes in at least two-thirds of the states for
victory) for its presidential elections. The third and sixth guidelines that I describe
in the present essay recommend a parliamentary system without a popularly elected
presidentand therefore no direct presidential elections at all.

9. Benjamin Reilly has come to Horowitzs defense, but only with significant
qualifications; for instance, Reilly dissents from Horowitzs advocacy of the alter-
native vote for the key case of South Africa. See Reilly, Democracy in Divided
Societies: Electoral Engineering for Conflict Management (Cambridge: Cambridge
University Press, 2001). Andreas Wimmer advocates the alternative vote for Iraq
in Democracy and Ethno-Religious Conflict in Iraq, Survival 45 (Winter 2003
2004): 11134.

10. Donald L. Horowitz, Constitutional Design: Proposals versus Processes,

in Andrew Reynolds, ed., The Architecture of Democracy, 25.

11. In contrast with plurality, the alternative vote (instant runoff) ensures that
the winning candidate has been elected by a majority of the voters, and it does so
more accurately than the majority-runoff method and without the need for two
rounds of voting.

12. Larry Diamond, Developing Democracy: Toward Consolidation (Baltimore:

Johns Hopkins University Press, 1999), 104.

13. All three of these systems use multimember election districts. The cumula-
tive vote resembles multi-member district plurality in which each voter has as
many votes as there are seats in a district, but, unlike plurality, the voter is allowed
to cumulate his or her vote on one or a few of the candidates. In limited-vote
systems, voters have fewer votes than the number of district seats. The single
nontransferable vote is a special case of the limited vote in which the number of
votes cast by each voter is reduced to one.

14. See Matthew Soberg Shugart and Martin P. Wattenberg, eds., Mixed-Mem-
ber Electoral Systems: The Best of Both Worlds? (Oxford: Oxford University Press,

15. This estimate is based on the T=75%(M+1) equationin which T is the

effective threshold and M the number of representatives elected in a district
suggested by Rein Taagepera; see Arend Lijphart, Electoral Systems, in Seymour
Martin Lipset, ed., Encyclopedia of Democracy, 417. There is considerable varia-
tion around the average of 8 representatives per district, but 9 of the 17 districts are
very close to this average, with between 6 and 9 seats. The open-list rules are very
complex and, in my opinion, make the lists too open. In addition to the 175 seats
described here, Greenland and the Faeroe Islands elect two representatives each. I
Arend Lijphart 109

should also point out that my recommendation of the Danish model entails a bit of
a paradox: It is a system that is very suitable for ethnically and religiously divided
countries, although Denmark itself happens to be one of the most homogeneous
countries in the world.

16. Parties below the 2 percent threshold may still benefit from the compensa-
tory seats if certain other requirements are met, such as winning at least one district

17. Juan J. Linz, Presidential or Parliamentary Democracy: Does It Make a

Difference? in Juan J. Linz and Arturo Valenzuela, eds., The Failure of Presiden-
tial Democracy (Baltimore: Johns Hopkins University Press, 1994), 18.

18. Seymour Martin Lipset, The Indispensability of Political Parties, Journal

of Democracy 11 (January 2000): 4855; E.E. Schattschneider, Party Government
(New York: Rinehart, 1942), 1.

19. John T.S. Keeler and Martin A. Schain, Institutions, Political Poker, and
Regime Evolution in France, in Kurt von Mettenheim, ed., Presidential Institu-
tions and Democratic Politics: Comparing Regional and National Contexts
(Baltimore: Johns Hopkins University Press, 1997), 9597. Horowitz favors a
president elected by the alternative vote or a similar vote-pooling method, but in
other respects his president does not differ from presidents in pure presidential
systems; see his A Democratic South Africa?, 20514.

20. Scholars have also indicated methods to minimize the problem of presiden-
tial-legislative deadlockfor instance, by holding presidential and legislative
elections concurrently and electing the president by plurality instead of the more
usual majority-runoff method. Such measures may indeed be able to ameliorate the
problem to some extent, but cannot solve it entirely. See Matthew Soberg Shugart
and John M. Carey, Presidents and Assemblies: Constitutional Design and Elec-
toral Dynamics (Cambridge: Cambridge University Press, 1992); and Mark P.
Jones, Electoral Laws and the Survival of Presidential Democracies (Notre Dame:
University of Notre Dame Press, 1995).

21. The 1998 Good Friday Agreement provides for a similar power-sharing
executive for Northern Ireland.

22. In my comparative study of the worlds stable democracies, defined as

countries that were continuously democratic from 1977 to 1996 (and had popula-
tions greater than 250,000), 30 of the 36 stable democracies had parliamentary
systems. See Lijphart, Patterns of Democracy: Government Forms and Perfor-
mance in Thirty-Six Countries (New Haven: Yale University Press, 1999).

23. This pattern was discovered by Rein Taagepera; see his The Size of Na-
tional Assemblies, Social Science Research 1 (December 1972): 38540.
Free and Fair Elections
An international round table
on election standards
organized by the Inter-Parliamentary Union,
Geneva, November 2004

(Ed. Michael D. Boda)

Democratic Principles and Judging
Free and Fair *
Richard S. Katz
Guy Goodwin-Gills (1994) Free and Fair Elections : International Law
and Practice was a path-breaking inquiry into a crucial question. The
book was significant in two respects. On the one hand, it was a
renewal of the attempt (e.g., Mackenzie, 1958) to distill from the
practices of the established democracies those standards that are
essential to free and fair elections. On the other hand, Goodwin-Gill
grounded his analysis specifically in international law, suggesting that
the international community would be entitled to take an interest in
the adherence of individual states to the standards.

As with all path-breaking studies, experience has shown that Free and
Fair Elections provides only partial answers to some of its questions, and
indeed raises additional questions whose importance only became
apparent later. Other participants addressed some of these. I address
two questions here. The first is whether democracy has a sufficiently
clear and unambiguous definition that free and fair elections can be
assessed on the basis of a uniform set of standards, or whether,
alternatively, there are still many competing understandings of
democracy, each with its own twist on the meaning of free and fair.
The second is whether, or more properly to what extent and how, the
two standards of free-ness and fair-ness are compatible.

Democracy and elections

Democracy is a messy concept, and there has been endless academic
debate concerning its true meaning. The definition of democracy is
not just a philosophical question, however, but also a question with
profound implications in the world of practical politics. Different
conceptions of democracy justify different institutional arrangements
and different standards for evaluating their performance and
ultimately differing distributions of authority. Who wins may be
determined by the rules and practices in place, and those, in turn,
depend at least in part on which understanding of democracy is

* This paper was presented initially in November 2004 as part of the

proceedings of Free and Fair Elections, Ten Years On : An International Round
Table on Election Standards. The author wishes to thank those who offered
comment on its contents, particularly David Beetham who acted as
18 Democratic Principles and Judging Free and Fair

Although in his title Goodwin-Gill only mentions free and fair as

criteria, legitimate elections must also be effective. But effective in
doing what? Answering this question requires that five big questions
be addressed.

Scope of Democracy
Most commonly, democracy is equated with the choice of government
through competitive elections. In Joseph Schumpeters (1962 : 269)
words, for example, democracy is that institutional arrangement for
arriving at political decisions in which individuals acquire the power to
decide by means of a competitive struggle for the peoples vote. In
this view, a democracy is neither more nor less than a political system
in which political leaders are chosen in reasonably free competition
among political parties.

In contrast to this, there is a tradition that identifies politics, and

thus democracy as a form of politics, with (in Abbie Hoffmans
words) the way you live your life. In this view, one would not talk
about a democratic government, but rather about a democratic
society, because to restrict attention to the method through which
political decision makers are chosen is totally to strip democracy of
its core meaning.

Choice of government or choice of representatives

The second big question is whether an election is properly
understood as the choice of a government or as a choice of
representatives. One key point, reflected in the grammar of the
preceding sentence, is that a government is singular, whereas there
may be many representatives. On one hand, this means that some
standards that are appropriate for governments might better be
applied to the parliament as a whole rather than to the individual
parties that make it up. On the other hand, because choice of a
government implies choice against some alternative, some forms of
inclusiveness that might be appropriate for parliaments would be
inappropriate for governments.

A second key point is that governments are expected to be able to

act both effectively (when the government makes a decision,
something happens) and coherently (the governments decisions, at
least while one government remains in office, are complementary
rather than contradictory). Representative assemblies, however,
Richard S. Katz 19

often are expected to be expressive rather than effective and to

reflect diversity rather than coherence.

Parties or candidates
Third, are elections contests among parties or among candidates, or
posed more realistically, are voters choosing among parties, each of
which has particular individuals as its standard bearers, or rather are
they choosing among individual candidates, each (or most) of whom
are associated with political parties ? Parliamentary democracy and
proportional representation (PR) electoral systems in modern states
are predicated not only on the idea that parties are cohesive units
but also on the idea that it is those cohesive units for which electors
vote and which therefore have a democratic mandate from the
voters. At the same time, many constitutions emphasize the personal
responsibility of individual members of parliament, either to their
own constituents or to their own consciences.

The mirror image of the question of whether parties or candidates

are the objects of voter choice is the question of responsibility.
Adapting the language of cabinet government, are parties to be
held collectively responsible, with each candidate of a party expected
and expecting to share in the blame for missteps taken by the party
or its leaders even if s/he personally opposed those steps, or is each
candidate to be individually responsible to his/her own constituents
independently of their views of the party as a whole ? As with
collective and individual responsibility of cabinet ministers, the
answer clearly is both, but in what mix?

Self-protection or direction
Fourth, is the objective of democracy to allow the citizens to protect
themselves by reactively punishing rulers of whose policies, or results,
they disapprove or is it to allow the citizens to rule themselves, by
affirmatively deciding the policies to be pursued?

One major strain in democratic theory suggests that the people

should decide what is to be done : that democracy means the will of
the people is to be put into effect. There is, of course, great
disagreement concerning how one can either define or identify the
will of the people, ranging in numerical terms from the unanimity of
Rousseaus volont generale to simple majority rule. Nonetheless, in
this view government is seen as an instrument of the people, taking
positive direction from them.
20 Democratic Principles and Judging Free and Fair

Theories that stress democracy as a means of self-protection begin

with recognition that although effective government is necessary to
the protection of individual rights, it may also be among the most
serious threats to those rights. As Macpherson (1977 : 34) wrote
describing what he called protective democracy, it follows from the
grand governing principle of human nature [that] every government
would be rapacious unless it were made in its own interest not to be
so. While we might argue about the degree to which liberal civil
rights (free speech, free press, free assembly) or a free market
economy are prerequisites for the inauguration or sustainability of
democracy, they clearly are not sufficient conditions, and hence a
liberal free market economy is not a synonym for liberal democracy,
let alone democracy tout court. For those who take the democracy
as self-protection view, one point of adding democracy to the
phrase liberal democracy would be a recognition that ordinary
people need some protection against the natural rapaciousness of
their leaders.

Even if governments are the unproblematic agents of the people,

however, that does not guarantee that they will be benevolent.
Assuming that the will of the people can be expressed by less than
unanimous consent, what is to protect the rights of the minority from
being trampled by the majority?

David Beetham, in response to the round table presentation of this

paper, argued that this dichotomy is overblown : that any democratic
system would prohibit the majority from denying fundamental
political rights to a minority, and recognizing that the existence of an
identifiable minority that is permanently excluded from executive
office would also be unacceptable. In simple, black and white terms,
he is correct, of course. The problem arises when the exploitation of
the minority is not so stark : they are not barred from competing, but
severely handicapped; they are not expropriated, but more heavily
taxed or less adequately served; they are not permanently excluded
from office because of race or gender, but always lose. In these cases,
it is not adequate to point to obvious democratic norms, and yet it is
also not obvious why those who are permanently on the short end of
the will of the people would unproblematically accept the legitimacy
of democracy defined simply as government in accord with the will of
the people.

When this problem is highlighted, the emphasis in the phrase

liberal democracy shifts ; instead of democracy (in particular
Richard S. Katz 21

elections) being a means of enforcing liberalism, liberalism becomes

a rationale for limiting the simple translation of the will of the
people into government action which in terms of the simple will of
the people definition would mean it becomes a rationale for limiting
democracy itself. The classic example here is the Madisonian concern
with majority faction, and the set of institutional prescriptions to
which that concern leads.

Role of citizens
The fifth big question concerns the primary role of citizens in the
context of an election campaign. Are they primarily to be judges
among the contestants or are they to be active participants and
partisans ?

In one view, democracy simply means popular choice among

alternatives that, while perhaps constructed in response to the
perceived or expressed needs, interest, or desires of the citizens, are
formulated by political (generally meaning party) elites. Provided the
range of options from which the choice is made is sufficiently broad,
this view sees electoral judgment, without other substantial
participation, as adequate to effective democracy.

The alternative view sees active involvement in the actual doing of

politics as an essential element of democratic governance. In part, the
contention is that only active involvement by the citizens in the
formulation of party programs, the selection of candidates, etc., will
ensure that the range of choices offered will reflect popular rather than
elite interests and concerns; in part, the contention is that comment and
criticism of party programs from civil society (e.g., interest
organizations, NGOs, etc.) is necessary if the citizens are to be
adequately informed before they make their judgment. More
fundamentally, however, it is informed by the idea that democracy
means self-government and not just government in the publics interest.

Varieties of democracy
In Democracy and Elections (Katz, 1997), I identified a large number
of models of democracy with profoundly different institutional
prescriptions derived from attention to four fundamental
democratic values, with an implicit fifth value, equality (a concept as
complex as democracy itself), assumed without detailed analysis. The
first two of the values considered explicitly were popular sovereignty
(the idea that the will of the people should determine government
22 Democratic Principles and Judging Free and Fair

personnel and policy) and liberalism (the idea that groups need not
just formal rights but practical power to protect themselves from
abuses by their governors).1 In addition to these two values, I
considered participation (the idea that participation in self-
government is an essential prerequisite for full human development),
and community (the idea that democracy both reflects and fosters a
single demos that both has, and perceives itself to have, a
commonality of interest).

The reason why I suggest many varieties of democracy, and indeed

the reason why I introduce an intermediate step between defining
democracy and evaluating institutions, is that the institutional
prescription appropriate to each of the democratic values depends not
only on the value itself, but on the structure of the political space2
and the structure and nature of social and other divisions in the society
in question as well. Combining some categories in the original analysis
(and omitting a few others as of little relevance here), the significant
varieties of democracy are summarized in Table 1. Each should be
regarded as an ideal type, based on stylized assumptions, but, like all
ideal types, they serve as points of reference for assessing the real

Free and fair : Alternative standards

What are the implications of these differing models for democracy for
the standards by which the freeness and fairness of elections should
be judged ? Some standards are universal : ballots must be counted
honestly; voters must be able to cast their ballots free of intimidation
or fear of reprisal; rules must be enforced in a neutral fashion; there
must be a mechanism for the non-arbitrary resolution of disputes.3
While no real election is likely to be perfect with regard to these
standards, there is not likely to be any question either as to their
importance or as to which end of the scale is good. For many of the
other criteria discussed by Goodwin-Gill, however, both the
importance of the criterion and in at least some cases even the
direction (is more or less better?) depends on the conception of
democracy about which one is thinking.

Proportionality of representation
One of the essential considerations that Goodwin-Gill (1994 : 28) cites
is that an election primarily must guarantee representation at the
national level of the countrys political forces, and reproduce in
Richard S. Katz 23

Parliament as faithful an image as possible of their relative strength.

If the primary purpose of an election is to produce a representative
assembly the members of which will be able to form coalitions that
might be justified as being equivalent to the coalitions the people
might form if Athenian style government by an assembly of the
citizens were possible (e.g., legislative popular sovereignty), then
proportionality and the multiparty system that PR is said to facilitate
are quite important. Similarly, if legislative seats are seen as public
goods that must be allocated proportionately among the pillars of
society (e.g., veto-group liberalism), a proportional electoral system is
again important.

If emphasis is instead placed on the other essential consideration

that Goodwin-Gill cites, the designation of a cohesive government
responsible for conducting a national policy, then proportionality is at
best of secondary importance, and from the perspective of several
models of democracy likely to be positively pernicious. On the one
hand, if proportionality facilitates a multiparty system, then it is clearly
undesirable from the perspective of theories (e.g., binary or Downsian
popular sovereignty) that derive their power from the assumption of
two party competition. If multiparty politics limits the heterogeneity
of individual party coalitions, then it would be undesirable from the
perspective of pluralist liberalism as well. On the other hand, from the
perspective of binary or Downsian popular sovereignty, what counts is
that the right party (the one with a majority of the popular votes) be
in effective control of the government, which means that the
tendency of non-proportional electoral systems to exaggerate the
support of the winning party in the translation of votes into
parliamentary seats would be an asset rather than a liability. Likewise,
from the liberal perspective, if the point of elections is to allow the
voters to reward and punish governments, the magnification of vote
swings in their translation into seat swings would simply be increasing
the power of the electoral weapon.

Stability and coherence of government

Governments issuing from parliaments elected by PR and coalition
governments on average have significantly shorter lives than single
party governments issuing from parliaments elected by First Past the
Post (FPTP) systems. Is this a problem?

From the perspective of theories that assume elections are about

choosing governments, the failure of the government chosen to last
Table 1 : Models of democracy

Model Assumptions Institutional Prescription

of Democracy

Binary Popular 1. All issues cluster into two complexes, so that choice is 1. A two-party system, with each party representing one of
Sovereignty between this and that. the two complexes of opinion.
2. The choice of the majority is the will of the people. 2. An electoral system (FPTP) that will support two-party
politics and provide reasonable assurance that the party
with the most votes will control the government.

Downsian Popular 1. All issues can be summarized in terms of a single dimension. 1. A two-party system, with each party free
Sovereignty 2. The first preference of the median voter along the assumed opportunistically to alter its policy position along the issue
single dimension of politics (the Condorcet choice) is the will dimension.
of the people. 2. An electoral system (FPTP) that will support two-party
politics and provide reasonable assurance that the party
with the most votes will control the government.

Legislative 1. The issue space is multidimensional. 1. A multiparty system, with each party representing a
Popular 2. While there is no Condorcet choice, the will of the people different combination of policy positions.
Sovereignty can be approximated through the formation of a parliamentary 2. An electoral system (PR) that will accurately reproduce
coalition representing a majority. in parliament the distribution of opinions found
in the electorate.

Majoritarian 1. Society is basically homogeneous, with no politically relevant 1. A two-party system, with each party representing an
Liberalism and stable subgroups. alternative team of leaders prepared to assume the
2. Majority rule is adequate to protect against elite tyranny, direction of government.
and majority tyranny is not a concern. 2. An electoral system (FPTP) that will support two-party
politics, and provide citizens with an effective opportunity
both to dismiss the party an power and to dismiss individual
politicians who are perceived to be abusing their positions.
Democratic Principles and Judging Free and Fair
Pluralist 1. Society is made up of stable, but cross-cutting, groups. 1. A party system in which each party (or at least one party
Liberalism 2. The problem of majority tyranny can be mitigated through in any potential coalition of parties) is to a significant
multiple veto points, privileging different combinations of degree dependent on the support of every significant
political resources. That at least one of these veto-points is group in society.
majoritarian is adequate protection against elite tyranny. 2. An electoral system (e.g., FPTP) that encourages parties
Richard S. Katz

to build broad coalitions rather than to mobilize a narrow

3. An institutional system that makes it difficult for a
majority to gain control of all of the institutions
of government (e.g.., separation of powers, federalism, etc.)

Veto-Group 1. Society is divided (pillarized) into stable and non-overlapping 1. A party system in which each segment of society is
Liberalism groups. represented by at least one party, but preferably by exactly
2. Majority tyranny can be avoided only by giving each politically one party.
relevant group unilateral veto power. 2. An electoral or other device to make the leadership
of each segment responsible to the members of that
3. An institutional system (e.g., government by grand
coalition) that will afford each segment of society veto
power over decisions that it perceives to be excessively

Participationist 1. Full human development requires taking active responsibility 1. Structures that afford maximum opportunities for direct
and in government. citizen involvement in decision-making.
Communitarian 2. Active involvement in the collective enterprise of governing 2. Institutional arrangements that move decisions to the
Democracy will foster the development of community. most local (i.e., smallest) units possible.
3. Structures that maximise political talk.
26 Democratic Principles and Judging Free and Fair

its full term is a clear failure. But this clearly is reflective of the logic
only of two-party popular sovereignty (or of presidentialism). With
legislative popular sovereignty, the purpose of an election is to choose
representatives who will choose the government; and because those
representatives, unlike the voters, are in more or less continuous
session, they can change the coalition in power without challenging
the underlying democratic legitimacy of the system.

If elections are to give the people through their representatives

the ability to limit their government, stability may be evidence of
failure rather than success. While the ideal of many 19th century
liberals that annual elections would be desirable to allow the voters
to [divest] of their power all unfit representatives before they have
had time to produce any lasting mischief, (Bentham, 1962 : 561) has
been generally rejected on practical grounds, the idea that a
government should never be too securely in office is the essence of
majoritarian liberal democracy, and only slightly less central to
pluralist liberalism.

Universality of suffrage and voter turnout

Clearly the first thing to be said about universal suffrage is that no one
either believes in it or practices it. All countries have minimum age
requirements. Many require citizenship or at least a lengthy period of
residence. Some citizens may be disqualified on account of mental
incompetence or criminal convictions. In Kuwait and Saudi Arabia,
women do not have the right to vote.*

The Kuwaiti example is illustrative of an important point. The

prototypical democracy of Athens was democratic only with respect
to a quite narrowly defined segment of the population. Exclusions of
women, members of various ethnic groups or religions, or people
without extremely long residence or property in the area have a long
history in the established democracies. That we now regard these as
illegitimate does not alter the fact that both their imposition and
their elimination reflect cultural biases concerning the proper nature
of a political community. That the international community may
regard itself as justified in imposing these modern values on cultures
that we regard as less advanced does not make them less the product
of culture.4

* Editors note : In May 2005, the Kuwaiti parliament approved constitutional

amendments to give women full political rights.
Richard S. Katz 27

Beyond this, there may be other reasons why less than universal
suffrage may be democratically acceptable. If elections are part of a
process of rational decision-making rather than mere expressions of
opinion, for example, one might legitimately limit the participation of
the incompetent. At the trivial (except in numerical terms) level, this
is the justification for denying the vote to children. The problem with
more substantive competency requirements (or toleration of
competency related conditions that impose a substantial burden on
some individuals) when applied to adults, however, is that the
incompetent may constitute an interest of their own, or be a
significant part of some interest. Particularly from the liberal
perspective, indeed, the incompetent may represent an interest that
is particularly in need of protection.

Low turnout often is interpreted as either indicative or causative of

low legitimacy of the outcome. Nonetheless, some efforts to increase
turnout may be a cure that is worse than the disease. On the one hand,
lessening the barriers to voting (easing registration procedures,
allowing postal voting, etc.) may also facilitate fraud. On the other
hand, they may exaggerate rather than mitigate biases in the
composition of the active electorate. (On both these problems, see
Katz, 2004.)

Is high turnout necessarily to be regarded as good? From the

perspective of participationist or communitarian models, true
citizenship is an achieved status : Citizens are neighbors bound together
neither by blood nor by contract but by their common concerns and
common participation in the search for common solutions to common
conflicts. (Barber, 1984 : 219) While participation in an election may
help to integrate individuals into a political community, the votes of
excessive numbers of individuals who are not committed to the
community may instead undermine the democratic legitimacy of the
election in the eyes of those who do feel such a commitment.

Constituency delimitation and equality of votes

One of the great mantras of the late 20th century became one-person,
one-vote, one-value. The first part (one-person, one-vote) is
unproblematic in theory, although sometimes more difficult to realize
in practice. The second part (one-vote, one-value), however, is not at
all straight-forward unless the electoral system is PR with large
districts. The greatest problems, however, arise in the context of
single-member districts.
28 Democratic Principles and Judging Free and Fair

It is often assumed that having each district with as near to the same
population (whether of residents, or citizens, or voters) as possible is
both a necessary and a sufficient condition for equality of influence.
This equality, however, may be illusory. This is illustrated in Figure 1,
which shows a hypothetical territory with 15,000 voters divided into
15 perfectly square and exactly equipopulous districts. If the numbers
in each square are the votes received by one of two parties competing,
then with exactly 7500 votes to its opponents 7500, that party will win
seven of the 15 seats. If it loses 60 votes in each district, it will win only
two seats with a total of 6600 votes, while if it gains 60 votes in each
district it will win ten seats with 8400 votes overall. But, by subtraction,
this means that while 6600 votes win this party only two seats, they
win its opponent five, and while 8400 votes win it ten seats, they win
its opponent 12.

Figure 1 :
15 Equal-population districts with the votes
for one of two parties.

900 900 450

550 550 550

550 550 350

450 350 350

450 350 200

This example was deliberately constructed to preclude appeal to the

claim of gerrymandering, and in a sense it is immune to that charge.
At the same time, however, it must be recognized that this immunity
stems entirely from the privileging of simple shape as an essentially
aesthetic judgment. Strange shapes (perhaps following a coastline or
a river valley or an ancient tribal boundary) might produce districts
that are more fair, either in the sense of including a more
homogeneous constituency or in the sense of being less biased in favor
of one party or another. For the purposes of this paper, the key point
is the frequent use of the word or in the preceding sentence; that is,
there are many different ways in which fair districting can be
understood, and the imposition of presumably neutral standards like
compactness does not obviate the problem. And, for precisely this
Richard S. Katz 29

reason, strict adherence to a standard of equal population may simply

make partisan gerrymandering easier.

Patronage and the buying of votes

There is general agreement that the buying of votes, whether for cash,
promises of favours, or threats of reprisals is unacceptable, but there
is little systematic consideration of the broader implications of this
position. The basic objection, of course, is that concentrated economic
power should not be converted into electoral power, which should
instead be based on numbers. Further reflection, however, raises two

The first problem is whether this form of retail corruption is

actually worse than what might be described as wholesale
corruption : corporate threats to relocate a major source of
employment out of a community or party promises to provide benefits
for groups that support them. Of course, it can be argued that the
wholesale forms of corruption involve the exchange of votes for public
goods whereas the reward for the retail vote-seller is purely private,
but whether this distinction can be maintained in cases of policies that
benefit only a narrowly defined set of voters is questionable. But if
the distinction cannot be maintained, the result is to give those in
control of government or corporate wealth an avenue of influence
denied to those without those assets.

The second problem concerns the purpose of voting. While the

communitarian model of democracy assumes that there is a single
common interest that is separate from and above the simple sum of
the citizens private interests, liberal models are based on the
fundamental primacy of private interests, and while popular
sovereignty theories talk about the popular will in the singular, they
define it as the aggregation of individual wills.5 But if it is assumed not
only that citizens will vote so as to advance their private interests, but
also that they should vote on this basis, why should they be barred
from acting on the view that their individual votes will advance their
private interest more effectively if regarded as private goods to be sold
on the private market ? Moreover, given the experience of machine
politicians, particularly in the United States but not only there, in
integrating new citizens into the political community through
Christmas turkeys, jobs, or outright bribes (see Riordon, 1963), one
might question the assumption that patronage and vote buying
undermine democracy.
30 Democratic Principles and Judging Free and Fair

Regulation of parties
Even when they do not accept the full equation of democracy with
party government (Katz, 1987), political scientists and political
practitioners alike generally accept E. E. Schattschneiders (1942 : 1)
dictum that political parties created democracy and that modern
democracy is unthinkable save in terms of the parties. This means, on
the one hand, that free and fair elections require the presence of
political parties a requirement that may be interpreted as imposing
an affirmative mandate to assure that there are parties and, on the
other hand, that some restrictions on parties may be justified. While
there are many aspects of this problem, I will mention only two :
registration of parties and regulation of their internal organization
(internal democracy). A third aspect, political finance, is raised later.

Registration of parties : Laws setting conditions under which

political parties will be recognized and will be given privileges like
assured ballot access, public subsidies, or representation in
parliamentary committees all can be justified as necessary to make
elections manageable. An excessive number of choices is likely to so
fragment the vote that the result will not be meaningful; similarly,
excessive fragmentation in parliament is likely to make the
maintenance of stable majorities impossible.

An official party registry facilitates timely and definitive resolution

of conflicts concerning the use of a party name and the right of
individuals to identify themselves as candidates of the party. At the
same time, however, one effect of such regulations is almost inevitably
to stack the deck in favor of the existing parties : requiring new would-
be parties to undertake extensive organizing efforts at a time when
politics is likely to be less salient; freeing existing parties from the need
to expend resources to collect petition signatures or otherwise to
demonstrate support in advance of the election in order to secure a
place on the ballot; giving established parties resources and a position
in parliament that are denied to independents or newcomers. The
problem is to strike the proper balance between unfettered entry into
the electoral arena and meaningfully structured competition.

Where that balance lies depends on the conception of democracy

with which one begins. In particular, while the model of legislative
popular sovereignty would incline the balance toward easy
qualification of parties, the models of binary or Downsian democracy
would incline it very much the other way. Similarly, the pluralist or
Richard S. Katz 31

majoritarian models of liberal democracy suggest raising the bar for

the entry of new parties in the interest of encouraging diverse
interests to coalesce within broader umbrella parties while veto group
liberalism would be more lenient, at least for parties claiming to
represent interests that do not already have their own party.

Internal party democracy : Does democracy require (or indeed is

democracy even furthered) by requiring that parties be democratic
with regard to their internal (policy formulation, candidate selection,
etc.) procedures ? Although it is appealing to assume that the answer
to this question must be yes, and although there have been moves in
this direction within the regulatory systems of some democracies, the
answer in fact is far from clear.

If democracy implies active citizen participation and parties provide

one of the venues for that participation, then internal democracy is
important. Similarly, if democratic elections are about choices of
representatives and among candidates, internal democracy may allow
groups of citizens to determine the individuals who will represent
them and the policy preferences that they will represent. On the other
hand, there is a significant body of democratic theory that takes the
opposite positionarguing, in Giovanni Sartoris (1965 : 124) words,
that democracy on a large scale is not the sum of many little
democracies. Anthony Downs (1957 : 25), for example, argues against
an inclusive definition of party, or internal party democracy, because
whatever policies emerge are likely to form a hodgepodge of
compromises. In this case, even if the representational function of
elections might be enhanced, the clarity of choice offered to the voters
would be sacrificed. And indeed in the Downsian model, the aim of
democracy is furthered by competition between parties that are
motivated solely by the private interest of their leaders, who generate
popular policies simply as a means of winning votes. It is the personal
disinterest in policy of the party leaders that leads them to converge
toward the first preference of the median voterwhich is the policy
package that has the best claim to the title will of the people. But
since individual party members would not share in the personal
rewards of office, the presumption is that they are motivated by
policy. Thus, even if internal democracy did not produce the
hodgepodge of compromise that Downs feared, it would produce
proposals near to the median preference of each partys members, and
therefore not at the median of the electorate as a whole. In other
words, democracy within the parties would prevent the democratic
32 Democratic Principles and Judging Free and Fair

outcome that is supposed to be furthered by competition between the


While the Downsian ideal of convergence requires a two-party

system, the requirement in multiparty systems is that the leaders of the
parties be willing and able to compromise with one another to form
coalitions. Particularly because internal party democracy is likely to
empower activists, who tend to be stronger if not necessarily more
extreme in their preferences, rather than either base party members
or party supporters in general, it is likely to make compromise less
rather than more easyand indeed in Lijpharts original work on what
he called in English consociational democracy (Lijphart, 1968) (kartel
democratie in Dutch), elite autonomy from their followers was
advanced as one of the secrets to maintaining liberal democracy in a
deeply divided society. From this perspective as well, the claim that
internal party democracy will further democracy at the system level is
at least suspect.

Are free and fair compatible ?

The previous section was concerned with the ways in which the
standards of free and fair are dependent on the understanding of
democracy adopted. In this section, I address a related questionthe
degree to which freeness (understood to mean the lack of restrictions
on those contesting elections or otherwise participating in electoral
politics) and fairness (understood as the metaphorical level playing
field) are compatible.

Limitations of party campaign practices

The idea that certain forms of campaign practice must be banned
because they risk offending groups or inflaming passions, or that
certain ideologies or individuals must be banned as being anti-
democratic may seem appealing at first glance but is problematic on
closer inspection. In Rousseaus (1947 : 91) democratic theory, for
example, the first question that should always be proposed, and never
on any account omitted was whether the present form of
government should be continuedin other words, democracy
requires that the continuation of democracy always be regarded as an
open question. To say, for example, that the people of a liberal
democracy may not choose to be governed instead by a theocracy
delegitimizes democracy in its own terms; consent is meaningless if
there is no way in which lack of consent can be expressed.
Richard S. Katz 33

Enforcement of good taste, civility, or truthfulness in campaign

activity all necessarily involve not just the protection but also the
constriction of democracy. While incitement to genocide, for example,
clearly cannot be protected,6 it also must be recognized that the
definitions of good taste, civility, or truth often are politically
contentious, so that to impose definitions is to bias the discussion.
Again, a balance is required rather than simply an attempt to impose
civility, respect for authority, or truthfulness through legislation.

Funding of politics
The conflict between freedom of speech and other democratic values
like community also arises with regard to regulation of party finance.
First, there is a conflict between the ideals of equality and majority
rule (which might suggest, among other things, strong limits on the
size of allowable contributions from individual citizens and perhaps a
total ban on political spending by anyone/anything except
individuals), on the one hand, and the ideals of freedom of speech and
the liberal pluralist notion that various groups, endowed with
different mixes of resources (numbers for some groups; wealth for
others; access to or ownership of strategic communications media for
still others) should be allowed to protect and advance their interests
as best they can.

Second, there is a conflict between the ideal of politics as a labor

intensive activity in which large numbers of citizens take part on a
regular basis (again suggesting strict limits on party finance) and the
reality that electoral politics has become a capital and expertise
intensive activity, in which citizens often can take part more effectively
by pooling their financial resources so as to hire experts. In particular,
this calls into question the idea that direct spending should be
protected as an expression of the right of free speech whereas
contributions can be regulated (e.g., Buckley v. Valeo 424 U.S. 1), as
giving a right to those who are rich enough to take effective action on
their own that is denied to those who must pool their resources in
order to be effective.

Third, there is a conflict between the idea that the effects of

regulations concerning party fund-raising, and especially concerning the
provision of public resources or privileges, ought to be in rough
proportion to current popular support, and recognition that those who
want to challenge the status quo often depend on a very few large
donors or access to public resources in order to build public support in
34 Democratic Principles and Judging Free and Fair

the first place. Moreover, where access to public resources is afforded to

parties simply by virtue of their qualification for the ballot, there is a risk
of candidacies motivated by desire for the resources rather than desire
to influence policy. If access is not equal however, this generally means
giving the established or larger parties access to public resources in
greater amounts and on more favorable terms than is given to new
parties. In other words, apparent equity based on demonstrated
support can be extremely conservative in its effect.

Fourth, there is the conflict between the fear that enforcement

powers will be used by those in power to repress, harass, or hamper
their opponents and the recognition that regulations without an
effective enforcement mechanism are unlikely to be effective and may
bring the entire notion of fairness through regulation into disrepute.
While this problem may be mitigated through the use of non-political
election management agencies, it still must be recognized that non-
political often is in the eyes of the beholder, and indeed that the very
notion of non-political administration has historically reflected a
profoundly bourgeois conservative bias.

Access to media
In large societies, freedom of speech without access to the media of
mass communications is worth very little. Goodwin-Gill (1994 : 24, 67),
for example, cites the final document of the CSCE 1990 Copenhagen
Conference that no legal or administrative obstacle [should stand] in
the way of unimpeded access to the media on a non-discriminatory
basis for all political groupings and individuals wishing to participate
in the electoral process. The problems are, first, that like the
universal in universal suffrage, no one means unimpeded to be
taken literally, and second, that as with public financial support the
standard of non-discriminatory is fraught with ambiguity. These
problems are manifested in a number of more specific questions.

The first concerns the allocation of time, particularly on state owned

media. This reproduces the problems cited above with regard to
financial subventions : Who is eligible ? Is the allocation made equally
to all qualified parties (with the danger of inspiring frivolous
candidacies or spurious multiplication of parties) or proportionately
based on strength (with the bias against new-comers that this
implies) ? Further, who will pay for the production costs (public
payment implying a restriction on the right of the well-endowed to
produce the most effective message, with private payment
Richard S. Katz 35

advantaging those who can afford to pay for more professional

presentation) and (in part in response to the first problem) should
form or content restrictions be imposed (e.g., allowing only a studio-
based talking head) ?

The second concerns the reporting of news. Reporting always

involves choices : which stories to cover; how to frame them; etc. News
reporting is important in a free and fair election because it gives the
voters the information they need to make informed choices. But this
is just another way of saying that news is expected to affect voters
choices. The problem is to reconcile freedom of the press with fairness,
given that fairness and balance are inherently subjective. Can there be
an objective standard of fairness in reporting ?

Because this problem appears to have particular purchase with

regard to public media, where the danger of journalistic bias being
imposed as a political choice by and in favor of those currently in
power is apparent, there has been a tendency to try to impose more
objective standards in these cases. Frequently, these take the form of
stopwatch-based equality of coverage. Although not specifically
related to elections, the length to which this kind of requirement can
be taken is illustrated by Italian televisions coverage of government
crises in the 1960s :

No politicians voice was ever heard, nor were his words quoted
directly. Instead, party leaders appearing for consultation with the
president of the republic were each shown from the same camera
angle and for the same amount of time; upon his exit, each was
shown in turn speaking at a microphone, but without sound. While
this went on, a disembodied a carefully written and
approved summary of what he had said... Italian politicians on
television [resembled] fish in an aquarium their mouths move, but
no sound emerges (Porter, 1977: 261-2).

The more one tries to impose the appearance of fairness, the more
one impinges on freedom, or effectiveness, or both.

On the other hand, although it may be more legitimate to impose

on the freedom of journalists who are, at least indirectly, in the
employ of the government, the problem of unfairness facilitated by
insistence on journalistic freedom may be far greater with regard to
privately owned media. A current example would be the plans of the
Sinclair Broadcasting Group in the United States in 2004 to air a
36 Democratic Principles and Judging Free and Fair

documentary film of questionable veracity that was highly critical of

John Kerry in the days immediately before the presidential election,
while labeling it as news. On a grander scale, one could point to
Italian Prime Minister Silvio Berlusconis ownership of much of the
Italian private television industry. When control of the media is both
concentrated and in politically interested hands, only the self restraint
of the owners can assure even an approximation of fairness in the
absence of regulation that must be recognized as an infringement of
freedom of speech. Ultimately, one can have one or the other, but not

If the analysis above is of any value, it is to highlight three facts that
must be central to any attempt to assess the freeness and fairness of
elections. The first is that if all one wants to do is identify electoral
events or practices that are grossly inadequate, the task is easy. If
ballot boxes are stuffed, or voters are credibly threatened with death,
or opposition candidates are barred from campaigning, or
government coffers are opened to fund the campaign of only one
party, a precise legal code is not required for the legitimacy of the
outcome to be rejected.

The second conclusion is that once one allows the possibility of

degrees of freeness and fairness, the task becomes very complicated.
Most obviously, the problems of establishing thresholds of
acceptability and measures that can be used to assess the placement
of actual electoral events relative to those thresholds are far from
trivial, and are only complicated by recognition that the reports of
election observers may have a significant bearing on the likelihood of
post-election violence and on the likelihood that a transition to stable
democracy will be continued. Is a flawed election better or worse than
no election at all ? Beyond these problems, however, what this paper
has shown is that even the standards by which acceptability might be
judged depend on the understanding of democracy that one employs.

The third conclusion is closely related, and equally troubling. It is

that while democratic legitimacy requires that elections be both free
and fair, there are a wide range of conditions and circumstances
under which freeness and fairness are incompatible. As a result, a
balance must be struck between the two desiderata. While the
particular balance that is appropriate will depend to a certain extent
on local conditions (limitations of freedom in the name of limiting
Richard S. Katz 37

concentrated economic power or media control are more justified

where there is more concentration in the first place), they cannot
determine the one most appropriate tradeoff, because there is no
one most appropriate tradeoff.

Few, if any, decisions of these types are politically neutral. Some

interests or parties are advantaged and others disadvantagedat
least relative to other decisions that might equally plausibly have been
proposed. Since there can be no unproblematic standard by which
freeness or fairness can be assessed, this means that to propose
reforms, even in the name of fairness or neutrality, is likely to be
perceived by some participants as taking sides in the substance of
political competition. Certainly there are some practices that are
unacceptable by any reasonable standard of democratic propriety, and
those should be opposed. Beyond that, however, those who advocate
standards for the evaluation of electoral practices would be well
advised to recognize that such proposals are not above politics, but are
of politics.

About the author: Richard S. Katz is Professor of Political Science at

The Johns Hopkins University in Baltimore. His books include: A Theory
of Parties and Electoral Systems, Democracy and Elections, Party
Organizations : A Data Handbook, and How Parties Organize (the
latter two co-edited with Peter Mair). Katz has served as chair of the
Representation and Electoral Systems section of the American Political
Science Association and as convenor of the Standing Group on Political
Parties of the European Consortium for Political Research. He is
currently on the editorial boards of Party Politics, Irish Political Studies,
and European Union Politics, and is co-editor of the Political Data
Yearbook for the European Journal of Political Research.
38 Democratic Principles and Judging Free and Fair

References :
Barber, Benjamin (1984) Strong Democracy: Participatory Politics for a New
Age, Berkeley : University of California Press.
Bentham, Jeremy (1962) Radical Reform Bill with extracts from the reasons, in
John Bowring (ed) The Works of Jeremy Bentham, New York : Russell & Russell.
Carver, Richard (2000) Broadcasting and political transition : Rwanda and
beyond, in Richard Fardon and Graham Furniss (eds) African Broadcast
Cultures : Radio in Transition, Oxford : James Currey, pp. 188-97.
Downs, Anthony (1957) An Economic Theory of Democracy, New York : Harper
and Row.
Elklit, Jrgen and Andrew Reynolds (2005) A framework for the systematic
study of election quality, Democratization, forthcoming.
Goodwin-Gill, Guy S. (1994) Free and Fair Elections : International Law and
Practice, Geneva : Inter-Parliamentary Union.
Katz, Richard S. (1987) Party government and its alternatives, in Richard S.
Katz (ed), Party Governments : European and American Experiences, Berlin : de
Gruyter, pp. 1-26.
Katz, Richard S. (1997) Democracy and Elections, New York : Oxford University
Katz, Richard S. (2004) Increasing turnout: Might the cure be worse than the
disease ?, paper presented at the Joint Sessions of Workshops of the European
Consortium for Political Research, Uppsala.
Lijphart, Arend (1968) The Politics of Accommodation, Berkeley : University of
California Press.
Lijphart, Arend (1999) Patterns of Democracy, New Haven : Yale University
Mackenzie, W.J.M. (1958) Free Elections : An Elementary Textbook, London :
George Allen & Unwin.
Macpherson, C.B. (1977) The Life and Times of Liberal Democracy, Oxford :
Oxford University Press.
Porter, William E. (1977) The mass media in the Italian elections of 1976, in
Howard R. Penniman (ed) Italy at the Polls : The Parliamentary Elections of 1976,
Washington : AEI, pp. 259-86.
Riordon, William L. (1963) Plunkitt of Tammany Hall : A Series of Very Plain
Talks on Very Practical Politics, New York : E. P. Dutton.
Rousseau, Jean Jacques (1947) The Social Contract, New York : Hafner. Sartori,
Giovanni (1965) Democratic Theory, New York : Praeger. Schattschneider, E.E.
(1942) Party Government, New York : Holt, Rinehart and Winston.
Schumpeter, Joseph (1962) Capitalism, Socialism and Democracy, New York :
Harper and Row.
Richard S. Katz 39

Notes :
Although there are some apparent similarities between what I have identified
as popular sovereignty models of democracy and Lijpharts (1999) model of
majoritarian democracy (in particular in the formers identification of the
popular will with the will of the majority itself identified in my popular
sovereignty models as the Condorcet choice, assuming that there is one),
there are also substantial differences. Likewise, although there are similarities
between my liberal models of democracy and Lijpharts consensual model
(indeed, I regard Lijpharts (1968) earlier model of consociational democracy
to be one of the liberal models), these two categories are far from identical.
I use the phrase political space as a short-hand to refer to the number of
issues in play, their nature (i.e., whether the options may best be
characterized as dichotomous, polychotomous, or continuous), their
dimensionality, and the degree to which attitudes concerning different issues
may be expected to cluster.
While not including all of the criteria that they consider, this list is adapted
from Elklit and Reynolds (2005 : Table 1).
This example was chosen deliberately to be provocative. In substantive terms,
I agree with the point that David Beetham raised in reply, that the exclusion
of women simply is wrong. On the other hand, like all questions of right and
wrong, the virtue of including women is not amenable to an empirical
answer. As with the teaching of evolution as the only scientific theory of the
origin of species (with theory here implying no more doubt than it does with
regard to the theory of gravity), overwhelming consensus on the part of
experts may have little probative value with those who, for cultural reasons,
do not accept the basic premise. But it is precisely in those circumstances in
which there is disagreement about culturally specific values that the problem
of defining free and fair is most important.
An exception is what I described (Katz, 1997) as collectivist popular
sovereignty theories.
For example, Georges Ruggiu pleaded guilty to incitement before the
International Criminal Tribunal for Rwanda. See Carver (2000).
Judging Elections and Election
Management Quality by Process*
Jrgen Elklit and Andrew Reynolds
At the heart of democratization attempts lie competitive elections,
often held during times of societal stress and under imperfect logistical
conditions characterized for example by administrative unreadiness.
The claim here, as argued in greater length elsewhere (Elklit and
Reynolds, 2002), is
that the relationship between the institutionalization of electoral
politicsin particular the administration of the electoral process
and the emergence of democracy in the developing world and
elsewhere is a much under-studied part within the fields of
governance and democracy studies; and
that a focused analysis and assessment of the quality of the various
elements in the electoral process will provide those interested in
electoral quality in general and election management quality in
particular with a useful instrument, which has so far been lacking in
the tool kit of electoral assessment.

It is difficult not to agree with those who claim that assessments of

elections must focus on the entire electoral process, as the component
parts of that process all have at least some bearing on the entire
outcome of the election, i.e., not only the results as such, but also
whether or not the entire electoral process is seen as legitimate and
binding by voters and other political players.

The quality of an election can thus be conceptualized as the

degree to which political actors at all levels and from different
political strands see the electoral process as legitimate and binding.
However, the operationalization of this theoretical concept is
not easy, even though the IPU publication of the landmark study
Free and Fair Elections : International Law and Practice by Guy
Goodwin-Gill (1994) has contributed substantially to the understanding
and structuring of the field.

* This paper was presented by Jrgen Elklit during Free and Fair Elections, Ten Years
On: An International Round Table on Election Standards, held in November, 2004.
The authors are grateful for the helpful comments received from our two
respondents at that time, Horacio Boneo and Ron Gould. We also appreciate the
comments offered at an Australian Electoral Commission workshop in June 2004 and
by an anonymous referee and Peter Burnell, editor of Democratization. Substantial
parts of this article have appeared in a previous article by Elklit and Reynolds
published in Democratization in 2005.
54 Judging Elections and Election Management Quality by Process

One important result of more than a decade of global diffusion of

multiparty politics and support for the holding of democratic elections
is that it has eventually been realized that the quality of election
administration has a direct and important impact on the way in which
elections and their outcomes are regarded, not only by international
observers and monitors and their organizations, but alsoand more
importantlyby domestic political actors such as voters, parties and
party leaders, media, and domestic observers. However, these groups do
not necessarily see things the same way; indeed, their differential
perceptions are useful as they allow us to gauge, at least partly, the
reasons why different groups arrive at variant assessments of the same
electoral process.

Focus here is on how institutional factors and institutional choices,

and the ensuing administrative and political behaviour, contribute to
the quality of the entire electoral process and therefore also to the
transition and the eventual consolidation of democracies. This inevitably
leaves aside a whole array of other issues, which also influence the way
in which elections are perceived and contribute to the way in which
democracy gradually becomes the only game in town, if that happens.

Elections play a crucial role in that development because they are a

necessary condition for having some kind of democratic regime. That is
why we focus specifically on the way in which elections are conducted
and formulate our questions so that they will enable us to gauge the
effectiveness and positive contribution of institutional choices related to
electoral process management and the impact of the various stages of
the implementation process. Following from this, our main claim is :
that individual experiences in a number of fields related to the
electoral process have a direct bearing on how the sense of political
efficacy develops in individual citizens; and
that this is an important factor behind the development of
democratic legitimacy as well as a principled commitment to
democracy, i.e., progression towards democratic consolidation (even
during the transition phase).

Assessing and observing elections

While an assessment of the quality of national elections (i.e., freeness,
fairness, and administrative efficacy) requires a fitting methodology, a
clear void exists in the academic and policy literature that focuses on this
problem. The field is ripe for the development of a systematic method
Jrgen Elklit and Andrew Reynolds 55

that pursues this goal, one that can be applied in the context of both
developed and developing world cases, whether during first or
subsequent electoral events.

Below, we go beyond previous work in the field of elections and

election administration assessment by suggesting a more operational
and empirically-oriented approach. We introduce and describe the
elements and the scoring methodology of our assessment framework,
explaining its rationale and offering the model to election practitioners
as well as to election observers and academics interested in these issues.
To illustrate its workings we have also scored six multiparty elections :
two in established democracies (Australia and Denmark 2001) and four
in fledgling democracies (South Africa 1994 and 2004, East Timor 2001,
and Zimbabwe 2002), so two of our cases are in the same country over
time. This degree of specificity is indicative of the route this field of
study must take if it is to contribute substantially to the empirically
based assessment and analysis of elections and election management.
We want to stress that the intention is not only to develop an
instrument for academic analysis. Our claim is that the same instrument
will allow useful specific comparisons of developments from one
election to the next in the same country and also, although requiring
more care, between countries.

We strongly hope that the presentation of this framework and this

approach will engender more debate and analysis, which will in return
facilitate the models use as a practical tool for both non-governmental
and governmental election observation missions and as a research tool
to better understand the issues that determine electoral quality and
legitimacy. We are mindful that the model will also allow election
managers and administrators to assess the quality of their own work on
a comparative basis.

The work has its seeds in our own previous work (Elklit and Reynolds,
2001; 2002) which was further developed during a workshop organized
by the Australian Electoral Commission in Canberra in June 2004. It is
only quite recently that students of democratization have begun to
acknowledge that governance issues must also encompass issues related
to the conduct of elections in both consolidated and emerging
democracies. Those analyses generally agree on conceptualizing
electoral governance as a set of closely linked activities, sometimes
categorized under the older headings of rule making, rule application,
and rule adjudication (Mozaffar and Schedler, 2002 ; Kjr, 2004 :
56 Judging Elections and Election Management Quality by Process

157-71 ; Elklit and Reynolds, 2002; Lpez-Pintor, 2000; Elklit, 1999;

Norris, 2004; Mozafar, 2002).

Mozaffar and Schedler (2002 : 5) claim that because elections in

established democracies tend to be routine events, usually producing
results within a narrow but fully-acceptable margin of error, systematic
analysis of electoral governance has not attracted much scholarly
interest. There will always be some margin of error as it is difficult to
envisage any large-scale operation such as a national election not being
occasionally infected by defective ballots, incomplete voter registers,
inaccuracies in counting and impersonation, etc. Humans make
mistakes, but if these errors are random and do not accumulate to
influence the outcome of the election, electoral credibility survives,
which is exactly why these credible routines themselves tend to obscure
how important electoral governance is. Electoral governance issues only
attract critical attention when something goes seriously wrong, or when
an electoral issue is taken up as part of a more general election-related
controversy (Mozafar and Schedler, 2002 : 6; Schedler, 2002).

It seems self-evident that good electoral governance contributes to

the democratic legitimacy of competitive elections, but it is not easy to
determine exactly how electoral governance in itself affects political
democratization and the development of democratic legitimacy. The
claim that electoral quality has a bearing on political legitimacy matters
is intuitive, but it is more difficult to offer convincing theoretical
arguments and empirical evidence. Indeed, previous attempts at
conceptualizing electoral manipulation have aimed at measuring
violations of democratic norms during the electoral process, and thus
have focused on electoral manipulation as an indicator, not a cause, of
illegitimacy (Elklit and Svensson, 1997/2002; Goodwin-Gill, 1994).

Obviously, attempts to hypothesize about the causes of political

legitimacy (or illegitimacy) require the inclusion of a number of
variables (including different forms of electoral manipulation), which
are difficult to operationalize and to measure empirically in such a way
that clear conclusions can be established. In spite of these difficulties,
one might still be able to use empirical observations as indicators of high
or low levels of political legitimacy. Interestingly enough, a number of
studies on elections and electoral issues in Latin America do touch on
these questions, even though often more indirectly. A recent example
analyses the differential turnout rates across Guatemala and tries to
identify reasons for the biases (Lehoucq and Wall, 2004).
Jrgen Elklit and Andrew Reynolds 57

We emphasize the intimate linkage between the entire electoral

process and democratization, but one can go beyond an analysis which
is focused on that context. Post-conflict elections must also be judged
on their contribution to bringing conflict to closure. Post-conflict elections
in cases such as Bosnia and Herzegovina, Mozambique, Angola,
Afghanistan and Iraq are obvious examples. However, the framework
presented below does not include a special category of post-conflict
war-torn societies alongside fledgling and established democracies
because we want, in this context, to focus on our main objective, which
is the presentation and discussion of the basic framework. Still, there is
reason to believe that the addition of a new category, akin to Lyonss
(2004) analysis of the role of election administration in some of the more
complicated post-conflict elections, could prove very valuable. At the IPU
round table, Ron Gould also argued that the use of only two categories
would most probably appear to be insufficient, and that one or more
in-between categories would need to be established.

Why the need for such a measurement tool ?

The lack of a robust and comprehensive framework of analysis has left
a space which has so far been filled by two equally-unsatisfactory
outcomes : Either election observers make judgments on the basis of
impressionistic and incomplete evidence focused on the conduct of the
vote and count on election day, or observation missions (often from
abroad and with their own governments lead) call an election in a
politicized way, detached from any relation to the truth of the process
itself. A case in point here is the pronouncement by the official South
African observation mission that the 2002 Zimbabwe presidential
elections were free and fair.

The greatest failing of election assessment to date has been the

tendency to see election quality in bimodal terms. The election is either
good or it is bad; or, when a fudge is required, it is substantially free
and fair. We claim that the quality of elections across cases and across
time can be seen as existing on a continuum, even though it makes
sense to approach this fuzzy concept as one of multidimensionality
(Elklit and Svensson, 1997/2002). Election management within a country
can be strong in some areas and weak in others. The playing field that
regulates the campaign can vary subtly in both de jure and de facto
ways, and elections clearly can improve as well as decline quality-wise
on a number of dimensions over time. One needs to look at both process
and outcome to gauge the full picture of election quality.
58 Judging Elections and Election Management Quality by Process

It would be too simplistic to apply a rigid methodology which gave,

for example, Sweden a 92, the United States a 78, and the Congo a 59
and see such scores as perfect indicators of the nuances of all that goes
into allowing for good electoral processes. At the same time, however,
there are clear clusters of electoral elements which we can assess, and
one is able to offer an overall assessment of election quality which is
more rooted in the evidence than previous impressionistic offerings.

There is also the important question of whether an elections failings

are great enough to affect the final result. Is an election during which
only 1% of the votes are lost or manipulated and the winner wins by
/2% any worse than an election during which 30% of the votes were
irregularly cast or treated but the winner wins by 35% ? Should election
quality primarily be assessed on the basis of the electoral process or the
electoral outcomeor both ?

We do not offer a foolproof method for categorizing election quality,

but rather lay out a framework which is more comprehensive and
meaningful than anything that has come before. Using a consistent and
over-arching assessment model allows not only for cross-country
comparisons but also for comparisons of elections within a single
country over time. We believe the framework will identify patterns of
success and failure in the fairness of elections and be able to spotlight
the weak areas of election administration that a government might
reasonably focus its subsequent quality improvement efforts on.

We are also aiming to develop a logically-sound methodology, one

that lends itself in the context of election management to an
uncomplicated application, straightforward mapping, and easily-
determined quantification. The use of the framework in very different
environments is a strong argument for not applying complex statistical
methods, which may not be appropriate in all cases. This form of
modesty is also warranted when the phenomenon under scrutiny is
characterized by a considerable number of constantly changing
variables, many of which are difficult to measure in a precise, valid, and
reliable way.

We expect the model to be refined by the expertise of academics and

practitioners both for its scope and scoring methodology. We think it is
useful to take an election that one is familiar with and for which one
has access to relevant data, and then to score the case using our
methodology. Eyeballing the results alongside the examples we offer
here will give a good feel for the strength, or not, of the framework.
Jrgen Elklit and Andrew Reynolds 59

It looks as if established democracies might tend to lose points on

areas of election management such as transparency, voter education,
campaign regulation, and appeals processes. Our guess is that effective
provisions covering these areas have atrophied as public trust in the
system has grown over time. The framework therefore identifies a
potential Achilles heel in elections, even within stable democracies. A
thorough assessment of both voter registration and complaint
procedures in Florida prior to the 2000 US presidential election (and in
a number of other US states in 2004) would most likely have identified
the issues which marred the two electoral processes.

In fledgling democracies, the niceties of election law may be quite

robust, at least in the first competitive election, but the playing field
of electoral competition is often deeply skewed in favour of
dominant parties and elites. This is something that becomes even
more problematic over time. Our framework attempts to capture
both sides of the equation, the de jure and de facto rules that shape
elections, the written laws and the practical realities, the freeness of
the vote and the fairness of the campaign, as well as the chance to
win and the ability to lose.

Introducing the framework

One of the chief questions when trying to gauge the freeness and
fairness of an electoral process is where to draw the boundary when it
comes to deciding what issues are relevant to the question. The
boundary lines are murky. While we feel it is important to go beyond
polling day and the vote count, we exclude from our analysis the very
broad determinants of political competition that speak only more
indirectly to elections and voting. For example, we include questions of
access to public media and boundary delimitation while excluding more
general issues of party funding and candidate selection.

When it comes to the electoral indicators, our rule of thumb is not

to pronounce upon the inherent fairness of an electoral system or
regulation (if it is generally perceived to be a legitimate democratic
option), but rather to assess whether the rules, as written, are
applied fairly and without partisan bias. Kenya, to take just one
example, would not lose points because it uses a majoritarian rather
than a proportional electoral system but because its majoritarian
single member districts are so massively malapportioned, in a manner
which gives rise to partisan bias.
60 Judging Elections and Election Management Quality by Process

There will probably be criticisms of our framework, just as there are

valid criticisms of any assessment method that combines elements of
objective and subjective assessments and weightings of various
elements. Through our pilot studies, however, we find our method
defensible on grounds of providing results that are prima facie intuitive
and reasonable. The expert panels which we envisage would use the
framework should be knowledgeable, detached, and diverse, and we
believe that the data indicators identified give us the best purchase on
the questions we seek to answer. Of course country experts may assign
different scores within the 54 survey questions, and we encourage them
to do so; this merely indicates the frameworks sensitivity to a
continuum of indicators.

After settling on relevant areas of election regulation and

administration the issue becomes : Which questions must one ask to
gain a clear view of the workings of the given area, and what data
will serve as good indicators of electoral performance? In our model
we have eleven steps ranging from the initial legal framework to the
closing post-election procedures. We incorporate the areas an
Election Management Body (EMB) usually has responsibility over:
districting, voter education, registration, the regulation and design
of the ballot, polling and counting along with some broader areas
such as campaign regulation, complaints procedures, and the
implementation of electoral results.

Each step includes three to ten questions, the answers to which will
gauge the quality of election administration and conduct for that step.
In sum, there are 54 questions that act as our indicators. Some of them
may be criticized for not providing sufficient discrimination between
cases, and one may therefore argue that they should be excluded. Still,
at least some of them will help clarify that some components of
election administration are performed more or less in an identical
fashionan important point to be made. Some steps are analysed
primarily through reference to data such as specific voter education
efforts, while others are by necessity scored more on the basis of
expert judgments (i.e., the perceived legitimacy of the EMB, even
though this variable can also, at least in some cases, be gauged from
survey data). These answers will to some degree be based on data, but
more likely on expert readings and assessments of events and the
domestic political climate. However, we also believe that the scoring
on the performance indicators can be done, at least tentatively, by
election observers (i.e., typically the long-term observers).
Jrgen Elklit and Andrew Reynolds 61

Table 1 : Election assessment steps and performance indicators

Step Performance indicators How to measure

1. Legal 1.1. Is a consolidated legal foundation easily available ? Expert panel

framework 1.2. Is a comprehensive electoral time table available ? assessment
1.3. Were the elections held without extra-legislative delay ?
1.4. Can the electoral legislation be implemented ?
1.5. Is the electoral framework broadly perceived to be
legitimate ?

2. Electoral 2.1. What is the perceived degree of legitimacy/acceptance Polling evidence

management of the Electoral Management Body (EMB) by parties for perceptions
and voters ? Expert panel
2.2. What is the perceived degree of the EMBs impartiality ? assessment for de
2.3. What is the perceived quality of the EMBs delivery jure and de facto
of service in these elections ? analysis of EMB
2.4. What is the perceived degree of the EMBs impartiality
transparency ? Survey of stake-
holders for EMB
quality and

3. Constituency 3.1. Is the constituency structure reasonable and broadly Expert panel
and polling accepted ? assessment
district 3.2. Is information about constituencies and lower level Stakeholder surveys
demarcation districts (demarcation, sizes, seats) easily available ?
3.3. Are fair and effective systems for boundary limitation
and seat allocation in place and used according
to the rules ?

4. Voter 4.1. What percentage of voters in need of voter education In need is here
education is exposed to voter education which facilitates their operationalized as
effective participation ? first time voters.
4.2. Have at-risk groups been recognized and their At-risk are
identified needs addressed ? historically margin-
4.3. What percentage of ballots cast is valid ? alized groups.
4.4. In terms of voting age population, what percentage Voter education
of those eligible to vote for the first time in this outreach assessed
election actually voted ? through surveys
Other data from
register, polling,
and election results

5. Voter 5.1. What proportion of the voting age population Data from register
registration is registered to vote ? Expert panel
5.2. Is the register free from serious bias based on gender, analysis
age, ethnic or religious affiliation, or region ?
5.3. Are qualified people able to be registered with a
minimum of inconvenience ?
5.4. Are there appropriate mechanisms for ensuring that
the information in the register is accurate ?
5.5. Are there appropriate mechanisms for ensuring that
the public can have confidence in the register ?
5.6. Are the criteria for registration fair and reasonable
and compliant with accepted international standards ?
62 Judging Elections and Election Management Quality by Process

Step Performance indicators How to measure

6. Access to 6.1. Are parties allowed, and can parties and candidates Expert panel
and design of that fulfil the requirements of registration be registered assessment
ballot paper. without bias ?
Party and 6.2. Are independent candidates allowed and registered
candidate if they fulfil legal requirements ?
nomination 6.3. Is the method of voting or the design of the ballot
and paper non-discriminatory ?

7. Campaign 7.1. If there is a system to provide access to state-owned Expert panel

regulation media, is it implemented equitably ? assessment
7.2. If a system for allocation of public funds to political
parties is in place, is it implemented ?
7.3. Is there an independent mechanism for identifying
bias in the state media and is identified bias subject
to swift correction ?
7.4. Are state resources by and large used properly
by the political parties and candidates ?

8. Polling 8.1. What is turnout as a percentage of total registration ? Data from elections
8.2. What is turnout as a percentage of the voting age results and observer
population ? reports
8.3. Is there a low level of serious election-related violence ? Expert panel
8.4. In how many polling stations did polling happen assessments based
according to rules and regulations ? on data
8.5. Are there systems in place to preclude and/or rectify
fraudulent voting ?
8.6. Is polling accessible, secure, and secret ?
8.7. If there is substantial desire for election observation,
is the desire satisfied ?
8.8. If there is substantial desire for political party election
observation, is the desire satisfied ?
8.9. Are there systems in place to preclude vote buying ?
8.10. Is the level of intimidation sufficiently low that voters
can express their free will ?

9. Counting 9.1. Is the count conducted with integrity and accuracy ? Expert panel
and 9.2. Is the tabulation transparent and an accurate reflection assessments
tabulating of the polling booth count ? based on data
the vote 9.3. Are results easily available to interested members from observer
of the general public ? reports
9.4. Does counting take place with no undue delay ?
9.5. Are parties and candidates allowed to observe
the count ?
Jrgen Elklit and Andrew Reynolds 63

Step Performance indicators How to measure

10. Resolving 10.1. Are serious complaints accepted for adjudication ? Expert panel
election 10.2. Is there an appropriate dispute resolution mechanism assessments
related which operates in an impartial and non-partisan Reports
complaints. manner ?
Verification 10.3. Are court disputes settled without undue delay ? Legislation
of final 10.4. Do election observation organizations confirm that the Expert panel
result and elections were without serious problems ? assessments
certification 10.5. If legislation prescribes a timeframe for the constitution
of parliament, is this timeframe met ?
10.6. Is a person with a reasonable case able to pursue that
case without unreasonable personal or financial risk ?
10.7. Are seats taken only by those properly elected ?

11. Post- 11.1. Are properly documented election statistics easily Expert panel
election available without serious delay ? assessments
procedures 11.2. Are EMBs audited and the results publicly available ?
11.3. Is there capacity for election review ?

For consistency each question is answered with reference to a four

point scale (very good [3], good [2], not satisfactory [1] very poor [0]).
Assigning a score from this scale is, of course, ultimately a subjective
call, but we can offer guidelines in some areas (and will do so when
the manual is eventually presented) when it comes to the use of
indicative data. For example, when it comes to scoring questions of
turnout (questions 8.1 and 8.2) one might make the score dependent
on the cases deviation from the peer group average. A turnout of
80% in the Congo might be considered wonderful when compared
to peer group cases, while a turnout of 80% might be considered not
quite as stellar in Australia.

This model is akin to methodologies used for comparative

measures of democracy, human rights, and corruption by bodies such
as Polity, Freedom House, and Transparency International. It shares
the qualities of these indices as well as their problems (Munck and
Verkuilen, 2004), which we will not elaborate on here.1 Each scoring
system depends on both objective data indicators and subjective
expert assessments and they are all, as David Beetham says,
democracy assessment comparisons based on league tables of
human rights and democracy (Beetham, 2004 : 2-3 ; Munck and
Verkuilen, 2004).

In our pilot cases, the original, relatively-simple scoring system

proved inadequate to capture the differing pressures pertaining to
64 Judging Elections and Election Management Quality by Process

established versus fledgling democracies. Weightings are therefore

used to reflect step importance relative to each of the two types of
polity. Our rule of thumb was to ask : If this element fails, will that
cause the catastrophic breakdown of the electoral process ?

This assessment enabled us to assign essential, important, or

desirable status to each step, as indicated in Table 2, where one also
notes that assignments are not identical for the two types of polities.
To take an example : The standard of election management per se is
in our opinion essential in fledgling democracies because of the
nature of the problems surrounding the entire electoral process,
while election management in established democracies has become
more business as usual. It is still important (as the Florida 2000 case
made so abundantly clear), but failure does not have the same
implications for stability as within democratizing post-conflict
polities. Voter education is another example of an element to which
different importance should probably be attached in established and
fledgling democracies.

Table 2 : Weighting systems for established

and fledgling democracies

Essential Important Desirable

(weight factor : 3) (weight factor : 2) (weight factor : 1)

Established 1. Legal framework 2. Election management 4. Voter education

democracies 6. Access to ballot 3. Constituency demarcation 7. Campaign Regulation
8. Polling 5. Voter registration 11. Post-election procedures
9. Counting the vote 10. Resolving disputes

Fledgling 1. Legal framework 4. Voter education 3. Constituency demarcation

democracies 2. Election management 5. Voter registration 7. Campaign regulation
6. Access to ballot 11. Post-election procedures
8. Polling
9. Counting the vote
10. Resolving disputes

Pilot cases
We have opted for including quite different pilot cases in order to
assess how the instrument performs in systems with high-quality
election management traditions, in transitional systems with
elections run by the international community, in transitional systems
where it is possible to compare two or more elections, and a polity
Jrgen Elklit and Andrew Reynolds 65

generally believed to have a dismal election management system.

Unless otherwise indicated, we are dealing with parliamentary
elections to the lower (or sole) house.

The cases to represent established democracies with well-

functioning EMBs are Australia and Denmark, both of which
happened to have ordinary parliamentary elections in the second
half of 2001. As an example of a parliamentary election in a
fledgling democracy conducted by the international community (in
this case, the United Nations) we chose East Timor, which also had
its election in the second half of 2001. South Africa is another
fledgling democracy, but with its own strong election
administration (Padmanabhan, 2002). South Africa provides an
opportunity to assess performance over a decade and on three
different occasions ; we have settled for the elections in 1994 and
2004 to allow for over-time comparison. At the other end of the
scale, we expect to find Zimbabwe. We have chosen the
presidential elections of early 2002 (and not the parliamentary
elections of 2000) to work on the basis of the more recent elections.
However, there are so many similarities between the 2000 and 2002
elections in Zimbabwe that we believe the scoring of the 2002
presidential election will also, to a very high degree, reflect the
situation as it was in 2000.

Even though we have also considered the inclusion of other pilot

cases from sub-Saharan Africasuch as Zambia 2001 (Burnell, 2002;
Kamemba, 2002), Ghana 2000 (Smith, 2002), or Lesotho 2002
(Southall, 2003 ; Elklit, 2002) in order allow more regional
comparisonswe have decided to leave that for another, later
analysis. The principal aim here is to present the instrument and
invite comments on the method and framework.

Scoring (by a select group of experts and experienced observers

well versed in election matters in relation to the specific cases) and
computation of the index values for the six pilot cases was done in
the following way :
1. The first step was to allocate a score (0 : very poor ; 1 : not
satisfactory ; 2 : good ; 3 : very good) to each indicator for the
election in question. In binary situations, 0 and 3 were used. The
tentative scores are all found in Table 3. They are all subject to
correction at this point, but have been provided by evaluators
with good factual knowledge and an understanding of the various
66 Judging Elections and Election Management Quality by Process

systems. Each score is supposed to strike a balance between the

expectations in a given polity and internationally recognized
norms and standards.2
2. The sum of scores for each of the eleven sets of indicators (e.g.,
7.1-7.4) is then standardized relative to the value ten to make the
index insensitive to the number of indicators used for each step
and for ease of comparison across steps. This procedure also has
the advantage of softening the importance of decisions about
scoring of border-line cases, of which there are a number (i.e., Is
this a 1 or a 0?).
3. This standardized value is then multiplied by three, if the step is
considered essential, two if important, and left as it is (i.e.,
multiplied by one), if it is only desirable, as categorized in Table
2 above. This procedure caters to the various areas being of
different importance in established and emerging democracies.
4. Because of this, the maximum values differ240 for established
democracies and 270 for fledgling democracies. A transformation
to a maximum value of 100 (i.e., a further standardization) is
conducted in order to have values that are as comparable as

The result is a scoring system in which it makes sense to compare

polities in relation to their level of democratization. Indeed, this was
one of our ambitions in constructing the framework.

For the purposes of transparency, professional exchange, and

improvement of our methodology, Table 3 gives the detailed
scorings for the six pilot cases. Commentators with special insight
may disagree on the inclusion or the particular focus of one or more
of the 54 individual items and they may also disagree on the
individual scores tentatively allocated. We are happy to be corrected
if incorrect or arguable scores have been allocated in any of the
cases. The weighting of the various areas in established as well as
fledgling democracies is also not above criticism, and it may
eventually appear in a different form than what one now sees in
Table 2.

Despite our willingness to consider all objections or suggestions

for improvement very seriously, we are comfortable with the
resulting scores and their assessment of the level of electoral quality
in the polities and elections included here.
Jrgen Elklit and Andrew Reynolds 67

Table 3 : Performance indicator scores for six pilot case elections

Performance indicator Denmark East South South
Australia 2001 Timor Africa Africa Zimbabwe
2001 2001 1994 2004 2002

1. Legal framework :
1.1. Consolidated legal foundation ? 2 3 3 3 2 1
1.2. Comprehensive electoral time table ? 3 3 3 3 3 1
1.3 Elections held without extra-legislative
delay ? 3 3 3 3 3 3
1.4. Can electoral legislation
be implemented ? 3 3 2 2 2 1
1.5 Electoral framework generally
considered legitimate ? 3 3 3 2 3 1
Intermediate step scores 9.3 10.0 9.3 8.7 8.7 4.7
2. Electoral management :
2.1. Perceived degree of EMB legitimacy? 3 3 3 3 3 1
2.2. Perceived degree of EMB impartiality? 3 3 3 3 2 0
2.3. Perceived degree of quality in EMB
service delivery? 3 3 2 1 3 1
2.4. Perceived degree of EMB transparency? 1 2 3 2 2 0
Intermediate step scores 8.3 9.2 9.2 7.5 8.3 1.7
3. Constituency and polling district demarcation :
3.1. Constituency structure reasonable
and broadly accepted? 3 3 3 2 2 2
3.2. Constituency and lower level district
information easily available? 3 3 2 2 3 2
3.3. Fair system for boundary delimitation
and seat allocation in place? 3 3 3 3 3 2
Intermediate step scores 10.0 10.0 8.9 7.8 8.9 6.7
4. Voter education :
4.1. Voter education provided to voters
in need? 2 2 1 2 1 1
4.2. At-risk groups with needs identified
and needs addressed? 2 2 2 2 2 1
4.3. Percentage of ballots valid? 1 3 2 3 3 2
4.4. Turnout among first time voters,
in terms of voting age population (VAP) 2 2 3 2 1 2
Intermediate step scores 5.8 7.5 6.7 7.5 5.8 5.0
5. Voter registration :
5.1. Registration rate among VAP? 3 3 3 3 2 1
5.2. Register free from serious bias? 3 3 3 3 2 0
5.3. Level of registration inconvenience? 3 3 0 3 3 2
5.4. Mechanisms for ensuring accuracy
of registers? 3 3 2 1 2 0
5.5. Mechanisms for ensuring public
confidence in register? 2 3 2 1 2 0
5.6. Fair registration criteria, compliant with
international standards? 3 3 3 2 3 1
Intermediate step scores 9.4 10.0 7.2 7.2 7.8 2.2
68 Judging Elections and Election Management Quality by Process

Performance indicator Denmark East South South

Australia 2001 Timor Africa Africa Zimbabwe
2001 2001 1994 2004 2002

6. Access to and design of ballot paper.

Party and candidate nomination and registration :
6.1. Parties allowed, and can register
without bias? 3 3 3 3 3 3
6.2. Independent candidates allowed? 3 3 3 0 0 3
6.3. Method of voting or ballot design
non-discriminatory? 3 3 3 3 3 3
Intermediate step scores 10.0 10.0 10.0 6.7 6.7 10.0
7. Campaign regulation :
7.1. Systems to provide access to state-owned
media employed equitably? 3 3 3 3 2 0
7.2. If a system of public funding of parties
exists, is it implemented? 3 3 3 3 3 3
7.3. Independent mechanism for identifying
bias in state media, and correction
of such bias? 2 2 3 2 2 0
7.4. State resources used properly by parties? 2 3 3 2 2 0
Intermediate step scores 8.3 9.2 10.0 8.3 7.5 2.5
8. Polling :
8.1. Turnout as per cent of registration? 3 3 3 3 2 2
8.2. Turnout as per cent of VAP? 3 3 2 3 2 2
8.3. Low level of election-related violence? 3 3 3 1 2 0
8.4. Polling according to rules
and regulations? 3 3 3 3 3 2
8.5. Systems for rectification of fraudulent
voting? 1 2 3 2 2 1
8.6. Polling accessible, secure, and secret? 3 3 3 2 3 1
8.7. If desire for election observation,
is it satisfied? 3 3 3 3 3 0
8.8. If there is desire for party election
observation, is it satisfied? 3 3 3 3 2 1
8.9. Anti-vote-buying systems in place? 3 3 2 3 3 2
8.10.Level of intimidation? 3 3 3 1 2 0
Intermediate step scores 9.3 9.7 9.3 7.7 8.0 3.7
9. Counting and tabulating the vote :
9.1 Count conducted with integrity
and accuracy? 3 3 3 1 3 1
9.2. Tabulation transparent and accurate? 3 3 3 2 3 0
9.3. Results easily available? 3 3 3 2 3 1
9.4. Counting with no undue delay? 3 3 3 2 3 3
9.5. Are parties and candidates allowed
to observe the count? 3 3 2 3 3 1
Intermediate step scores 10.0 10.0 9.3 6.7 10.0 4.0
Jrgen Elklit and Andrew Reynolds 69

Performance indicator Denmark East South South

Australia 2001 Timor Africa Africa Zimbabwe
2001 2001 1994 2004 2002

10. Resolving election related disputes.

Verification of final results and certification :
10.1. Serious complaints accepted
for adjudication? 3 3 3 3 3 3
10.2. Appropriate dispute resolution
mechanism? 3 3 3 3 3 0
10.3. Disputes settled with no undue delay? 3 3 3 3 3 0
10.4. Election observation confirmation
if no serious problems? 3 3 3 2 3 0
10.5. Is timeframe for constitution
of parliament (if any) met? 3 3 3 3 3 3
10.6. Can persons with reasonable cases pursue
them without personal or financial risks? 2 3 3 2 2 0
10.7. Are seats taken only by people
properly elected? 3 3 3 3 3 3
Intermediate step scores 9.5 10.0 10.0 9.0 9.5 4.3
11. Post-election procedures :
11.1. Election statistics available with
no serious delay? 3 3 3 2 3 1
11.2. Are EMBs audited and results
made publicly known? 3 3 1 3 3 1
11.3. Is there capacity for election review? 3 3 2 3 3 1
Intermediate step scores 10.0 10.0 6.7 8.9 10.0 3.3
Weighted and standardized scores 89 93 83 72 77 41

The legitimacy and broad acceptance of any election depends on

the quality of the election management process, but only to a
certain degree. A fairly clean, well-managed election may
produce results completely unacceptable to losers in one country,
while a deeply-flawed election may be accepted in another. There
are a host of political, strategic and other factors that come into
elite legitimization of an election, which go well beyond the
process itself.

Therefore, it is not sensible to establish a certain total aggregate

score as the dividing line between legitimate and illegitimate
elections, where the latter then would be rejected. As argued by
Horacio Boneo at the IPU round table, two cases could both score
75 out of a 100 and have problems in quite different areas. It
makes more sense to sensitize the scoring methodology to give
primacy to the most important electoral process components in
70 Judging Elections and Election Management Quality by Process

both established and fledgling democracies, but even so local

realities will impact greatly on the perception of how good an
election (and the administration thereof) actually is (or was).

The final weighted and standardized scores offered in Table 3

do not constitute the final word in our analysis. They are included
for purposes of transparency and as a further indicator of election
and election administration quality both over time and across
nations. The intermediate standardized scores for each of the
eleven steps are also included for the purpose of more focused

The aggregate scores make sense intuitively. Yet in our view

(which was shared by participants at the round table), the
individual (and intermediate) scores are considerably more
interesting. They offer particular insight in comparing elections in
a specified country, point out areas where improvements in
election administration are particularly needed, and help observers
assess an elections influence on political legitimacy and

As argued by Eric Rudenshiold of IFES during the IPU round table,

the analysis becomes particularly tricky when a democracys status
changes from fledgling to established and assessment scores
must be altered. One possibility is to introduce a system with finer
categorizations (containing three or more categories) in the
context of developing the overall analytical framework. Such
changes will obviously complicate the over-time comparisons,
with the consequence being that category weightings within a
country may change as time progresses.

The framework functions well and offers a useful starting point for
future attempts of identifying levels of electoral governance
performance in all kinds of democracies. It provides a tool that
enables us to assess electoral processes in a more systematic way
than has previously been possible, allowing for comparisons of
electoral quality within and across regions and across time. In itself,
the implementation of this approach should prove useful. At the
same time, however, its implementation may also advance our
understanding of what fosters the development and stability of
democratic legitimacy.
Jrgen Elklit and Andrew Reynolds 71

We invite general comments on the framework presented here,

along with more specific evaluations that may contribute to a
greater understanding of our analysis of the six elections and the
assessments and scorings documented in Table 3. A subsequent step
is to invite interested colleagues, election practitioners, and others
to join in on our efforts to evaluate an even broader coverage of
countries and elections than are offered in this paper. This kind of
cooperation is necessary to ensuring the success of such a project,
one that solicits country- and case-specific expertise and insight.

We foresee the formation of assessment teams (also expert

teams) for individual countries, consisting of two to three
international and two to three domestic assessors coming together
in a specified country to discuss their case(s) and its performance in
relation to the various indicators. The basic guidance in this work
will come from a general (first) manual to be developed within the
project providing the cues for the decisions (in line with some of
the indications in Table 1 above, but obviously more detailed and
building on established international standards, when such
standards exist, a point strongly underlined by Ron Gould at the IPU
round table). This should allow the various expert teams to work
towards a common goal. At the same time, however, we foresee
the analyses going beyond the simple scoring mechanisms outline
above. Behind each score, we will expect an indication in clear
language of the reasons for the level suggested. This
documentation should be available in writing and accessible to
interested individuals via a home page on the Internet.

Based on our initial contacts in a number of countries, we have

become convinced that considerable enthusiasm exists for gaining
access to this kind of instrument. While interest is evident in
academic circles, we expect election administrators to express an
even greater desire to participate, given their genuine interest in
designating areas where improvement is needed, whether in
established or in fledgling democracies. Perhaps of greatest
importance, we expect that election observers and their various
organizations can also make use of such an instrument, providing
a better means for assessing a given election and facilitating
meaningful comparisons between elections.

Once this work has developed further, a more definitive manual

will be produced for use in assessing any election where domestic
72 Judging Elections and Election Management Quality by Process

and/or international observer organizations and political actors,

etc., express interest in systematically judging an electoral event
across systems, within and across regions, and over time.

About the authors : Jrgen Elklit is Professor in the Department

of Political Science at the University of Aarhus, Denmark. He has
acted as an advisor on election assistance, election administration,
electoral systems and democratization in Africa, Europe, and Asia.

Andrew Reynolds is Associate Professor in the Department of

Political Science at the University of North Carolina in the United
States. He has served as a consultant on electoral and constitutional
design issues in Africa, Europe, the Pacific, and South America.

References :
Beetham, David (2004) Towards a universal framework for democracy
assessment, Democratization, 11 (2) : 1-17.
Burnell, Peter (2002) Zambias 2001 elections : The tyranny of small decisions,
non-decisions and not decisions, Third World Quarterly, 23 (6 ) : 1103-20.
Diamond, Larry and Marc F. Plattner (eds.) (2002) The Global Divergence of
Democracies, Baltimore : The Johns Hopkins University Press.
Elklit, Jrgen (1999) Electoral institutional change and democratization : You
can lead a horse to water, but you cant make it drink, Democratization, 6 (4) :
Elklit, Jrgen (2002) Lesotho 2002 : Africas first MMP elections, Journal of
African Elections, 1 (2) : 1-10.
Elklit, Jrgen and Andrew Reynolds (2001) Analysing the impact of election
administration on democratic politics, Representation, 38 (1) : 3-10.
Elklit, Jrgen and Andrew Reynolds (2002) The impact of election
administration on the legitimacy of emerging democracies : A new comparative
politics research agenda, Commonwealth and Comparative Studies, 40 (2) :
Elklit, Jrgen and Andrew Reynolds (2005) A framework for the systematic
study of election quality, Democratization, 12 (2) : 1-16.
Elklit, Jrgen and Palle Svensson (1997) What makes elections free and fair?,
Journal of Democracy, 8 (3) : 32-46. Reprinted in Diamond and Plattner (eds.)
(2002), pp. 200-14.
Goodwin-Gill, Guy S. (1994) Free and Fair Elections : International Law and
Practice, Geneva : Inter-Parliamentary Union.
Kamemba, Kambuya (2002) Looking at the management of the 2001 Zambian
tripartite elections, Journal of African Elections, 1 (2) : 11-26.
Kjr, Mette (2004) Governance, Cambridge : Polity.
Jrgen Elklit and Andrew Reynolds 73

Lehoucq, Fabrice and David L. Wall (2004) Explaining voter turnout rates in
new democracies: Guatemala, Electoral Studies, 23 (3) : 485-500.
Lpez-Pintor, Raphael (2000) Electoral Management Bodies as Institutions of
Governance, New York : UNDP.
Lyons, Terrence (2004) Post-conflict elections and the process of demilitarizing
politics : The role of the electoral administration, Democratization, 11 (3) :
Munck, Gerardo L. and Jay Verkuilen (2004) Conceptualizing and measuring
democracy : Evaluating alternative indices, Comparative Political Studies,
35 (1) : 5-34.
Mozaffar, Shaheen (2002) Patterns of electoral governance in Africas
emerging democracies, International Political Science Review, 23 (1) : 85-101.
Mozaffar, Shaheen and Andreas Schedler (2002) The comparative study of
electoral governance Introduction, International Political Science Review,
23 (1) : 5-27.
Norris, Pippa (2004) Electoral Engineering : Voting rules and Political Behavior,
Cambridge: Cambridge University Press.
Padmanabhan, Vijay (2002) Democracys Baby Blocks: South Africas Electoral
Commissions, New York University Law Review, 77 : 1157-94.
Schedler, Andreas (2002) The nested games of democratization by elections,
International Political Science Review, 23 (1) : 103-22.
Smith, Daniel A. (2002) Consolidating democracy? The structural
underpinnings of Ghanas 2000 elections, Journal of Modern African Elections,
40 (4) : 621-50.
Southall, Roger (2003) An unlikely success: South Africa and Lesothos election
of 2002, Journal of Modern African Studies, 41 (2) : 269-96.

Notes :
Munck and Verkuilen (2004) demonstrate the various problems of
conceptualizing and measuring democracy one finds in the different indices
of democracy currently available. A comparison with their stringent approach
and useful analyses of previous attempts of indexation is somewhat
discouraging because it is not (yet?) possible to live up to the high and very
reasonable standards to which they subscribe. However, the framework
suggested is only a first attempt to deal with some specific issues and the
outcome should in any case be assessed on whatever merits it has.
As reflected in publications such as Principles for Election Management,
Monitoring, and Observation in the SADC Region (Johannesburg: Electoral
Institute for Southern Africa, 2004) and International Electoral Standards.
Guidelines for reviewing the legal framework of elections (Stockholm:
International IDEA, 2002).
July 2013, Volume 24, Number 3 $12.00

Latin Americas Authoritarian Drift

Kurt Weyland Carlos de la Torre Miriam Kornblith

Putin versus Civil Society

Leon Aron Miriam Lanskoy & Elspeth Suthers

Kenyas 2013 Elections

Joel D. Barkan James D. Long, Karuti Kanyinga,
Karen E. Ferree, and Clark Gibson

The Durability of Revolutionary Regimes

Steven Levitsky & Lucan Way

Kishore Mahbubanis World

Donald K. Emmerson

The Legacy of Arab Autocracy

Daniel Brumberg Frdric Volpi Frederic Wehrey Sean L. Yom
Kenyas 2013 Elections

Is Not DemocracY
Joel D. Barkan

Joel D. Barkan is professor emeritus of political science at the Univer-

sity of Iowa and nonresident senior associate with the Africa program
of the Center for Strategic and International Studies. He is the author
of the 2011 CSIS report Kenya: Assessing Risks to Stability.

E lections are a cornerstone of democracy, but elections that are flawed

or perceived to be flawed retard rather than advance democratization.
Elections work as a procedure to select state officials only when the
losers accept the results as the outcomes of a legitimate process. This is
particularly true when the electoral contest is close and for the highest
office in the land, as was the case in Kenya in 2007 and 2013.
The 2013 elections were Kenyas fifth since the country returned to
multiparty politics and the direct election of its president in 1992. Three
of the preceding four elections were marked by violence. But in only
onethe 2007 presidential contestwas the violence triggered by deep
dissatisfaction with the electoral process and the electoral commission
that administered the balloting. In that election, the losing candidate,
Raila Odinga of the Orange Democratic Movement (ODM), had led ev-
ery preelection public-opinion poll by decisive margins. He had also
led in the reporting of more than two-thirds of the polling stations by a
clear margin of 400,000 votes. Yet he lost the election by a margin of
2.3 percent (46.4 to 44.1) to Mwai Kibaki, the incumbent president, who
was hastily sworn into office behind closed doors following a media
blackout on the last day of the count. The rest is well known: Odinga
supporters attacked Kibaki supporters, who responded in kind. By the
time the violence was all over, more than 1,100 people were dead, and
over 660,000 had been displaced from their homes.
The 2007 election started off as one of the most peaceful in Kenyas
history. Turnout was high, and, despite minor logistical glitches, the
election proceeded smoothly through the vote count and the announce-
ment of the results at more than 24,000 polling stations. The agents

Journal of Democracy Volume 24, Number 3 July 2013

2013 National Endowment for Democracy and The Johns Hopkins University Press
Joel D. Barkan 157

of losing candidates, though disappointed, generally accepted the out-

comes with grace. The problem was with the transmission of the re-
sults from the polling stations to the district electoral headquarters, the
tabulation of the polling-station results at the district level, and finally
the onward transmission of the vote count to the headquarters of the
Electoral Commission of Kenya (ECK) in Nairobi. The chair of the ECK
publicly acknowledged that he had lost contact with senior staff at the
district level, where the votes from all polling stations were first aggre-
gated before being forwarded to Nairobi. Yet Kibaki, who had packed
the ECK with his own appointees, was soon declared the winner under
mysterious circumstances. Not only was there a breakdown in commu-
nication and alleged fraud in reporting the vote, but the vote totals for
many areas exceeded the number of registered voters.
In the wake of the violence and the National Accord that restored peace
in February 2008, the ECK was disbanded and replaced by two interim
commissionsone for the delimitation of constituency boundaries and
another for administering elections. Following the August 2010 referen-
dum approving the new constitution, an exhaustive process began to se-
lect and appoint a single new Independent Election and Boundaries Com-
mission (IEBC), as specified in the document. An elaborate public-vetting
process ensued to ensure the independence, competence, and legitimacy
of the IEBC for the next elections. The IEBC then went about delimit-
ing 290 new constituencies for the National Assembly, as required by
the new constitution. The IEBC also conducted several parliamentary by-
elections, all of which went well and generated no postelection disputes.
Public confidence in the IEBC was high. According to several public-
opinion surveys, 80 percent of the public trusted the IEBCa higher level
of public confidence than for any other major government institution, in-
cluding the presidency, the judiciary, and the parliament.

Embracing Technology
As the IEBC prepared for the next election, it looked to the latest
technology to avoid repeating the two biggest failings of the 2007 poll:
1) The lack of an accurate, up-to-date, and secure register purged of de-
ceased Kenyans; and 2) the failure to transmit the vote accurately from
thousands of polling stations to constituency counting centers, to ag-
gregate those results properly, and then to forward them on to the ECK
headquarters for final tabulation and the announcement of the outcomes.
Whereas Kenyas previous elections had relied on a voter register
compiled from paper forms that were later scanned into a computer
database, the new system would be completely computerized from the
point of registration to the production of the voter list. Most important,
to guarantee against possible fraud the new national register would in-
clude not only the names, national identification numbers, gender, and
158 Journal of Democracy

addresses of those registered, but also biometric data in the form of

digitally scanned fingerprints and a digital photo of each person regis-
tered. This biometric voter-registration
system known as BVR would pre-
Technology would ensure vent fraud by requiring all voters ar-
an election that was free riving at their assigned polling station
and fair and thus legiti- on election day to verify their identity
by having their fingerprints digitally
mate. Losing candidates
rescanned and compared with the bio-
would accept the out-
metric data stored on the computer-
come without claiming ized national register. As a safeguard
that it had been rigged. against multiple voting, the national
database would be updated throughout
election day. Once a voter signed in at
the polls, his or her presence would be recorded and struck off the list
of those eligible to vote.
The use of the BVR and national database would also enable the
IEBC and the public to monitor the progress of the vote in real time.
The system would provide a national moving-picture map of turnout
(using geographic information system software) showing what percent-
age of the electorate had voted in each parliamentary constituency and
county and in the country as a whole. This would all be accessible on-
linethe ultimate in transparency.
Finally, the BVR and the computerized voter-ID system would be
complemented by another hi-tech solutiona computerized results-
reporting system that would ensure the fast and accurate transmission of
the vote from 33,100 polling stations to the IEBCs reporting headquar-
ters in Nairobi. The system would include 33,100 mobile phones pro-
grammed with a special application designed for this purpose. The presid-
ing officers at each polling station would be issued a phone with a unique
password that they would use to securely log in to the IEBCs central
computer and text the results for each of the six offices being contested
in the election.1 The data from each polling station would be captured by
the IEBCs computer servers, after which they would be aggregated for
each electoral contest and made available immediately online and to the
press. The system was highly sophisticated in order for the IEBCs servers
to capture, tally, and keep separate the results for the six different offices
being filled. For the presidential contest and other high-profile races, such
as those for the new governorships, the results would also be displayed in
real time on big screens at the IEBC and broadcast by Kenyas television
networks. This, too, would be the ultimate in transparency.
By opting for these systems, the IEBC sought a twenty-first-centu-
ry solution to Kenyas history of electoral mismanagement and fraud.
Technology would ensure an election that was free and fair and thus
legitimate in the eyes of the people. Losing candidates would accept the
Joel D. Barkan 159

outcome without claiming that it had been rigged. The IEBC would
maintain its legitimacy. And, most important, those elected would enter
office with the full authority of their mandate.
The IEBCs approach was not surprising given the recent strong em-
brace of technology by Kenyas private sector. The adoption of mobile
phones and related applications in Kenya is arguably the most impressive in
all of Africa. Out of Kenyas population of 43 million, there are more than
38 million mobile-phone usersnearly every adult, rich or poor, now has
one. Mobile-phone networks cover all but the sparsely populated northern
areas of the country. The development and spread of the Safaricom mobile
networks M-pesa money-transfer system, along with its imitation by other
mobile-phone providers and banks, has been a model for other developing
countries. For the first time, millions of Kenyans, including those in the
informal sector, have become part of the national economy. Mobile-phone
coverage has improved the lot of small producers, such as the now-myth-
ical fisherwoman on the shores of Lake Victoria who brings fresh fish to
the Kisumu market on demand via mobile-phone orders rather than taking
her entire catch to rot in the market from dawn until dusk. Small entre-
preneurs using solar charging devices have set up shop in areas without
electricity to facilitate the use of mobile phones across the countryside. In
Kenya, new communication technologies have displaced the corrupt and
poorly maintained state-owned landline phone system inherited from the
big man era of former president Daniel arap Moi.
Kenya also has one of the highest rates of Internet penetration in
Africa. Roughly a fifth of the population has regular Internet access, a
rate that is rising as fiber-optic cable is laid across Nairobi and between
major population centers. Kenyas domestic Internet system is linked to
the outside world by two undersea fiber-optic cables, with a third on the
way. None of this connectivity existed five years ago. Kenyas press,
arguably the freest and most sophisticated in Africa, is available online
to Kenyans at home and to the large diaspora abroad. Kenyas television
stations are increasingly streaming content online. The countrys back
office and software industries, though dwarfed by Indias, are beginning
to establish themselves in world and regional markets. As noted earlier,
Kenyas software-development capacity has powered the expansion of
financial services not only in Kenya but increasingly across Eastern Af-
rica as Kenyan banks move into adjacent markets.2
The World Bank estimates that information technology (IT) now ac-
counts for a full percentage point of Kenyas recent annual economic
growth of 4 to 6 percent. Its development and spread is driven mainly
by a young generation of entrepreneurs and professionals who are part
of Kenyas expanding urban middle class. Given this culture of tech-
nology adoption, it is not surprising that the IEBCitself chaired by a
43-year-old lawyeropted for the latest. Though not articulated in
these words, technology was viewed as a shortcut to democracy.
160 Journal of Democracy

Kenyas political classincluding Kibaki and Odinga, president and

prime minister, respectively, at the timesupported the IEBCs plans
for the BVR, the computerized national register of voters, and the elec-
tronic results-reporting system. Notwithstanding their deep suspicions
lingering from the 2007 polls, Kenyas politicians believed that the new
technologies would save the country from another election debacle.

Problems of Implementation
Though tantalizing, the BVR system was expensive and logistically
complicated to implement. First, the IEBC needed to acquire and program
15,000 laptop computers, each of which would be equipped with digital
scanners to scan the fingerprints of those registering to vote. The equip-
ment also included cameras to record digital photos of those registered
and appropriate software to facilitate quick and reliable use of the equip-
ment by election officials at each registration station. Generators were also
needed to guarantee continuous power for the laptops used in areas without
electricity. The IEBC also recruited and trained 30,000 workers to con-
duct the registrationtwo for each registration centerso that the centers
could operate seven days a week during the month-long enrollment period.
The choice of BVR also meant that a second set of laptops and finger-
print scanners needed to be purchased and loaded with the appropriate
software as well as the national-register data so that voters could be prop-
erly identified on election day. Known as electronic voter-identification
devices or EVIDs, 33,100 were eventually purchased by the IEBC, one
for each polling station. Because generators were cumbersome and ex-
pensive, the IEBC decided instead to equip each EVID laptop with three
rechargeable batteries to ensure sufficient power on election day. The
number of EVIDs was more than double the number of laptops required
for the BVR because the number of polling stations was more than double
the number of voter-registration centers.3 As noted previously, the results-
reporting system also required the procurement of 33,100 mobile phones.
The final cost of the BVR ballooned to over US$72 million, while the es-
timated cost of the EVIDs was between $27.2 and $34 million. At roughly
$20 per mobile phone, the estimated cost of the results-reporting system,
including the central server, was relatively cheap, no more than $1 million.
The total cost of all the computers, mobile phones, and accessory equipment
is estimated to have run as high as $120 millionabout $10 for each of the
12.2 million Kenyans who ultimately took part in the election. The total
equipment cost, however, did not include all the other expenses associated
with the election, such as the operating costs of the IEBC itself, salaries for
the officials who registered voters and staffed the polling stations on elec-
tion day, and the cost of printing ballots, ballot boxes, and other material.
The cost of going high-tech also presented potential equipment pro-
viders with the prospect of highly profitable contracts. The result was
Joel D. Barkan 161

a procurement nightmare. The IEBC received bids to supply the BVR

equipment from 27 contractors in December 2011, more than a year
before the elections. But the IEBC then extended the deadline for bids
by 90 days and dawdled until May 2012 in reviewing them. All but four
bids were judged to be technically flawed, while those that remained
were deemed wanting on cost and other fiduciary grounds. Thus in early
August 2012, the original target date to begin registration, the IEBC was
faced with the challenge of starting over.
At this point, the technical advisors to the Commission provided by the
International Foundation for Electoral Systems (IFES) and UNDP advised
the IEBC to abandon the BVR and replace it with a highly reliable and
significantly cheaper alternative, Optical Marked Registration (OMR). A
manual system, OMR records a prospective voters name, age, national
ID number, and other information by filling in the blanks on a preprinted
form, much as one fills in the answer sheet for an electronically graded
examination. The system requires no special equipment or electric power
at the locations where the data is collected because the information sheets
completed for each registrant are then processed by a small number of
optical scanners at a single central facility or small group of disbursed
facilities to produce the voter roll. The OMR process requires less training
and, at an estimated cost of $20 million, OMR costs less than a third of
what the BVR did. To their credit, IEBC chairman Issack Hassan and the
other members agreed to the switch. Unfortunately, they were overruled
by President Kibaki and his cabinet, including Prime Minister Odinga.
A second round of procurement for the BVR thus began in Septem-
ber, at which point the Canadian Commercial Corporation stepped for-
ward to offer the IEBC assistance in obtaining the necessary equipment.
Though initially understood by the IEBC to be in-kind assistance or a
loan from the Canadian government at concessional rates, the deal was
in fact a commercial loan to the Kenyan government at a 5 percent inter-
est rate. The loan was also tied to the purchase of Canadian equipment,
though the eventual supplier, Safran Morpho, was actually a French
company that maintained a subsidiary in Montreal. Financing for the
deal was not finalized until late October after lengthy negotiations that
denied Safran Morpho proprietary rights to the registration database.4
Of the 15,000 registration kits, none arrived in Kenya until November
5, four months before the election. The registration process finally got
underway on November 19, more than three months late and now slated
for a period of thirty rather than sixty days as required by law.
To the surprise of many knowledgeable observers, the BVR process
went fairly smoothly. Yet because of the late start and compressed time
period, the IEBC had registered only 14.3 million people, or two-thirds
of the estimated 21.5 million Kenyans eligible to vote, by the close of
registration on December 18. Using outdated census figures, the IEBC
declared victory by claiming to have registered 79 percent of those eli-
162 Journal of Democracy

TablePercentage of Constituencies by Level of Success at

Reporting Results Electronically
Constituencies where: Presidency Parliament
020% of polling stations reported successfully 27 59
2140% of polling stations reported successfully 24 12
4160% of polling stations reported successfully 17 15
6180% of polling stations reported successfully 19 10
81100% of polling stations reported successfully 13 8
Number of observations (290) (290)
gible; somewhat surprisingly, it was not criticized for what was consid-
ered by seasoned observers to be a distortion of the truth.
Once registration was complete, the IEBC had little more than two
months to finish all other preparations for the poll. The Commission was
in a race against the clock. The IEBC thus quickly went about procur-
ing the ballots (from a British firm), the EVIDs, and the mobile phones
required for the results-reporting system. Because time was short, the
normal tendering and bidding process was short circuited.
The EVIDs, which were supplied by the South African firm Face Tech-
nologies Ltd., arrived less than two weeks before the election. Procurement-
related delays once again held up delivery and also changed the configu-
ration of the equipment. As noted previously, for the EVIDs to work as
intended, the laptops had to be equipped with the proper software and a
copy of the national voter register, including the BVR data. IEBC staff
toiled around the clock in a warehouse the week before the election to load
each of the 33,100 laptops. They finally finished on Saturday, March 2, less
than 48 hours before the polling stations opened on Monday, March 4. The
machines were to be sent out with three fully charged batteries for each,
but in the rush most were shipped to their assigned polling stations with
only one battery fully charged, resulting in a loss of power and complete
breakdown. Half of the EVIDs stopped working midway through election
day. The IEBC responded in the only way that it could, by abandoning the
system and reverting to a backup paper registera printout of the informa-
tion and photo contained in the computerized register but not the fingerprint
scans central to the BVR. Presiding officers at nearly all polling stations
reacted with aplomb, perhaps because they had always been skeptical of
whether the EVIDs would work. But they had, in fact, worked as designed
and touteduntil they ran out of juice! After that, the election was depen-
dent on the resourcefulness of the presiding officers in the field.
A similar problem befell the special mobile phones with which the
presiding officer at each polling station was to transmit the vote totals to
IEBC headquarters. In the rush to distribute the phones, some presiding
officers were not given the password needed to log in to the IEBCs cen-
tral server. Others were not provided with proper SIM cards to access
the mobile-phone network in their area. Still others were given phones
whose batteries had not been fully charged.
Joel D. Barkan 163

Table 1 above summarizes the extent of the breakdown. In only 13

percent of Kenyas 290 constituencies were 80 percent or more of poll-
ing stations able to report the vote for president electronically; the level
of successful transmissions for the parliamentary votes was even less.
The electronic results-reporting system was thus also abandoned a day
and a half into the count, while the Kenyan public was avidly follow-
ing the results on television and online. As the hours passed, the system
first slowed and then stopped updating the results. Before that, however,
television viewers and commentators had noticed that the number of
rejected ballots for the presidential race was approaching 5 percent
of the total votes cast, an unusually high figure that raised doubts over
whether Uhuru Kenyatta could pass the threshold of 50 percent plus 1
vote. The IEBC acknowledged that it had discovered a software error
that had mysteriously multiplied the number of rejected ballots by eight.
Here, too, the promise of technology proved too good to be true. Indi-
vidual components of the system had tested fine. Scaled-down versions
of the system had also tested fine.5 But in the rush to put the many thou-
sands of moving parts together, the system as a whole failed. Simply
stated, the IEBC was overwhelmed by the challenge of completing too
many tasks in too short a time. In the process, the Commissions cred-
ibility and the credibility of the election itself took a hit.6

Lessons Learned
The foregoing discussion would be interesting only to Kenyans and
Kenya watchers were it not for the broader lessons to be learned. There
are severalfor Kenya and for other countries contemplating high-tech
solutions to their own election problems. There are also lessons for the
international community, which has become heavily involved in promot-
ing these and other solutions, such as parallel vote tabulations (PVTs), to
enhance the credibility and fairness of elections in emerging democracies.
For Kenya and Other High-Tech Adopters: Four lessons are relevant
for the future of election administration in Kenya. First, the challenge
is not one of making the technology work, but rather one of embedding
technology in an organization involving thousands of people that oper-
ates for only short periods of timethat is, the month during which voter
registration takes place, election day itself, and the period immediately
thereafter. In a country like Kenya, where mobile phones and computers
are part of daily life, the challenge is not a tech challenge, but a human
systems and managerial one. Second, like institution-building generally,
building a viable and independent election commission that ticks along
like the proverbial Swiss watch takes time. The process cannot be rushed.
Third, Kenyans, especially Kenyan professionals and members of the
political class, need to make two cultural adjustments: 1) Notwithstanding
the many marvelous things that IT has brought to the country over the past
164 Journal of Democracy

decade, Kenyans must maintain a healthy skepticism and ask hard ques-
tions, particularly the question of whether the most sophisticated forms of
technology (BVR) are more appropriate than less-advanced forms that are
easier to use, less expensive, and more reliable (OMR); and 2) they must
resist the hakuna mutata (Swahili for no problem) mantra that enables
millions of Kenyans, especially the poor, to get through lifes daily strug-
gles, but which grossly minimizes the challenges associated with imple-
menting high-tech solutions, especially the requirement of time. This was,
arguably, the most serious failing of the IEBCthe repeated overestima-
tion of what could be accomplished by election day, even as the original
deadlines for setting up the IT systems for voter registration, voter sign-in,
and results-reporting continued to be missed.
The fourth lesson relates to procurement. By opting to go high-tech,
the IEBC compounded the first three challenges because the high cost of
equipment attracted a wave of profiteers (some qualified, but most not)
who made big promises in order to land lucrative contracts. The first at-
tempt at procuring equipment for the BVR sounded the alarm loud and
clear, but the IEBC ignored it. When procuring such equipment, there
must be a set plan for a lengthy and proper review of every proposal
before a contract is signed so that all the components can be on site
months before they are used. Only then can technology be successfully
embedded in the organization recruited to carry out the required tasks.
For the International Aid Community: Having spent hundreds of
millions of dollars to support multiparty elections in Kenya and other
countries since the early 1990s, the international donor community,
especially the bilateral-aid agencies, needs to take stock of what has
been accomplished and where this enterprise is headed. Only a minor-
ity of elections in emerging democracies are regarded as competently
run free and fair contests. Moreover, the outcomes of these elections
and the manner in which they are conducted are largely a function of
variables that are outside donor control. Nearly all African countries
have conducted four or five multiparty elections since the resumption
of multiparty politics two decades ago. It is no longer an option for the
international community to conduct elections on behalf of African states
as it did in Angola and Mozambique in the early 1990s, as that would be
a violation of national sovereignty and pride. At best, the international
community can assist from the sidelines and at the margins.
There are four broad types of assistance that the international com-
munity can appropriately provide, and Kenya has over the years re-
ceived every one: 1) technical assistance to the election commission
in designing better ways to conduct election-related processes and in
building the organizational capacity of the election-management body;
2) funds or equipment for conducting the election; 3) financial and tech-
nical support to civil-society organizations for mounting voter-educa-
tion programs and for monitoring the electoral process, from the earliest
Joel D. Barkan 165

preparations a year or more before the event through the vote count; and
4) support for the presence of international observers both to monitor the
election and to backstop domestic-observer operations. None of these
options involves taking direct control of the process. At the same time,
none has the ability to shape the local culture of citizens, political elites,
or the heads and members of electoral commissions.
Given these realities, donors should be more cautious than they have
been in the past, especially when it comes to recommending or financ-
ing high-tech solutions, given the procurement baggage that they bring.
In the Kenyan case, the technical advisors provided by UNDP and IFES
(which was funded by USAID) advised against the most high-tech op-
tion, but that advice was turned down. Should those advisors then have
been withdrawn? Were they right to remain and give their best efforts
to make the Kenyans choice work? Should the Canadian Commercial
Corporation have financed the deal for the BVR? The international do-
nor community needs to develop a checklist of requirements before be-
coming part of a process that it can never control. Doing no harm
should be the first item on the list.

The author gratefully acknowledges the comments made on an earlier draft of this article
by Michael Yard, senior elections advisor at IFES. Responsibility for this analysis, how-
ever, is the authors alone.

1. In addition to the presidency, Kenyans elected members of a new bicameral national

legislature, governors of the new counties, members of the new county assemblies, and 47
women county representatives.

2. An example is the expansion of Kenyan-owned Equity Bank into Uganda and South
Sudan, where it operates branches as well as a network of ATMs linked via satellite to its
central computer in Nairobi.

3. The number of polling stations was more than double the number of registration
centers because the registration process lasted one month, whereas the election took
place on one day. The number of polling stations was set at 33,100, on the calculation
that no station or voting stream should process more than 450 voters during the eleven-
hour period that the polls were open.

4. In its proposed contract to the IEBC and the Government of Kenya, Safran Morpho
included a clause asserting a proprietary right to all data captured during the registration
process. By demanding control of the database, the firm sought to lock in an advantage for
future work. The clause was ultimately dropped.

5. The results-reporting system was tested in a mock election two weeks before the
actual election, but with test data from only a handful of constituencies.

6. Defenders of the IEBC point out that the electronic results-reporting system was
from the start intended only to apprise Kenyans of the provisional results of the elec-
tion, and that the official count was always intended to be undertaken manually over a
period of one week. While true, the mere existence of the electronic results-reporting sys-
tem raised expectations on the part of both citizens and candidates that went unfulfilled.
January 2011, Volume 22, Number 1 $12.00

The Impact of the Economic Crisis

Marc F. Plattner Larry Diamond

Two Essays on Chinas Quest for Democracy

Liu Xiaobo

The Split in Arab Culture

Hicham Ben Abdallah El Alaoui

Migai Akech on the Uses and Abuses of Law in Africa

Harold Trinkunas & David Pion-Berlin on Latin American Security
Christoph Zrcher on Peacebuilding
Ngok Ma on Pact-Making in Hong Kong
Shadi Hamid on Islamist Parties

Latin America Goes Centrist

Michael Shifter Javier Corrales Eduardo Posada-Carbo
constraining government
power in africa
Migai Akech

Migai Akech is an independent legal scholar based in Nairobi, Ke-

nya. He has taught at the New York University School of Law and at
the University of Nairobi, where he was senior lecturer.His writings
include Privatization and Democracy in East Africa: The Promise of
Administrative Law (2009). He was a Reagan-Fascell Democracy Fel-
low at the National Endowment for Democracy in Washington, D.C.,
from October 2009 to February 2010.

Most analysts have long contended that African politics can largely
be explained by reference not to formal but to informal institutions, and
above all to neopatrimonialism. This term, coined in the early 1970s to
identify a seeming variation on Max Webers notion of patrimonialism,
is meant to identify a hybrid political regime in which informal patron-
client relationships both underlie and overshadow legal-rational norms.1
This literature privileges the patrimonial; legal-rational norms hardly
matter. According to this binary view in which the formal can be neatly
separated from the informal, Africa is a place where formal institu-
tional rules are largely irrelevant.2
Recent revisions of this dominant thesis acknowledge that formal in-
stitutions, or legal-rational norms, are beginning to matter. For example,
Daniel Posner and Daniel Young see the Nigerian Senates 2007 rejec-
tion of a proposed constitutional change that would have permitted Pres-
ident Olusegun Obasanjo a third term in office as a sign of formal rules
growing importance.3 In their view, it matters that leaders who want to
get around constitutional restrictions such as term limits feel the need to
use formal institutional channels rather than extraconstitutional ones.
But even as it acknowledges the growing importance of formal insti-
tutions, such revisionism asserts that progress remains marginal. Thus
Goran Hyden observes that political leaders are not yet bound by consti-
tutional norms.4 Further, Richard Joseph sees systems of personal rule
as continuing to clash with formal institutions.5 And Larry Diamond

Journal of Democracy Volume 22, Number 1 January 2011

2011 National Endowment for Democracy and The Johns Hopkins University Press
Migai Akech 97

contends that the political struggle in Africa remains very much a con-
flict between the rule of law and the rule of the person.6
The flaw in such accounts is their failure to grasp the role that law
and its guardian, the judiciary, now play and have always played in Af-
rican politics. Precisely because Africas formal legal systems tend to
feature broad grants of poorly circumscribed discretionary powers, law
and legal processes often become important tools in political contests.
Indeed, the sheer breadth of formal power is what facilitates informal
and unaccountable uses of it. Hence it is time that students of African
politics began paying closer attention to the nature and uses of formal
laws and legal processes. Empirical studies of how formal and informal
institutions interact are particularly needed, in part to help us understand
what sorts of reforms will help citizens to take part in politically salient
legal processes and hold them to account.
Neopatrimonialism has become a synonym for informalism and per-
sonal rule, or the antithesis of the rule of law. Thus Larry Diamond ob-
serves that in African politics, the informal always trumps the formal.7
Extreme, if not cynical, versions of this perspective even suggest that
formal rules are essentially epiphenomenal and unable to alter the un-
derlying structural dynamics of African politics.8
Neopatrimonialism is also synonymous with presidentialism. This is
the dominant form of government in Africa, where it is characterized by
the extreme concentration of power in the president, often known as the
Big Man. The Big Man often stays in power until the end of his life,
distributes public-sector jobs and resources to his followers, and makes
little distinction between public and private funds.9 His lieutenants act
as patrons to lower-level power brokers. Politics becomes a matter of
clientelism, patronage, and corrupt, lawless, personal rule.10
While it cannot be denied that neopatrimonialism and extremely pow-
erful presidencies are prevalent in Africa, the existing literature largely
fails to account for the role that formal law has played in the emergence
and persistence of these phenomena. Neopatrimonialism in Africa is not
merely evidence of the absence or failure of law; it is enabled or facili-
tated by formal law. Likewise, the African imperial presidency is not
epiphenomenal; it is a creature of formal law.
Formal laws facilitate neopatrimonialism for the simple reason that
discretion is inevitable in any grant of power, whether formal or infor-
mal. Indeed, the inevitability of discretion in any governance arrange-
ment means that neopatrimonialism will be a characteristic of all po-
litical systemsif only to the extent that everywhere political power is
exercised via both formal institutions and informal (or personal) rela-
tions. It can therefore be expected that some level of informalism will
be prevalent, if not desirable, in any political system. After all, politics
is about negotiation. Ideally, however, such informalism and the nego-
tiations that go with it are kept transparent and open to public scrutiny.
98 Journal of Democracy

In established democracies, where this ideal is relatively well approxi-

mated, norms of transparency and accountability remain strong enough
to stop patrimonial institutions from dominating legal-rational ones, and
citizens enjoy fairly ample opportunity to question how political power
is being used. In African countries, by contrast, the legal-rational do-
main is not sufficiently transparent or accountable. This results in legal-
rational systemswith their already-broad grants of powerthat are
subject to penetration by patrimonial forces bent on self-serving rather
than public-regarding outcomes.
Examples abound of how African legal systems grant wide discre-
tionary powers. For example, statutory laws and regulations typically
grant officials broad powers without establishing effective procedural
mechanisms or limiting principles to circumscribe their exercise. In the
absence of effective regulation, the formal law therefore often aids the
practice and maintenance of neopatrimonialism.
Africas imperial presidencies are creatures of formal law. Soon after
independence, postcolonial elites set about carefully building up these
offices and their powers. Between 1960 and 1962 alone, thirteen newly
sovereign African states, beginning with Kwame Nkrumahs Ghana,
amended or replaced their independence constitutions in favor of new
rules of the game that centralized public power in a one-person presi-
dency.11 Such reconstitutions of the state were informed by an instru-
mental view of law that saw the primary purpose of the constitution as
facilitating state power, not controlling it. To the extent that the imperial
presidency is a creature of formal law, it should not be perceived as an
informal institution, which is how the conventional literature on neopat-
rimonialism often sees it.
Examining the ways in which law is used as an instrument in elec-
tions, state-security matters, media regulation, and control of the public
service will show that formal rules have always mattered in African
politics. Typically, those who wield political power straddle the divide
between patrimonialism and the realm of legal-rational norms, and will
base their actions on formal rules, informal considerations, or some
combination of the two, as expediency dictates.
For a long time, African political elites won or kept power through
extraconstitutional means, including violence and coups dtat. Increas-
ingly, however, they are capturing or retaining power through formal in-
stitutional channels such as elections. What explains this turn to formal
institutions? First, the democratization drives of the past two decades
have made citizens more politically potent and aware, leaving incum-
bents with fewer options. No longer able simply to ignore formal law,
incumbents must seek to manipulate it instead. Their turn to law is the
stuff of calculation, not conversion. Law has become the next political
frontier. Second, formal law may lend political decisions an air of legiti-
mate authority. If the law proclaims that a president has been validly
Migai Akech 99

elected or can serve a third term or can unilaterally name the members
of an electoral-oversight agency, then perhaps citizens will swallow the
pill thus handed to them.
Such calculations have probably figured in the decisions of presi-
dents who have sought third terms by
means of constitutional amendments.
In contemporary Africa, Some, such as Yoweri Museveni of
many nominally private Uganda, have succeeded. Others,
purveyors of physical- such as Nigerias Obasanjo and Zam-
bias Frederick Chiluba, have failed.
security services are joined
But even they have gone on to ma-
in complex networks nipulate formal rules in order to bar
involving formal and rivals or impose handpicked succes-
informal arrangements sors. In this way, Obasanjo stopped
alikewith the public his rival, Vice-President Abubakar
authorities. Atiku, from gaining the ruling par-
tys 2007 presidential nomination. In
this endeavor, Obasanjo received the
aid of the Independent National Electoral Commissiona body whose
members the president appoints by lawwhich issued politically mo-
tivated corruption indictments aimed at disqualifying Atiku and other
targeted candidates.12
In Zambia, President Chiluba disqualified his predecessor, Ken-
neth Kaunda, from contesting the presidency in 1996 by amending the
constitution to require Zambian ancestry of all presidential candidates.
Subsequently in 2001, when Chilubas efforts to amend the constitution
to allow a third term were thwarted, he nevertheless manipulated the
formal rules to guarantee victory for his handpicked successor, Levy
Mwanawasa. Thus voter registration, a process overseen by political ap-
pointees loyal to Chiluba, began suspiciously late in opposition strong-
holds. In Zambia, as in Nigeria a few years later, a sitting presidents
manipulation of formal rules had enabled him to shape the presidential
Kenyas bungled 2007 presidential elections can also be attributed to
presidential manipulation of the rules governing the electoral process.
Here, although an informal 1997 agreement of the so-called Inter-Par-
ties Parliamentary Group (IPPG) had stipulated that all major political
parties would thenceforth be represented on the Electoral Commission,
subsequent governments ignored the accord, arguing that it was not le-
gally binding. Thus President Mwai Kibaki opted to unilaterally appoint
members of the Electoral Commission in the months preceding the 2007
elections. As authority for this, he could cite the constitution, brushing
aside the IPPG agreement that he thought less likely to aid his quest for
a second term.
The instrumental use of the formal law is also prevalent in the state-
100 Journal of Democracy

security domain. In many African countries, regime maintenance domi-

nates the realm of public-security enforcement. This means using a strict
law-and-order approach and repressive statutes to hold on to power in
the face of competition from groups and factions that reject the regimes
claims to legitimacy. Invariably, postauthoritarian governments in Afri-
ca have retained and zealously enforced many existing repressive laws.
When Museveni took power in Uganda in 1986, for example, he refused
to repeal any of the repressive laws then in place, and indeed reinforced
them.13 It is arguable that the neoliberal privatization reforms that came
along with the democratization initiatives of the early 1990s have also
made regime maintenance easier. Thus in contemporary Africa, many
nominally private purveyors of physical-security services are joined
in complex networksinvolving formal and informal arrangements
alikewith the public authorities. In these circumstances, distinguish-
ing public from private security actors can be difficult.
In order to escape scrutiny, many African governments employ pri-
vate security operatives. Indeed, examples of joint ventures between po-
litical actors (including government ministers) and private security or
military companies abound.14 Thus in Angola, a major private security
company counts prominent public officials among its shareholders. In
Uganda, a politically influential army officer also owned a security firm
that was often seen as an extension of the army. In Nigeria, public secu-
rity forces, including the military, have been integrated into the security
arrangements of the oil industry to an extent that makes it difficult to
determine where public policing ends and private security provision be-
gins. Such hybrid arrangements are found in Sierra Leone as well.
Typically, the laws of these countries do not forbid such straddling
of the formal-versus-informal divide. In practice, the tactic has proven a
useful tool for regime maintenance. Many of these countries have set up
(admittedly imperfect) systems of accountability in order to help regu-
late the relationship between the uniformed military and civilian public-
security agencies, but have no such accountability frameworks to gov-
ern how their militaries relate to private security providers. As private
entities, such providers can and often do seek to shield themselves from
scrutiny by resorting to claims of privacy and confidentiality. The dan-
ger that regimes and their state-security agencies can use such private
actors to evade legal obligations and do dirty work is obvious.

Exploiting Legal Gaps

Among the regimes which have thus exploited gaps in the law is
that of Musevenis National Resistance Movement (NRM) in Uganda.
There the method is to use private security agencies as operational arms
and task forces of formal security structures, with officials prepared to
disown their private partners should public disapproval be roused. Simi-
Migai Akech 101

lar trends can also be observed in neighboring Kenya, where vigilante

groups have been touted as carrying out a form of community policing.
They often consult with actual police officers, and there is a percep-
tion that the government condones vigilantism whenever it is politically
expedientnot least because members of certain vigilante groups may
carry illegal firearms without fear of arrest. In both countries, it is evi-
dent that such groups succeed only thanks to the support that the formal
security apparatus gives them.
Media activities too have become targets of legal manipulation. Elites
naturally fear having their illicit activities exposed by a vigorous media,
and eagerly seek ways to curb it. One useful tool is the body of criminal
libel statutes left over from the colonial era. In Ghana and Kenya, jour-
nalists have found themselves prosecuted under such laws on charges of
defaming the government and influential public figures. Constitutional
guarantees of press freedom notwithstanding, the courts in Ghana have
upheld such laws as necessary to protect the dignity of public office.15
Finally, there is the legal manipulation of the public service. In many
African countries, independence constitutions that mandated a non-
political public service were changed in short order to give presidents
vast powers over state agencies. These, accordingly, have often become
entangled in corrupt dealings and schemes for keeping incumbents in
power. Presidential authority typically includes the right to constitute
and abolish offices, and to name or dismiss their holders at will.
Anticorruption campaigns stand little chance in the face of laws that
bind civil servants to obey all orders from above (even if merely verbal
and patently illegal). As a result, even well-meaning public employees
can find themselves forced to act as accomplices in grand corruption
schemes meant to finance regime-maintenance projects. In Kenya, for
example, public servants who refuse to go along are transferred to ir-
relevant departments or fired, often without due process. Thus the head
of the Central Bank of Kenyas tender committee was promoted to the
newly created Ministry of Development of Northern Kenya after object-
ing to a new currency issue on the grounds that it violated the public-
procurement law. When public complaints arose, the attorney-general
quickly issued a legal opinion pointing out that, under the constitution,
the tender-committee chair served purely at the presidents pleasure.
Likewise in Nigeria, the head of the Economic and Financial Crimes
Commission, who by all accounts was doing a fine job of tackling cor-
ruption and had brought to justice many untouchable figures, was sent
on study leave by the president when political winds shifted.
Longstanding official-secrets acts and more recent ethics laws aug-
ment presidential control over the civil service in many an African
country. Thus Kenyas Public Officer Ethics Act is proving a double-
edged sword. Its declared purpose is to improve ethics standards, but
it contains provisionsincluding stiff fines or even a jail term for
102 Journal of Democracy

divulg[ing] informationthat allow the president or his minions to

intimidate subordinates into silence. At the Kenyan central bank, for
instance, the tender-committee controversy was followed by the bank
governors decision to have the police launch a leak investigation under
the Ethics Act. A plausible interpretation is that lower-level officials
were being signaled to keep their mouths shut, or else.
As African regimes have watched their informal or extralegal options
growing narrower under pressure from more vigilant citizens and do-
nors, incumbents have found ways to work through formal institutions,
including courts. Kenya again provides a case in point. Not long after
his December 2002 election, President Mwai Kibaki began reshaping
the judiciary, citing corruption as a reason to sack numerous judges and
replace them with his own candidates. Because the Kenyan justice sys-
tem was poorly institutionalized until the promulgation of a new con-
stitution in August 2010there were no accountability or due-process
mechanisms for appointing or disciplining judges, for instancethe
courts were fairly easy to turn into instruments whose rulings advanced
regime objectives. These included Kibakis plans to stymie and then re-
configure a constitutional-revision process, inherited from the last days
of his predecessor Daniel arap Mois administration, that Kibaki feared
would dilute his power as president.

The Case for Administrative-Law Reform

It should by now be evident that the failure of democratization initia-
tives to tame African presidents is not merely a matter of rampant infor-
malism, but in fact has much to do with statutory laws that give these
chief executives and their subordinates vast discretionary powers. As
we have seen, these powers are instrumental to the regime-maintenance
endeavors of these presidents, who are adept at manipulating law. The
fundamental problem with African politics, therefore, is not that formal
rules are inconsequential or neglected. On the contrary, they are insuf-
ficiently institutionalizedat least in the sense that they are all too of-
ten open-ended and neither transparent nor accountable. They therefore
avail authoritarian or quasi-authoritarian regimes as tools with which to
subvert the progress of democracy and constitutionalism. The deliberate
laxity of the formal rules explains why leaders such as President Ab-
doulaye Wade of Senegal, who spent many years in opposition sharply
criticizing incumbents as undemocratic, begin to behave autocratically
once they gain power.16
This suggests that friends of democracy in Africa should worry
about how the operation of formal law can be made more participa-
tory, transparent and accountablein short, more power-constraining
than power-enabling. A key challenge is to transform the statutes that
give the legal order its imperial or authoritarian character so that this
Migai Akech 103

order conforms to the demands of constitutionalism. H. Kwasi Prempeh

takes the view that simply leaving these laws to the vagaries of consti-
tutional litigation and judicial reviewhoping that courts will strike
them down, in other wordsmay not be enough, especially given that
judicial review is a reactive institution.17 According to him, a more
effective approach would be to repeal by express provision in the text
(of the new constitution) all the repressive statutory laws upon which
authoritarian regimes rely. Although the adoption of such an approach
could help to tame the imperial presidency, I believe that the reform of
administrative lawa field often overlooked amid the dominant con-
cern with constitutional lawalso offers a useful avenue for democra-
tizing the exercise of power.
Administrative law, as its name suggests, regulates public adminis-
tration. On the one hand, this type of law reflects an appreciation of the
inevitability of discretion in the exercise of power, and empowers pub-
lic officials (or administrators) to implement government policies and
programs. On the other hand, it seeks to regulate the exercise of power
by requiring that all administrative actions meet certain requirements of
legality, reasonableness, and fairness. It performs this latter function by
setting out general principles and procedures that all administrators must
follow, and by providing remedies for people affected by administrative
decisions. In common-law jurisdictions, the courts have developed these
principles and procedures.
Nevertheless, countries such as the United States and South Africa
have codified them in statutes such as the federal Administrative Proce-
dure Act and the Promotion of Administrative Justice Act, respectively.
In South Africas case, the statute provides a mechanism for the realiza-
tion of a constitutional provision that confers on every person the right
to administrative action which is lawful, reasonable, and fair. Kenya
has followed South Africas lead, and its new constitution gives every
person a right to fair administrative action. The inclusion of a similar
provision in Zimbabwes draft constitution of 2007 suggests that the
idea of fair administrative action is also gaining acceptance in other
African countries.
Key principles of administrative law include the following require-
ments: decisions of administrators must be reasonable or justifiable; pri-
or to making major decisions, administrators must consult those likely
to be affected by them; decision-making processes must be free of real
or apparent bias; administrators must explain their decisions in writing;
administrators must not act arbitrarily or outside their powers; adminis-
trators must act in good faith; and there must be a right to judicial review
of administrators decisions. Another critical principle of administrative
law is to require checks and balances in decision making. For example,
in the context of criminal-law enforcement, administrative-law princi-
ples would decree a separation of functions between those officials who
104 Journal of Democracy

are responsible for conducting investigations, and those who are respon-
sible for prosecuting any offenses thereby uncovered.
Crucial procedures also include requirements that administrators must
give adequate notice of proposed ac-
tion to those likely to be affected by
Administrative law is their decisions, and give them rea-
a critical tool for the sonable opportunities to make rep-
creation of a limited gov- resentations. These procedures can
take the form of public inquiries or
ernment that does not rule
notice-and-comment procedures in
arbitrarily, but instead which the affected people are given,
respects the rule of law. say, thirty days to make comments
prior to the taking of a decision. Typ-
ically, such procedures are tailored to
suit the circumstances of the particular case. Through these procedures,
administrative law fosters participation by interested parties in the deci-
sion-making processes of government.
Administrative law is therefore instrumental to the realization of day-
to-day democracy, since it requires that decisions of government must
not only be subjected to checks and balances, but must also be explained
or justified to the people that they affect. In this way, administrative law
ensures that public officials do not abuse their powers, thereby under-
mining the liberties and livelihoods of citizens. Administrative law, in
short, is a critical tool for the creation of a limited government that does
not rule arbitrarily, but instead respects the rule of law.
The method of administrative law is to give those likely to be affected
by a governmental decision an opportunity to participate in its making,
or to contest it once it is made. For this to work in practice, members of
the public should be ready, willing, and able to utilize these opportuni-
ties. In that regard, it is encouraging to witness African civil societies
increasing vibrancy and readiness to call public officials to account.
Further, recent democratization initiativesespecially in the context of
constitutional reforms and such technological changes as the prolifera-
tion of mobile phoneshave made citizens more aware of civic mat-
ters. In significant ways, then, civil society actors and members of the
public are now better able than ever to make good use of the opportuni-
ties that the establishment of administrative-law regimes would create.
The publics capacity to engage officials, moreover, could benefit from
legal-empowerment initiatives, including training in legal literacy and
methods of finding and using legal assistance. Finally, disadvantaged
or marginalized groups would need to have their capacities enhanced in
order to be able to take advantage of the greater amounts of democratic
space that an administrative-law regime would create.
In particular, administrative-law reforms can stem the instrumental
use of law for political purposes. For example, an administrative-law re-
Migai Akech 105

gime would enhance the reasonableness and fairness of the decisions of

electoral bodies, ensuring that they are not used as instruments of regime
maintenance. In the security domain, administrative-law reforms would
enable citizens to hold accountable private security providers who are
integrated into public security networks. Private providers, the reason-
ing would go, are performing an essentially public function in this case
(that is, suppressing crime) and must therefore be held to normal public-
law obligations such as respecting the civil liberties of citizens. By sepa-
rating the function of bringing charges from that of following through
on prosecutions, administrative law would also ensure reasonableness
and fairness in the exercise of the prosecutorial power. This would, for
instance, prevent governing elites from abusing criminal statutes for the
purpose of silencing critical journalists such as those exposing corrup-
tion in government. And by requiring that administrators consult mem-
bers of the public before making decisionsincluding regulationsthat
are likely to affect the public, administrative law would enhance the
prospect that the exercise of power will display a higher regard for pub-
lic interests and concerns.
In these ways, administrative law would draw limits around the ex-
ercise of the broad swaths of power found in the statutory order. If the
framers of the U.S. Constitution deserve the thanks of posterity for hav-
ing adopted such a visionary basic law for their country, the later and
less famous figures who wrote the federal and state Administrative Pro-
cedure Acts deserve thanks of their own for giving U.S. citizens the
tools with which to scrutinize and hold the powers of government to
account on a day-to-day basis. The enactment of similar statutes in Afri-
can countries, and the legal empowerment of citizens to take advantage
of them, would contribute immensely toward taming the imperial presi-
dency. Administrative-law reform is thus critical to enhancing citizens
abilities and opportunities to challenge what remain all-too-common
abuses of power in African countries.


1. The first use of the term occurs in Shmuel N. Eisenstadt, Traditional Patrimonialism
and Modern Neopatrimonialism (Beverly Hills, Calif.: Sage, 1973).

2. Daniel N. Posner and Daniel J. Young, The Institutionalization of Political Power

in Africa, Journal of Democracy 18 (July 2007): 126 (emphasis added).

3. Posner and Young, Institutionalization of Political Power in Africa, 126.

4. Goran Hyden, African Politics in Comparative Perspective (New York: Cambridge

University Press, 2006), 111.

5. Richard Joseph, Progress and Retreat in Africa: Challenges of a Frontier Region,

Journal of Democracy 19 (April 2008): 95 (emphasis added).
106 Journal of Democracy

6. Larry Diamond, The Spirit of Democracy: The Struggle to Build Free Societies
Throughout the World (New York: Holt, 2008), 26061 (emphasis added).

7. Larry Diamond, Progress and Retreat in Africa: The Rule of Law versus the Big
Man, Journal of Democracy 19 (April 2008): 145 (emphasis added).

8. Nicolas van de Walle, The Democratization of Political Clientelism in Sub-Saharan

Africa, paper prepared for delivery at the Third European Conference on African Studies,
Leipzig, Germany, 47 June 2009, 2.

9. Michael Bratton and Nicolas van de Walle, Democratic Experiments in Africa: Re-
gime Transitions in Comparative Perspective (Cambridge: Cambridge University Press,
1997), 63, 66.

10. Diamond, Spirit of Democracy, 247 (emphasis added).

11. H. Kwasi Prempeh, Africas Constitutionalism Revival: False Start or New

Dawn? International Journal of Constitutional Law 5 (July 2007): 474.

12. Rotimi T. Suberu, Nigerias Muddled Elections, in Larry Diamond and Marc F.
Plattner, eds., Democratization in Africa: Progress and Retreat, 2nd ed. (Baltimore: Johns
Hopkins University Press, 2010), 124.

13. Andrew M. Mwenda, Personalizing Power in Uganda, in Diamond and Plattner,

Democratization in Africa, 238.

14. Migai Akech, Privatization and Democracy in East Africa: The Promise of Admin-
istrative Law (Nairobi: East African Educational Publishers, 2009), 97103.

15. H. Kwasi Prempeh, A New Jurisprudence for Africa, Journal of Democracy 10

(July 1999): 138.

16. Joseph, Challenges of a Frontier Region, 100.

17. Prempeh, Africas Constitutionalism Revival, 469.




This Report has been drafted by Darryl Biggar (Australian Competition and Consumer
Commission, ACCC) and Alberto Heimler (Italian Antitrust Authority and co-chair of
subgroup 1 of the AERS Working Group). The competition authorities of Brazil, Hungary,
Indonesia, Mexico, South Africa and South Korea provided very useful inputs. Comments and
suggestions were received from ICN members and from the following individuals: Ginevra
Bruzzone (Assonime), Frdric Jenny (Professor of economics at ESSEC and co-chair of
subgroup 1 of the AERS Working Group), Paul Wachtel (New York University) and, in
particular, Massimo Marchesi. The ICN AERS Working Group, subgroup 1, thanks very much
those that contributed.

1. Banking regulation originates from microeconomic concerns over the ability of bank
creditors (depositors) to monitor the risks originating on the lending side and from micro and
macroeconomic concerns over the stability of the banking system in the case of a bank crisis. In
addition to statutory and administrative regulatory provisions, the banking sector has been
subject to widespread informal regulation, i.e., the governments use of its discretion, outside
formalized legislation, to influence banking sector outcomes (for example, to bail out insolvent
banks, decide on bank mergers or maintain significant State ownership).

2. Banks in one form or another have been subject to the following non exhaustive list of
regulatory provisions: 1) restrictions on branching and new entry; 2) restrictions on pricing
(interest rate controls and other controls on prices or fees); 3) line-of-business restrictions and
regulations on ownership linkages among financial institutions; 4) restrictions on the portfolio
of assets that banks can hold (such as requirements to hold certain types of securities or
requirements and/or not to hold other securities, including requirements not to hold the control
of non financial companies); 5) compulsory deposit insurance (or informal deposit insurance, in
the form of an expectation that government will bail out depositors in the event of insolvency);
6) capital-adequacy requirements; 7) reserve requirements (requirements to hold a certain
quantity of the liabilities of the central bank); 8) requirements to direct credit to favored sectors
or enterprises (in the form of either formal rules, or informal government pressure); 9)
expectations that, in the event of difficulty, banks will receive assistance in the form of lender
of last resort; 10) special rules concerning mergers (not always subject to a competition
standard) or failing banks (e.g., liquidation, winding up, insolvency, composition or analogous
proceedings in the banking sector); 11) other rules affecting cooperation within the banking
sector (e.g., with respect to payment systems).

3. In recent years regulation in banking has become less pervasive and has shifted from
structural regulation to other more market oriented forms of regulation. As a consequence
competition has come to play a very important role in the allocation of credit and in the
improvement of financial services. The capital requirements framework created in the context of
the Basel committee paved the way to the development of stronger competition in banking. It is
unquestionable that all over the world banks now face greater competition both from new
entrants in the banking sector and from other financial companies.

4. Competition authorities have not been much involved in the process of liberalization of
banking. Moreover, in several countries the enforcement of antitrust rules until very recently has
not been applicable to banking because of sectoral exceptions.

5. In this light, the purpose of this report is:

to assist policy makers and enforcement authorities (in their competition advocacy
function) in their efforts to promote competition oriented regulatory reform in

to assist policy-makers and enforcement authorities (in their competition advocacy

function) in promoting an environment where competition law is fully applicable to
banking and where there is an appropriate institutional setting to that end; and

to assist competition enforcement authorities in the enforcement of competition law
in this sector, with a special emphasis on merger control.

6. The structure of the report is as follows. First, it briefly reviews the recent history of
banking regulation (section II). Second, it discusses (under the perspective of competition
authorities) the market failures banking are exposed to, their macroeconomic consequences
(section III), and the most common regulatory instruments introduced to address them (section
IV). Then, the report examines the impact of recent liberalizations on market power in banking
(section V). A brief description of banking issues in developing countries follows (section VI).
Finally, the report turns to competition issues, addressing first the application and scope of
competition law (section VII) and then examining issues of enforcement of competition law,
with a particular emphasis on merger control (section VIII). The final section concludes with a
number of recommendations.


7. In the early 70s financial systems were characterized by important restrictions on market
forces which included controls on the prices or quantities of business conducted by financial
institutions, restrictions on market access, and, in some cases, controls on the allocation of
finance amongst alternative borrowers. These regulatory restrictions served a number of social
and economic policy objectives of governments. Direct controls were used in many countries to
allocate finance to preferred industries during the post-war period; restrictions on market access
and competition were partly motivated by a concern for financial stability; protection of small
savers with limited financial knowledge was an important objective of controls on banks; and
controls on banks were frequently used as instruments of macroeconomic management.1

8. Since the mid 70s there has been a significant process of regulatory reform in the
financial systems of most countries. This process involved a shift towards more market-oriented
forms of regulation and involved partial or complete liberalization of the following:

interest rate controls

Until the early 1970s controls on borrowing and lending rates were pervasive in
most countries. These controls typically held both rates below their free-market
levels. As a result, banks rationed credit to privileged borrowers. By 1990 only a
handful of countries retained these controls.

quantitative investment restrictions on financial institutions

Investment restrictions on banks took a variety of forms, including requirements to

hold government securities, credit allocation rules, required lending to favored
institutions and controls on the total volume of credit expansion. Compulsory
holdings of government securities, as well as having a prudential justification, also
acted as a disguised form of taxation in that it allowed governments to keep
security yields artificially low. With some exceptions these controls were largely
eliminated by the early 1990s.

Edey and Hviding (1995), p4.

line-of-business restrictions and regulations on ownership linkages among
financial institutions

Although important line-of-business restrictions still remain in place in many

countries, the role of these restrictions has been significantly eroded or, in some
cases, entirely eliminated. For example, the separation of savings-and-loans and
commercial banks has been largely eliminated in many countries, as has the
distinction between long-term and short-term credit institutions in Italy and the
legal separation of various types of credit suppliers in Japan. Bank branching
restrictions were phased out in a number of European countries by the early 1990s.
In the US breaking down the barriers imposed by the (1933) Glass-Steagall Act
the Gramm-Leach-Bliley Financial Service Modernization Act of 1999 permits
banks, securities firms, and insurance companies to affiliate within a new structure
the financial holding company2.

restrictions on the entry of foreign financial institutions

There has been significant liberalization of cross-border access to foreign banks. In

particular, there are now in place a number of international agreements on trade in
banking services, including GATS, NAFTA and the EC. In particular, in the
European Union, the second banking directive (89/646/EEC) forbade the
obligation for banks established in one Member State to seek authorization from
other Member States when they intended to establish a branch in their territory. In
many countries however the entry of foreign banks is still made more difficult than
that of domestic ones.

controls on international capital movements and foreign exchange transactions

Liberalization of controls on capital movements is now virtually complete in

OECD countries and in many developing countries as well3. Some controls remain
on long-term capital movements, particularly with respect to foreign ownership of
real estate and foreign direct investment. There also remain important restrictions
on international portfolio diversification by pension and insurance funds.

The origins of regulatory reform

9. Regulatory reform was driven by a number of inter-related factors, including:

the diminishing effectiveness of traditional controls due to financial innovation

(including the difficulty of isolating domestic markets) and rapid technological

the development of various types of regulatory avoidance (such as the development

of offshore financial centers and off-balance-sheet methods of financing);

competition between international financial centers;

See Crockett et al. (2003)
The beneficial effects of capital movements liberalization for developing countries are still controversial.

competition with non banks for many services (consumer credit; small business
loans; mortgages; etc.);

competition between financial institutions under different regulatory environments;

and, finally,

multilateral agreements liberalizing cross-border banking activities.

Benefits from regulatory reform

10. Regulatory reform has raised efficiency and lowered costs in the financial services sector:
First, the removal of regulatory restrictions gave financial firms more freedom to
adopt the most efficient practices and to develop new products and services.

Second, regulatory reform increased the role of competition, which in turn spurred
reductions in margins in financial services and raised efficiency by forcing the exit
or consolidation of relatively inefficient firms and by encouraging innovation4.

11. Regulatory reform furthermore contributed to:

declining relative prices for financial services and productivity growth well in
excess of that for the economy as a whole5;

considerable improvements in the quality, variety and access to new financial

instruments and services;

improved world allocation of resources due to the removal of the barriers to

international capital flows;

significant improvements in growth performance in a number of developing


Regulation has been maintained but has progressively been reformed

12. The progressive liberalization from structural regulatory restrictions such as the ones
mentioned above has not led to the deregulation of banking activity, but to the adoption of new
instruments of prudential regulation more compatible with competition in the banking sector.
The first and most known milestone of this new trend in regulation is the Basel Accord of July
1988 which required the major international banks in a group of 12 countries to attain an 8%
ratio between capital and risk-weighted assets from the beginning of 1992.

13. Subsequently, the increasing range and sophistication of financial instruments made the
limitations of the probably too simple design of the 1988 capital-adequacy framework become
apparent. Already in 1997 the Basel Committee on Banking Supervision, seeking to further
enhance banking supervision in both G10 countries and a number of emerging economies,
released a set of Core Principles which set out minimum requirements for banking

OECD (1997a), p83.
Estimates indicate that overall financial service sector productivity increased at an annual rate of nearly 4 per cent in the US over
1980-93, nearly three times the rate of the economy as a whole. OECD (1997a), p. 84.
See among others and more recently Claessens amd Laeven (2005).

supervision. The document also sets out an extensive list of recommended powers of banking
supervisory authorities.

14. Finally, in June 2004 the Basel Committee on Banking Supervision issued a revised
framework (Basel 2) for measuring capital adequacy and for identifying new minimum capital
requirements for banks (Pillar 1). The new framework encourages banks to develop their own
in-house risk-management systems to compute in a much more precise and sophisticated way
their minimum capital requirements , with supervisory oversight present in the endorsement of
the adequacy of the system. The proposals of the Committee, expected to be progressively
implemented from the end of 2006, also introduce two additional pillars for banking regulation
that are expected to become more and more important in complementing capital adequacy
requirements. Pillar 2 introduces a continuous dialogue between banks and their supervisor in
order to follow and accommodate changing and evolving business practices. Pillar 3 calls for
improving the flow of information to the public on banks financial conditions, so that market
discipline can exercise a greater role in reducing excessive risks in banking activities.



15. It is widely accepted that in the absence of market failures, open and competitive markets
yield strong incentives to efficiently meet the demands of consumers and to adapt to changing
demands and technology over time. With very few exceptions, in the absence of a market failure
there is no economic justification for regulation.

16. The most important rationale for regulation in banking is to address concerns over the
safety and stability of financial institutions, the financial sector as a whole, or the payments
system. The description and the evaluation that follows necessarily reflect the views of
competition authorities. With only one exception, no bank regulator has reviewed this Report
which, therefore, does not necessarily reflect the positions and the opinions of bank regulators.

The risk of bank runs

17. All banks operate in conditions of fractional liquidity reserve. The great majority of banks
liabilities are very liquid deposits redeemable on demand. The great majority of their assets are
instead much more illiquid loans. This situation leads to the problem that if all depositors
demanded their deposits back at the same time, any bank (even if perfectly solvent) would face
serious problems in meeting its obligations vis vis its depositors. A single bank might obtain
refinancing on the financial market but the problem would severely persist in cases of low
liquidity on the market or if the issue concerned a big portion of the banking sector.

18. It is well known in the literature that whenever depositors start fearing the insolvency of
their bank, their first most common reaction is to go and withdraw their deposits creating
serious problems to the banks. Such behavior is normally referred to as a bank runI7.

There are two alternative theories that have been proposed for explaining bank runs. Some authors, for example Diamond and
Dybvig (1983), consider bank runs as a sunspot phenomenon, unrelated to any underlying economic variables. Others, for example
Bryant (1980), suggest that bank runs originate from some negative information (either right or wrong) depositors have on the
quality of the assets of their bank.

The risk of excessive risk taking (moral hazard) in banking

19. Banks grant loans normally financed by the deposits they received. This is by itself a
powerful incentive for banks to grant credit in a not sufficiently prudent way and to take in too
much risk. In fact it is well known in the literature that with debt financing, while the risk of
failure of the financed investment is mostly carried out by the bank depositors, in the case of
success profits accrue mostly to the bank8..A good example of this deviating behavior is the
Asian financial crisis of 1997 that is mentioned further below. In general, however, this
incentive is somehow mitigated by the possibility that the market, both via depositors and other
banks, could monitor the risks assumed by the banks management.

20. The main purpose of regulation is to avoid the highly negative consequences for the
economy of widespread bank failures. There are two main strands of arguments for banking
regulation. The first focuses on the systemic dangers of bank failures, while the second on the
need for security and stability in the payments system.

Systemic dangers of a bank failure

21. The main argument for bank regulation focuses on the possibility of systemic or system-
wide consequences of a bank failure. i.e., the possibility that the failure of one institution could
lead to the failure of others. This argument is summarized by Feldstein as follows:

The banking system as a whole is a public good that benefits the nation over and above
the profits that is earns for the banks shareholders. Systemic risks to the banking system
are risks for the nation as a whole. Although the management and shareholders of
individual institutions are, of course, eager to protect the solvency of their own
institutions, they do not adequately take into account the adverse effects to the nation of
systemic failure. Banks left to themselves will accept more risk than is optimal from a
systemic point of view. That is the basic case for government regulation of banking
activity and the establishment of capital requirements.9

22. It is possible to distinguish two mechanisms by which the failure of one bank could lead
to the failure of other banks or other non-bank firms:

(a) the failure of one bank leading to a decline in the value of the assets sufficient to
induce the failure of another bank (consequent failure) and

(b) the failure of one bank leading to the failure of another fully solvent bank, through
some contagion mechanism (contagion failure)

Consequent failure

23. The failure of a bank, like the failure of any other firm in the economy, may, of course,
lead to the failure of other firms exposed towards the failing bank. The loss of value associated
with the failure leads to a reduction in the value of assets held by its stakeholders. If this loss in
value is sufficiently large and/or the stakeholders themselves are near enough to failure, the
stakeholders may, in turn, fail.10

See Dewatripont and Tirole (1994).
Feldstein (1991),
See Kaufman (1996) , p. 25.

29. Although there are, in general, strong incentives to diversify, in the case of a large firm
there may be a number of other firms (such as its suppliers) who are unable to diversify
adequately and whose survival is threatened by the large firms insolvency. However, in
general, banks are able to diversify. They are not constrained to retain their assets with a bank
that is in difficulty. The decision to invest in a distressed bank is a risk-management decision
like all other investment decisions. Provided the investing bank is aware of the risk it is taking,
there is no externality. The externality can however originate from the fact that full information
on potential risk is not immediately, correctly and easily achievable.

Contagion failure

24. A majority of authors argue that there is an important asymmetry between the information
available to banks and the information available to depositors and other outside investors.
Banks can utilize economies of scale and specialization to reduce the transactions costs of
determining the probability that a borrower will not repay a loan as promised, to monitor the
borrowers performance and circumstances, and to take effective actions to reduce the
probability and cost of defaults. Thus banks have information about the value of loans that
depositors and other outside investors do not have..11

25. In the most extreme case of this information asymmetry, depositors cannot distinguish
solvent from insolvent banks. As a result, news that one institution is failing can be interpreted
as information that other institutions are in difficulty. Depositors rush to make withdrawals
from solvent as well as insolvent banks since they cannot distinguish between them.12 It is
possible that the failure of one institution may lead to a generic flight of funds from all
institutions. The available evidence does not always suggests that this has happened.

Dangers to the soundness of the payments system

26. We turn now to the arguments relating to the stability and soundness of the payments
system. These arguments are summarised in the following comment:

An efficient payments system, in which transferability of claims is effected in full and

on time, is a prerequisite for an efficient macroeconomy. Disruptions in the payments
system carry the risk of resulting in significant disruptions in aggregate economic
activity. To some observers, instability in the payments system is more threatening than
instability in deposits. This fear appears to reflect the larger dollar volume of daily
payments, the speedy movement of the funds and the unfamiliarity of the clearing
process. 13

27. Until recently the standard form of settlement between banks was end-of-day net
settlement. Banks would accumulate their obligations to other banks throughout the day in order
to settle the smaller net obligations at the end of the trading day. The risk of this form of
settlement is that it usually requires participants to grant unsecured and unlimited credit to other
participants during the day until final settlement occurs. Credit extended to a single party can
sometimes exceed a banks entire capital. Like other forms of credit, the potential exists for
default. If the expected payment to the bank extending credit does not materialize in full and on

Benston and Kaufman (1995), p216.
Mishkin (1991),.
Benston and Kaufman (1995), p37.

a timely basis, previous payments may need to be reversed or unwound. This may be complex
and time-consuming and cause gridlock in the payments system that interrupts the smooth
flow of trade. Moreover, if the losses to the paying bank from customer defaults were large
enough to drive it into insolvency, the receiving banks would experience losses, which might be
sufficient to drive them to insolvency if these losses exceed their capital.14

28. An obvious candidate solution to this kind of problems is to prevent the intraday build-up
of credit exposures by insisting that inter-bank payments occur at the same time as the exchange
of the corresponding assets. This is known as real-time settlement. Such real-time trading
systems have already been implemented in some countries.

29. In the case of international transactions, the problem of intra-day exposure is however
somewhat more complicated. The problem arises because of the different timing for daily
settlement in each national bank system. For example, in a NZ Dollar/US Dollar foreign
exchange transaction, the NZ$ leg must be settled even before the US system for settling the
US$ leg is open for the day. This gives rise to what is known as Herstatt risk, named after the
failure of a small German bank, Bankhaus Herstatt in 1974. This bank was active in the foreign
exchange markets. It defaulted after receiving deutschmarks from international banks but before
the matching US dollar leg was processed later in the day. This left its counterparties exposed to
the full value of the Deutsche marks delivered. This event severely disrupted CHIPS, the main
clearing system for US dollars, led to a collapse in trading in the US dollar/deutsche mark
market and even resulted in disorder in the inter-bank money markets. This problem is widely
recognised and is a focus of attention of central banks around the world.


30. This section of the paper provides a description of the most standard instruments of bank
regulation: deposit insurance, capital adequacy requirements and lender of last resort. These
three policies are linked one with the other. Deposit insurance protects the smallest depositors
from a bank bankruptcy and prevents bank runs. Capital adequacy requirements are necessary in
order to make sure that bank managers follow a responsible credit policy, in the absence of an
effective control on the part of depositors. Lender of last resort policies further reduce the risk
of banks bankruptcies providing banks with Emergency Liquidity Assistance facilities that are
designed to avoid that temporary situations of illiquidity lead to the insolvency of the bank.

Deposit insurance

31. Deposit insurance is a guarantee that all or part of a depositors debt with a bank will be
honored in the event of bankruptcy. The specific form of insurance schemes can vary in a
number of ways, including the fee structure (flat fee versus variable, risk-related fees); the
degree of coverage (full versus partial coverage, maximum limits); funding provisions (funded
versus unfunded systems); public versus private solutions; compulsory versus voluntary

32. Deposit insurance reduces (and in most cases eliminates entirely) the incentive to run
on the bank in the event of financial difficulty. Therefore it reduces the possibility that a

Benston and Kaufman (1995), p37.

temporary situation of illiquidity and rumors on the insolvency of the bank actually lead to the
failure of the bank. Furthermore, deposit insurance prevents the chain reaction that can also be
started associated by the run on a single bank, so that it reduces the possibility of contagion in
the banking system.

33. A drawback of its introduction is however the fact itself that from the point of view of the
depositor, deposit insurance makes all banks equally attractive. It almost completely removes
the incentive on the depositor to determine the risk of a bank and the need for the bank to
compensate the depositor for bearing bank-specific risk by including a bank-specific risk
premium in the interest paid to the depositor. Similarly, the depositor faces little incentive to
diversify her portfolio of assets held in banks.15

34. The effect of deposit insurance on the incentives of the bank depends upon the nature of
the insurance contract (and also on any other complementary regulatory measures). In
particular, the effect of the deposit insurance on the bank will depend on whether or not the
insurance premium paid by the bank depends on the individual banks risk.

35. In the case where the premium is completely unrelated to the risk of a particular bank
(i.e., the fixed fee system), there is clearly an incentive for the bank to attempt to increase its
profits by either increasing its revenues (by lending to higher return but riskier projects) or by
reducing its costs (by reducing its reserves). Both actions increase its risk. This is the well-
known moral hazard problem of deposit insurance. Fixed fee deposit insurance creates
incentives for banks to take on more risk in their operations than they would without deposit
insurance. This effect was apparent almost as soon as deposit insurance was adopted in the
1930s, when bank capital ratios dropped from 15% to around 6%16.

36. Deposit insurance, especially if extended to all deposits, by reducing the market
incentives for prudent management, may have the perverse incentive of making banks riskier.17
When this moral hazard extends across all financial institutions, the macroeconomic
consequences can be very significant. .

37. The problem of moral hazard and the need for additional regulatory measures can be
reduced if the insurance premium is related to the risk of the insured bank. An efficiently
organized insurer would graduate insurance premia according to the risk of the banks asset
portfolio and the adequacy of its capital holdings. Such a system would minimize the danger of

At least as long as the size of the deposit is less than any ceiling on the amount per deposit insured (100,000 dollars in the US).
Note however how this ceiling is only 20,000 Euros in the EU exactly in the attempt not to exacerbate such problem.
Parry (1992) , p 14. The consequences of the moral hazard can be clearly seen in the S&L crisis in the US of the early 1980s.
In the case of S&Ls the insurance premium was set by statute in 1950 at 1/12 of 1% of the assessable deposits and was the same for
all insured institutions regardless of the riskiness of their assets or the size of their equity capital or the capability of their
management. The holder of an insured account had no reason to be concerned with the safety or soundness of the particular
institution in which he had invested, or to require a higher return commensurate with higher risk. ... From the standpoint of the
management and owners of an insured S&L, this system created a constant inducement to take added risks with their expected
higher returns, depositors would not demand higher interest and the FSLIC could not raise its premium in response. Scott (1989),
Ironically, the introduction of government regulations and institutions in the US intended to provide protection against the
fragility of banks appears to have unintentionally increased both the fragility of banks and their breakage rate. By providing a poorly
designed and mis-priced safety net under banks for depositors, first through the Federal Reserves discount window lender of last
resort facilities in 1914, and then reinforced by the FDICs deposit guarantees in 1934, market discipline on banks was reduced
substantially. As a result, the banks were permitted, if not encouraged, to increase their risk exposures both in their asset and liability
portfolios and by reducing their capital ratios. . T(t)his represents a classic and predictable moral hazard behavior response.
Kaufman (1996), p22.

adverse incentive effects ... Under such a system, the individual bank bears the consequences of
a higher risk portfolio or a lower capital-deposit ratio, in the form of a higher insurance fee.18

Capital adequacy requirements

38. One regulation which exists in most countries is some form of capital adequacy
requirement. Capital adequacy requirements can take a variety of forms. Most countries know
a minimum level of required capital (an absolute amount). Beyond that, many countries require
the maintenance of some capital - or solvency - ratio; that is, a minimum ratio between capital
and an overall balance sheet magnitude, such as total assets or liability, or some weighted
measure of risk assets.19

39. However, capital-adequacy requirements do have certain difficulties:

(a) First, it is difficult to design capital-adequacy requirements in a sufficiently

sophisticated way. For example even though the 1988 Basel rules on capital
adequacy for banks categorizes assets and assigned a risk-weighting inevitably
differences in risk were overlooked between individual assets. One consequence
was that banks tended to search for the most risky assets within a risk class,
encouraging banks to go up the yield curve in pursuit of a return on capital.20 In
effect, the moral hazard problem re-emerged within the constraints of each
regulatory risk class.

(b) A particular problem can arise with inter-bank lending. If inter-bank lending is
treated favorably for capital-adequacy purposes in order to promote the liquidity on
the market, banks may, perversely, be given incentives to lend to other banks in
difficulty, increasing the risk of contagion and removing one of the more important
disciplines on bank risk-taking.

(c) Third, with technological advances, innovation in financial products is rapid.

Regulations, in contrast, might be changed not sufficiently frequently and only
catch up with current developments.

(d) Fourth, in some cases the adoption of new financial products is hindered by
lagging regulatory developments, delaying and stifling the pace of innovation.

40. Partly as a result of an increasing recognition of these problems, the Basel Accord was
modified in 2004 introducing more sophisticated ways of computing capital requirements and
increasing the focus on risk-management policies and systems in banks. In particular the new
regulation, which will start to be implemented from the end of 2006, encourages banks to
develop, with supervisory oversight, their own systems to compute minimum capital
requirements. Furthermore Basel 2, by improving the flow of information to supervisors and the
public on banks financial conditions, assigns a greater role to supervisory and market oversight
in reducing excessive risks in banking activities.

Baltensperger (1989), p8.
Baltensperger (1989), p13.
Charles Dallara, Chief Executive of the Institute of International Finance, reported in Financial Times, Wednesday, November 19,

Lender of last resort

41. In most countries the central bank or the government have an explicit (or implicit) policy
of providing assistance to banks facing financial difficulties

42. These lender of last resort interventions should be strictly limited to illiquid banks, easing
only very temporary liquidity problems faced by banks (Emergency Liquidity Assistance), not
extending also to help insolvent banks. In fact, whenever the lender of last resort assists
insolvent banks, its intervention has the same consequences of a flat-rate unfunded deposit
insurance, giving banks a strong incentive to adopt a riskier position than otherwise.21 As with
deposit insurance, when such incentives extend across the financial system, the macroeconomic
consequences can be severe.

Moral hazard and the Asian financial crisis

In the mid 1990s, several countries in South-East Asia experienced a severe currency and financial crisis,
on a scale that was almost entirely unforeseen, involving collapses in domestic asset markets, widespread
bank failures, bankruptcies on the part of many firms and a very severe economic downturn.

The crisis represents something of a puzzle for macroeconomists. None of the fundamentals that drive
traditional currency crises seem to have been present in any of the afflicted Asian economies. On the eve of
the crisis all of the governments were more or less in fiscal balance; nor were they engaged in irresponsible
credit creation or runaway monetary expansion.

In a paper written at the bulk of the crisis, Paul Krugman attempts to explain this puzzle, focusing on
problems with bank (non) regulation in these countries. He argues that a key common feature was that the
liabilities of financial intermediaries in these countries were perceived as having an implicit government
guarantee, but that the financial institutions themselves were essentially unregulated and therefore subject
to severe moral hazard problems.

To be sure, the government guarantees were not explicit. However, press reports do suggest that most of
those who provided Thai finance companies, South Korean banks, and so on with funds believed that they
would be protected from risk - an impression reinforced by the strong political connections of the owners
of most such institutions. In practice, moreover, these beliefs seem to have been for the most part validated
by experience.

In the presence of government guarantees and a complete absence of prudential regulation, Krugman
shows that banks have an incentive to continue lending as long as there remains any possibility at all that
the lending will yield a positive return. This has the effect of bidding up asset prices to the point where
they reflect their highest possible return, which can be several times higher than prices in an efficient
market. The inflated value of assets improves that apparent financial position of the financial institutions,
permitting more lending, and so on.

Krugman argues that a widespread perceived risk that government would decide to abandon the implicit
debt guarantees is sufficient to lead to a financial crisis in which plunging asset prices undermine banks,
and the collapse of the banks in turn ratifies the drop in asset prices. The self-fulfilling prophecy
component of this story can help explain why an asset value down-turn in one country can rapidly spread

It must however be said that it may be very difficult in practice to immediately distinguish an illiquid from an insolvent bank.

to others, in what is traditionally been called contagion. The moral of the story is either to impose
stringent prudential regulatory controls or abolish the government guarantees.


43. On the credit side competition between banks has led to lower spreads and greater care in
financing sound projects. Claessens and Laeven (2005) write:

More competitive banking systems are better in providing financing to financially

dependent firms. . There is support for the view that more competition may reduce hold
up problems and lower the cost of financial intermediation, making financially dependent
firms more willing to seek (and more able to obtain) external financing

Furthermore in most countries, including developing ones, recent market developments have led
to strong rivalry by non bank financial institutions for the supply of some banking services, for
example consumer credit or factoring services to small and medium size firms. This implies that
banks market power is somehow disciplined also by non banks.

Ensuring that banks are properly informed of the debt exposure of potential borrowers

44. Especially in developing countries, however, competition among banks may be impaired
because information on the credit worthiness of potential borrowers is not readily available.
Without a proper supplier of information on borrowers credit worthiness, each single bank has
an informational advantage over any other bank on the credit worthiness of its customers. New
banks will be very reluctant to lend to customers of other banks, if they are not fully and readily
informed on the total debt exposure of each potential borrower. A competitive financial market,
where banks compete for customers and potential borrowers choose among alternative banks as
suppliers of funds, can only develop if banks are fully informed on the total exposure of each
customer. Otherwise, if information is privately held by each bank, the market for credit will be
segmented and banks will only lend to customers they personally know.

45. Relationship banking is particularly efficient when firms are small and accounting rules
are not very effective. On the other hand a marked based system is particularly effective when
firms are relatively large and accounting statements transparent. Moreover, limitations on
competition in a relationship-based system do not just give the financier (market) power, but
also strengthen his incentive to cooperate with the borrower22. This implies that a relationship-
based system tends to smooth firm specific shocks intertemporally, while an arms length
system is much less able to provide such contingent insurance. On the other hand relationship-
based systems, because of the illiquidity of the financed assets, have an incentive to increase
financial risk more than arms length systems. Market based financing permits more
flexibility in explicit contracts, which allows the system to absorb adverse shocks. Moreover the
healthy can be distinguished from the terminally ill after a shock and can be dealt with
differently not everyone has to sink or swim together as in the relationship system23.

See Rajan and Zingales (2003) p. 12.
See Rajan and Zingales (2003) p. 19.

46. Relationship banking does not imply that potential borrowers do not have but one choice
with respect to the bank that would assist them. There can be strong competition among banks
also with relationship banking. In fact, in some countries, where the banking industry is
sufficiently competitive and the industrial sector is sufficiently developed, each local bank may
be willing to invest in order to develop a credit relationship with each local firm.

47. In many ways the two systems (arms-length and relationship banking) coexist in the same
economy. Regulators should therefore not impose or favor one system over the other and should
introduce regulatory provisions that are as much as possible neutral with respect to the type of
relationship between banks and their creditors.. Regulators should therefore maintain a
centralized system of monitoring the full exposure of different firms with respect to the banking
system, and more in general with respect to the financial sector at large, requiring all financial
institutions to communicate to the regulator all loans granted to a given (consolidated) borrower
and their degree of utilization. The increase in transparency that such a system of centralized
monitoring of debt exposure would provide, may help the development of arms-length
financing, and in any case reduce the market power of each bank with respect to its customers.

48. Antitrust authorities should use their advocacy powers to ask for such centralized reporting
of debt exposure to be undertaken. Their role can be very important because they would advice
on how to collect the information centrally without, at the same time, promoting collusion
among market players.

Regulatory reform, competition and depositors switching costs

49. While, in many countries banks benefited from the new opportunities originating from
regulatory reform by offering new and improved financial services to customers, switching
costs for consumers remained quite high, so that competition between banks did not increase
proportionately. There is now substantial evidence that the widening range of services offered
by banks was not associated with a significant increase in the elasticity of each bank residual
demand (as should have been expected because of greater competition). The effect of
liberalization on the market power of banks with respect to customers of banking services was
probably not too strong.

50. In recent decades, besides the traditional deposit-taking banks have entered quite a
number of new related markets, such as (among others):

Credit cards services, paying bills for depositors

Consumer loans
Life insurance
Financial consulting; Management of investment funds; Asset management

By providing all these services under one roof, banks reduce the transaction costs depositors
would have faced had they been obliged to negotiate for receiving these services with a number
of different providers. At the same time, however, by offering all these services, banks have
made it more costly for depositors to switch bank. In fact should depositors decide to move to a
new bank they would need to: 1) receive new credit cards (with a different number and expiry
date) that would need to be communicated to any service provider, for example the cable TV
company, should its bills being paid by credit card; 2) inform the new bank about all utilities
whose bills were being paid by debiting the depositor checking account; 3) transfer the deposit

of all purchased stocks or bonds to the new bank; 4) maintain the checking account of the old
bank just to service the mortgage; 5) communicate to all correspondents the new banking
coordinates. The increase in switching costs tends to make steeper the residual demand curve
each bank faces, so, even though competition may be increased in each of the markets where the
bank expanded, the overall market power of each bank is increased, at least with respect to
existing depositors. Or, to say it differently, in order for a bank to convince depositors of
another bank to switch, the improvements in the quality of services it offers must be much
larger than it would be the case in the absence of switching costs.

51. Depositors may also face switching costs because of strategic behavior on the part of
banks. For example while opening a checking account may be free, banks may require that a
high fee be paid when closing an account. There are good reasons why a policy of charging for
closing an account would be followed by all banks and would not be competed away: Each
bank benefits by market segmentation and no bank benefits by unilaterally reducing exit costs.

52. This is why it is unlikely that banks would engage autonomously in switching costs
reducing activities, given that this would imply reducing profits for each bank and also for the
industry as a whole. Pro-competitive rules and regulations may contribute to make switching
easier, so as to ensure that all the benefits originating from greater competition actually reach

53. Regulation could impose on all banks disclosure rules with respect to all the costs
involved in switching, so that consumers are made aware of these costs and competition among
banks may indeed prove to be very useful.

54. With the advent of the internet, banking is no longer necessarily a local industry, not even
for the smallest depositor, at least in countries with widespread internet literacy. Since banking
technology is the same across the world it is extremely important that regulation does not limit
the extent of the market with unjustified restrictions. This is particularly important in
jurisdictions that use the same currency. For example, the introduction of the Euro in 2002
could have made depositors indifferent as to the nationality of the bank where they would
deposit their savings, leading to a very significant enlargement of consumer choices and of
competition. Notwithstanding the regulatory interventions in such directions, such as with
regulation (EC) 2560/2001 on cross-border payments in the Euro area, the high costs
traditionally associated with dealing with foreign banks have remained. As a consequence, the
residual demand of a bank localized in one country remained substantially equal to what it was
before the Euro, while the removal of the higher costs associated with cross border transactions
would have probably led to a significant increase of the elasticity of its residual demand.

55. Antitrust authorities should use their advocacy powers to push forward the pro-consumer


56. Cross country comparisons show the importance of a well developed banking sector for
achieving both long term economic growth and the reduction of poverty. Countries with better
developed banking systems and capital markets have shown higher growth rates24. However the

See World Bank (2001)

direction of causality is not always clear. In particular, need property rights and contract laws be
firmly in place before a viable financial sector is developed? Is the modernization of the
banking sector a prerequisite for economic growth or is the other way round? What is the role of
the public sector in the financing of development? This section will try to provide the
competition authorities view, drawing on the existing literature and on the responses to a
questionnaire delivered to six countries: Brazil, Hungary, Indonesia, Mexico, South Africa,
South Korea.

57. Finance is always necessary for growth. In particular ongoing business need finance for
operation and for expansion. The same is true for launching new business enterprises.
Households need to have safe deposits, access to the payment system, to mortgages and
consumer loans. In this respect the experience of many developing countries show that the
banking sector is generally responding well to the needs of the wealthy households and of the
established firms. More in general, banking seems to develop well with firms and people that
are able to offer a collateral or have formal employment so as to provide some guarantee with
respect to future income, less well with people and firms that are unable to offer guarantees.
However, while in developed countries this second group of customers is relatively small, in
developing countries it represents the majority, so that banks tend to provide services only to the
minority of the population. In banking, while the competitive solution with little regulation is
appropriate for these existing banks so as to eliminate distortions, favoritism and high interest
rate spreads. As ana example, the Pakistani competition Authority in its submission to the
OECD Global Forum on Competition in February 2005 writes:
The financial sector was deregulated and with the economic liberalization, new
banks, financial institutions, leasing companies, housing finance, investment companies
and foreign banks have come up, which has created a competitive milieu25.

58. Regulatory reform and competition are able to expand the reach of banking to the
underprivileged. On the one hand, especially in countries where the majority of potential
borrowers do not have a collateral to offer, conventional banking may lead to a non optimal
equilibrium, where quite a number of low risk project are not financed and high-risk borrowers
end up having to pay higher interest payments. On the other hand technical progress and
flexible regulation have made it possible to provide banking services also to the poor. For
example Dymski (2003) writes:
Lemon Bank (a microcredit bank) offers credit and debit cards and savings accounts
to the unbanked. Its minimum amount are tiny, and checking services are available
without annual fees. Lemon Bank, which has 3600 access points, many in favelas and
in drugstores, is about to launch a media campaign aimed at opening 100,000 new
accounts by years end.

59. As for the lending side, in recent years in many developing countries specialized lending
institutions started to use unconventional methods to lend successfully to the poor, starting what
is now known as microcredit. Considerable evidence shows that such unconventional lenders
were able to lend to borrowers that no conventional borrower was willing to attract and
nonetheless performed much better, in terms of financial self sufficiency and repayment rates,
than would conventional banks in comparable loans. The reason of this success, that is not
limited to the Grameen Bank in Bangladesh, is the use of unconventional methods of risk
reduction: forming groups of borrowers that are jointly responsible for each others loans (joint
liability) and intense monitoring of clients, relying heavily on the promise of repeating the loan.

See OECD (2005)

60. A recent World Bank report on rural financial services26, comparing the competitive low
interest rates that microcredit offers with the regulatory solution of subsidized low interest rate,
concludes that the competitive solution of allowing microcredit institutions to develop is far
superior. Indeed subsidized credit leads to excess demand and the decision on which firm to
lend does not depend so much on the relative profitability of the underlying project, but mainly
on other considerations (political connections, corruption etc.). The World Bank report outlines
in the following table, the cost and benefits of the old and the new paradigm:

Table 1 Primary features of the old and new paradigms in rural finance

Features Directed Ag. Credit Paradigm Financial Systems Paradigm

1. Chief aims Boost agricultural production Reduce market imperfections and

Reduce poverty Transaction costs for income
expansion and poverty reduction
2. Role of financial markets Help the poor Intermediate efficiently
Stimulate production
3. View of users Beneficiaries: borrowers Clients: borrowers and depositors
4. Subsidies Heavily subsidy dependent Increasingly independent of subsidies
5. Sources of funds Vertical: governments and donors Horizontal: primarily voluntary
6. Associated Dense, fragmented, and vertical Less dense and mainly horizontal
information systems
7. Sustainability Largely ignored Major concern
8. Outreach Mostly ignored Primary concern
9. Evaluations Credit impact on beneficiaries Performance of financial institutions
mainly primary data
Mostly secondary information

Source: World Bank (2003)

61. In the past decade many micro-credit supplying institutions, which originally were State
owned and loss making, were progressively privatised and deregulated, increasing both their
efficiency and their profitability. Besides the Grameen Bank in Bangladesh which is well
known, BancoSol in Bolivia, Bank for Agriculture and Agricultural Cooperatives (BAAC) in
Thailand, Bank Rakyat Indonesia (BRI) and the National Micro-finance Bank (NMB) in
Tanzania are all successful examples of efficient micro-credit. They all show the important role
micro-credit institutions in developing countries can play in fostering rural development and
how more effective market based institutions can be with respect to direct Government
interventions for directing credit to specific markets at regulated low interest rates. Important
conditions for success include independence of decision-making and a high level of
accountability for financial performance27.

Three examples of successful micro-credit

Banco Sol started in Bolivia in 1987 as a non-profit foundation and in 1992 was turned into a private
bank, the first bank in the world dedicated exclusively to microfinance. By 2002 Banco Sol became the

World Bank (2003)
See World Bank (2003)

largest institution in Bolivian financial markets in terms of the number of loan contracts (35% of the total)
with an outstanding loan portfolio of $ 67 million (see Santos 2003). The profitable strategy of Banco Sol
was to lend to previously unbanked firms and individuals, reducing risk with joint liability contracts and,
as a consequence, charging much lower rates than those available on the informal money market, before its
entry the only available source of funds for its clients. (Andersen and Nina, 2000).

The experience of BAAC in Thailand shows the important role that competition oriented regulatory reform
in banking can have on the profitability of microcredit institutions. BAAC depended initially exclusively
on capital from government, and in the early 1970s displayed a chronic funding shortage and loan
recovery rates as low as 51%. At that time the solution was additional regulation and the Bank of Thailand
adopted an agricultural credit policy in 1975, by which commercial banks were obliged to lend a share of
their portfolio to agricultural sector. Many of these banks, instead of lending directly to agriculture,
deposited their funds with BAAC. As a consequence, the structural shortage of funds suffered by BAAC
disappeared. Banking reforms undertaken between 1988 and 1996 eliminated interest rate ceilings and
restrictions on the opening on new branches, eliminating also the constraints on commercial banks on
agricultural lending. Nonetheless the efficiency of Baac strongly increased and rural deposits became its
main source of funds. By the late nineties its branches had grown from 82 to 535, its outreach and savings
mobilization had raised at such point that it did not even suffer from the financial crisis of 1997 (see
Seibel, 2000).

BRI in Indonesia has been a major provider of microfinance since 1984. By 1989 BRI was able to finance
its lending activity with rural deposits. According to Seibel (2000)

BRI benefited form interest rate deregulation and a management initiative to commercialize
operations by transforming its sub-branches into self-sustaining profit centers. For example it
offered its staff profit-sharing incentives. The bank covers its costs form the interest rate margin and
finances expansion from its profits; its long term loss ratio is only 2.1 percent.

BRI, like BAAC, remained profitable even during the Asian crisis. As Seibel (2000) reports it was the only
profitable entity among the government-owned banks. .

NMB was created in Tanzania after the privatisation in 1997 of the loss making rural branches of the
National Bank of Commerce. After an internal restructuring and a thorough reform of NMB pricing policy,
by 2002 NMB had become profitable without having to close any of its branches. As the World Bank
(2003) reports:

A key initiative has been the development and rolling out of microfinance products, mainly small
(average $400). As of June 2002, 10.000 loans had been disbursed through 36 of the banks 104
branches, with a level of arrears below 2%.


62. We turn now to the interaction between competition law and banking regulation and, in
particular,to an explanation of why the full application of competition law in the banking sector
by a national competition authority is desirable, and in no way incompatible with an effective
regulatory framework.

Competition law should fully apply to banks( in parallel with banking regulation)

Item 4 of the OECD Policy Recommendations on Regulatory Reform specifies that
sectoral gaps in coverage of competition law should be eliminated unless evidence
suggests that compelling public interests cannot be served in better ways. This is
echoed in the Financial Services chapter:

It is important that the rigorous concern for the pursuit of competition policies that
has been a key element of past policies toward the financial services industry be
continued. Basic principles of competition policy should be applied in financial
services as should competition law, subject only to clearly justified exceptions
needed for prudential reasons or other overriding public policy objectives.28

As an aside it is, of course, necessary that the national competition laws are up to
the task29. In particular, the national competition laws must be generally-applicable,
flexible enough to take full account of differences in different sectors, and must be
designed to promote economic efficiency objectives.

Banks should not be subject to their own, special competition rules but should
be subject to general competition rules.

Very often it is proposed that a sector be subject to its own particular set of
competition rules on the grounds that the sector is unusually important or in some
other sense special. The proposal should be treated cautiously. Violations of
competition rules fall within very general categories and are flexible enough to
accommodate any sector specific characteristics. Special competition rules are not
only unnecessary, but they may also undermine enforcement. There is a very thin
line between sector-specific competition rules and continued regulation, especially
if the special rules are to be enforced by the former regulator. There is a danger that
sector-specific enforcers may adopt an understanding of competition that is overly
congenial to the industrys traditional mode of operation instead of promoting a
competitive regime.30 As explained in the following box, sector-specific laws are
more vulnerable to being changed and enforced in the interest of the regulated
industry, rather than in the interest of the economy at large. General laws, on the
other hand, tend to be more immune and therefore more robust and long-lived.

Sector-Specific Or Generic Regulation?

Are sector-specific competition rules preferable to generic competition rules? Is it preferable to have a
sector-specific competition enforcer or an economy-wide competition authority? The answer is that,
wherever possible, generic regulation and generic enforcement is preferred to the sector-specific approach.

The reason is straightforward. Sector-specific institutions encourage sectoral lobbying and are more
vulnerable to industry capture. Experience suggests that firms in a regulated industry will, over time, seek
to influence their governing regulatory regime to their own purposes - for example, to restrict competition.
In particular, the regulated firms will seek to use political pressure - on policymakers and regulators - to
influence the legislation or the enforcement of the regime.

OECD (1997b), p98.
See OECD (1997b) p255.
OECD (1997b), p256.

In contrast to an economy-wide regulatory regime, sector-specific regulation is much more vulnerable to
this form of lobbying. The larger and more diverse are the affected firms, the harder it is to form the
coalition of common interests necessary to maintain a sustained lobbying effort. Generic legislation, which
applies to a large number of firms with different interests, is therefore more stable and more immune to the
tendency for regulation, over time, to operate for the benefit of the regulated industry.

The same is true for the regulatory body itself. Experience suggests that over time, through the sustained
lobbying efforts of the industry, sectoral regulators tend to be influenced by the specific interests of the
industry they regulate. That is, it becomes increasingly harder for the regulatory body to distinguish the
public interest from the interest of the industry. Regulators, in direct contact with those whom they
regulate rather than with consumers, tend to identify more with suppliers and their problems than with the
general public and its problems.31 A generic regulator with experience in a large number of industry
sectors is able to more easily discern self-interest in the arguments of the regulated firms and is less likely
to be peopled with staff who see a bright future for themselves in the regulated industry.

Importantly, a sector-specific regulator may also become an obstacle for regulatory reform. Over time, the
interests of the regulatory body and the regulated industry may converge - both have a strong interest in the
continuation of the sector-specific regulation, even where the underlying reason for the regulation no
longer exists. Indeed, where the underlying reason for the regulation disappears, sector-specific regulators
have a strong incentive to find alternative reasons for regulation, in order to ensure its continued survival.
A generic regulator, in contrast, has little interest in the continuation of any specific regulation and
therefore can act as an important influence, where appropriate, for regulatory reform.32

More generally, a sector-specific regulator has incentives to argue against structural reforms or other policy
actions which the expand the role of competition (and therefore reduce the responsibility of the regulator)
within the regulated sector.33
In addition, a generic law is likely to develop a larger body of case law more quickly than a sector-specific
Where generic competition rules apply to the financial sector, banking supervision authorities, if charged
with their enforcement, may be naturally led to take into account, in a non-transparent way, concerns
relating to the stability of banks and to adopt an improper regulatory approach in the application of
competition rules, for instance, as far as the choice of remedies is concerned.
Finally, also due to the removal of most regulatory line of business restrictions in many countries, it is
becoming increasingly difficult to design an effective and stable system in which a subset of markets or
firms is not under the jurisdiction of the economy-wide competition authority but of a sector-specific
competition law enforcer.

Antitrust law should be enforced by the general antitrust authority, not by the
specialized sectoral regulator

Benston (1973), p221.
Sector-specific agencies may resist the pro-competitive thrust of reform because of self-interest. An agency whose chief purpose
is to regulate an industry ensures its own survival by keeping regulation in place. The general jurisdiction competition-enforcement
agency, which has no such concern with respect to any particular industry, may be able to assess competitive conditions and
opportunities more impartially. OECD (1997b), p256.
There are other arguments in favour of generic legislation. For example, the development of a body of case law is likely to be
more rapid under industry-generic legislation, enhancing industry certainty. It might be argued that industry-specific measures are
preferable when there are serious shortcomings in the generic competition law. In this case, however, rather than introduce sector-
specific rules, these shortcomings should be addressed as soon as possible.

Again, as the box emphasizes, there are strong reasons for preferring that
competition rules be applied by the antitrust authority and not by the sector-
specific regulator. Should sectoral expertise be necessary for competition
decisions, this can be addressed through formal or informal consultation of the
sector regulator by the competition authority. The OECD Report on Regulatory
Reform notes:

Reformers should pay special attention to experiences of agencies such as the US

Interstate Commerce Commission and the Civil Aeronautics Board. Though
originally charged with ensuring competition, these two regulators became means
for maintaining cartels. The problems persisted after the old agencies were
abolished. For several years after the US airline industry was deregulated,
jurisdiction over airline mergers rested with the Department of Transportation,
rather than the antitrust agencies. The Department approved several combinations
leading to significant market power in several city-pair markets, despite vigorous
objections from the antitrust authorities. The same thing happened in the case of a
railroad merger approved by a special Board within the US Department of

63. The process of regulatory reform in the banking sector, which has occurred over the past
two decades, has significantly increased the role of competition in the banking sector. At the
same time, there has been a movement (in those countries which had partially or totally
exempted their banking systems) to extend the jurisdiction of national competition laws to
include banks:

Finnish legislation has been largely emended in 1998, removing special provisions for
bank mergers. The Irish Competition Act 2002 assigns to the Irish competition Authority
all powers on mergers, including banks.
In France bank mergers fall now fully under the general antitrust provisions. The French
Authorities have to consult with the banking regulator before taking a decision and
should provide a full explanation, should they decide to deviate.
In Canada the Competition Act of 1986 brought bank mergers and interbank agreements
within the scope of the general competition law (subject to a general right of
authorization of mergers by the Minister of Finance). Prior to this new Act, interbank
agreements and mergers involving banks were exempted from competition law.
In Germany special treatments for banks under the competition Act have been
progressively eroded and all remaining privileges have been lifted as of January 1 2000.
In Portugal the new Competition Act applies fully to banks.
The European Court of Justice confirmed in 1981 that EC competition law has
always fully applied to the banking sector.

64. In almost all jurisdictions Ministries of Finance or Central Banks have the duty to control
bank mergers for stability reasons and for ensuring the safety and soundness of the
institution and its managerial competency, while competition authorities control them on
competition grounds. Only in very few jurisdictions competition and stability concerns are
pursued by the same institution:

OECD (1997b), p256.

In Brasil the Central Bank has full responsibility over bank mergers (both for stability and
for competition considerations).
In South Africa, the Minister of Finance for public interest objectives can exclude the
competition authorities jurisdiction over bank mergers.
In the US, under section 18(c) of the Bank Merger Act of 1966, the Comptroller of the
Currency (OCC) for national banks, the FDIC for federally-insured, state-chartered banks
that are not members of the Federal Reserve System and the Board of Governors of the
Federal Reserve System for state-chartered banks that are system members, must conduct
their own competitive analysis of bank mergers. However in most transactions only
DOJ and a single bank regulatory agency actually are involved and obtain a
competitive factors reports from the Attorney General of the United States before
approving a bank merger.
In Italy the antitrust law provisions apply to banks but they are enforced by the Central
Bank (only in so far as the conduct or the merger produces effect on credit-making and
deposit-taking markets). In such cases the antitrust authority is obliged to provide an
advice. In all other circumstances the antitrust authority is fully responsible.
In Korea, the Financial Supervisory Commission, when considering an approval of a
merger or an acquisition, has to have prior consultation with the Korea Fair Trade
Commission on the effect of the operation on competition.



65. We turn now to the issues that arise in applying competition law in the banking sector. In
particular we will address some of the problems arising in merger control, as an example of how
a competition authority applying competition law can bring added value (for example, in the
field of market definition, which has been under discussion for quite some time). For reasons of
concision, restrictive agreements and abuse of dominance in banking are not analyzed in detail
in this report.

66. During the last fifteen years there has been a decline in the number of banks in many
OECD countries.35 Reasons for the consolidation of banking activity include (amongst other
factors) the relaxation of restrictions on the geographic area that a bank can serve, and
elimination of other structural regulations that may have served to shelter relatively inefficient
banks from competition.36 An additional factor is the adoption of new information processing
technologies which has increased the efficient scale of operation in some bank activities.37

Framework for analyzing bank mergers38

In the U.S., for instance, the number of banks declined monotonically from 14,230 in 1983 to 10,313 in 1994. Over this twelve
year period, entry of 2,416 newly chartered banks more than made up for the 1398 banks that failed and exited. The net decline
represents a wave of merger activity among banks in the U.S. which has no parallel since the Great Depression. Not only has there
been a large number of mergers in the recent past, but a number of individual mergers that have taken place during the 1990s rank
among the largest U.S. bank mergers ever, in terms of the real value of the assets involved and also in terms of the share of total U.S.
bank assets accounted for by the merging banks. Rhoades (1996a) , Rhoades (1997) ,.
Rhoades (1996b) , Rhoades (1997), Berger, Kashyap, and Scalise (1995)
Description and some discussion of changes in regulations and other forces relevant to the competitive analysis of banking
markets in Europe can be found in Gual and Neven (1992) .
This section closely follows Rozanski and Rubinfeld (1997).

67. In assessing the likely effect of a bank merger on competition, in principle one should
consider whether the merger could create or facilitate the exercise of market power, where
market power is defined as the ability of firms to increase price or reduce quality from pre-
merger levels. A merger could have anticompetitive effects by making it profitable for a leading
firm to exercise market power unilaterally, or by increasing the likelihood that firms in a market
could successfully maintain a collusive outcome.

68. To evaluate the effect of a merger, it is essential to analyze the mergers impact on the
range of services provided by banks. Banks sell a wide range of services or products, including
deposit, loan, and investment services sold to retail customers; deposit, loan, and various other
services sold to businesses, and also correspondent services, which are specialized services
supplied by a relatively limited number of banks to other banks, often for resale to the ultimate
purchaser. Trade finance, custody, check clearing services, and foreign exchange services are
examples of correspondent services. Banks in some countries are restricted in their ability to
offer underwriting services, insurance, and some investment products. There are fewer
limitations on the ability of banks to offer these products in most other countries.

69. In general, the analysis of the likely effects of a merger on competition must take into
account a number of factors. One factor is the possibility that prospective purchasers of a
product would choose to substitute to alternative products in response to a small but significant
increase in the relative price of the product. If such substitution would not occur in an amount
sufficient to make the price increase unprofitable then the product constitutes a relevant product
market. A second factor is the possibility that prospective purchasers could turn to alternative
sources of supply, including firms that currently produce and sell the product in other
geographic areas. If such substitution away from firms located in a given area would not be
significant, then the area constitutes the geographic market. The possibility of significant new
competition from entry by firms that dont currently produce or sell the product is a third

70. The structure of competition in the relevant product and geographic market, including the
number and relative competitive effectiveness of current market participants, affects the
likelihood that a merger be anticompetitive. Other characteristics of competition in the market
also affect the likelihood of anticompetitive effects. For example, if there is significant product
differentiation, and if products sold by the merging firms are perceived by purchasers to be
relatively good substitutes, than there is a greater possibility of unilateral anticompetitive
effects. If firms have good information about the competitive actions of their rivals, and if
competitive strategies can be revised quickly, then coordinated anticompetitive effects are more
likely. Finally, in some countries competition law allows consideration of a possible efficiency
defense - if a proposed merger holds the promise of real efficiencies that could not reasonably
be achieved through other means, these efficiencies could serve to lessen concerns about the net
effects of the merger on competition.

71. The analytical framework described above will result in different policy
recommendations for bank mergers in different countries, because of significant differences in
the structure of competition, the preferences of purchasers of bank products (and the set of
alternatives they face) and the institutional context. The following paragraphs set out an
indicative approach to the analysis of competition in the markets for small business loans and

These are the arguments used by the DOJ/FTC in their merger guidelines in their hypothetical monopolist test for market

consumer bank products, two bank products for which competition concerns tend to be the

Small Business Loans

72. Small businesses40 typically have obtained a variety of credit products from banks,
including mortgages on commercial property, and loans to purchase or lease vehicles,
equipment, and other capital goods. In recent years however non banks have started to enter into
this filed offering a number of credit products to small businesses, such as factoring, leasing and
mortgages. On the other hand businesses that have a need for a line of credit for startup or
working capital are likely to have a limited ability to substitute away from their bank.

73. It is not uncommon for small businesses to rely to a significant extent on personal credit,
such as general purpose consumer credit cards or a second mortgage on a personal residence.
These alternatives are likely to be viewed as inferior, however, because they are relatively high
cost, and they put personal assets at risk. The question for antitrust analysis is whether, as a
result of a merger, banks are likely to find it profitable to raise prices with respect to small
business loans. The answer to this question depends on the willingness of businesses that would
obtain a line of credit from a bank at prevailing prices to substitute to another bank or to
alternatives in response to an anticompetitive price increase. The fact that some businesses use
these alternatives at prevailing prices demonstrates the feasibility of substitution, but does not
establish that such substitution would occur in an amount sufficient to make an anticompetitive
price increase unprofitable; the analysis must attempt to quantify the likely magnitude of such

74. The next step in the analysis of the likely effects of a proposed merger on competition to
supply small business lines of credit is the determination of which banks and which bank
locations are able to compete effectively to supply the product. In the past there have been
strong reasons why small businesses tended to obtain lines of credit and some other key bank
products from nearby suppliers. In part, this was due to the information advantages a nearby
supplier would have on local enterprises, coupled with a strong preference that some services
used on an almost daily basis, such as transaction services (the provision of currency and coin,
acquisition of credit card receipts, night deposits, and electronic funds transfers) and demand
deposit accounts, be quickly accessible. The internet and internet banking may change all this,
considering that credit to small businesses is mainly based on collaterals.

75. To the extent that banks finance investment on the basis of its profit opportunities, than
local banks are relatively better placed, considering their superior knowledge of local business
conditions which tends to make them better informed about the risks associated with a new
business startup, while their proximity to local businesses tends to lower costs of monitoring
performance and updating information about credit risk. Local banks are therefore likely to be
able to identify small businesses that are better credit risks and compete successfully to win
their business by offering relatively favorable terms. It is true that some banks and other
providers of credit to small businesses are sometimes located a great distance away.41 In the
case of vehicle or equipment loans that are secured by the capital good being financed, the
riskiness of the loan is reduced and the informational advantage of local banks is eroded. In the

"Small" businesses are defined, e.g., by the US DoJ to be those with annual revenues in the range of one to ten million dollars.
Wells Fargo & Co., a California bank, initiated a strategy in 1995 of marketing lines of credit to small businesses nationwide
using direct mail. Some other banks have imitated this strategy. Oppenheim (1996) , Oppenheim (1997) More recently, Wells
Fargo has solicited applications through its web page.

case of lines of credit, distant suppliers lacking a branch network or significant presence in a
local market are likely to regard all but the most well-established small businesses as relatively
high risks. Distant suppliers may compete successfully to make loans that the better informed,
local lenders also identify as high risk, but they may not be competitive in the case of
borrowers that local lenders identify as relatively good risks. It is competition to supply
services to these borrowers that is at issue from a merger of local banks.42

76. In regard to the analysis of entry conditions, studies of entry in local banking markets
show that entry appears to be driven largely by factors such as the growth of economic activity
in the area and the current density of banks and branches, rather than by the measured
profitability of incumbent banks. It seems unlikely that the entry decision would turn on
increased profit opportunities in a relatively small activity such as small business lines of credit.
In addition, new entrants may require several years to establish themselves as effective
competitors to make small business loans, because of the importance of private information,
reputation, and long-standing business relationships in this activity. The possibility of
exogenous entry is an important factor to consider, but it may not be possible to count on quick
and effective entry to counter the effects of an otherwise anticompetitive merger.

Consumer bank products

77. In the case of some important consumer bank products, such as home mortgages, car
loans, and credit card loans and transactions services, distant banks and specialized non-banks
are increasingly demonstrating their effectiveness as competitors. The analysis of consumer
home mortgages and car loans bears some similarity to asset-backed loans made to businesses:
the fact that the collateral is relatively easy to evaluate makes competing in this market easier
for non-local suppliers. Credit cards in many countries are often marketed on a national basis
by direct mail and telephone. Such credit card issuers rely on credit histories assembled by
third-parties (where they exist) and on credit-scoring software that predicts credit risk. Credit-
scoring algorithms have so far proven to be more useful in this application than in the case of
small business lines of credit.

78. Consumers tend to prefer to obtain checking account services from a conveniently located
supplier. Because many consumers who commute a significant distance to work consider a
bank location near their workplace to be a good substitute for a bank location near home, the
geographic market is relatively large. Also, in some countries (in contrast to the analysis of
small business bank products) there are other non-bank depository institutions (such as thrifts or
credit unions) which are active suppliers of consumer bank products. The advent and spread of
ATMs, electronic funds transfer, and the development of home banking via computer or
telephone raise the likelihood that local banks with branch networks lose their competitive

In the case of a market such as that for small business lines of credit in which suppliers are significantly differentiated based on
their locations, competitive interactions among firms located along a geographic continuum can be sufficient to conclude that the
geographic market is much larger than would be suggested by the strong preferences of customers for local sources of supply. Each
firm is constrained only by the few competitors in its immediate neighborhood, but the effects of competition at one end of the
spectrum are transmitted from local area to local area and may be felt at a great distance. In theory, however, even if there is no
break in the geographic chain of substitutes, the exercise of market power over a limited portion of the spectrum may be
profitable because the profits that can be earned by increasing price to inframarginal customers who lack good alternatives more
than makes up for the loss of business at the margin. In the case of bank loans, the possibility of price discrimination simplifies the
analysis, and may make it possible to define geographic markets that are quite narrow. Price discrimination in the case of small
business loans is likely to be a successful strategy: significant arbitrage among borrowers is implausible, and banks can use
information obtained in the loan application process to develop good information about the willingness of customers to substitute
toward other suppliers. Banks can meet competition at the margin by lowering prices selectively to some customers.

advantage, and that geographic markets for consumer bank products become much larger. Also
internet banking is quickly developing in many countries.

Cluster market approach

79. The methodology described above considers separately the effects of a bank merger on
competition to supply each bank product. An alternative approach views the relevant product
for analyzing bank mergers as the cluster of products and services that constitutes "commercial
banking."43 This cluster includes consumer loans and consumer banking services as well as
business loans and products.44

80. Some have argued that the cluster approach is not appropriate because banks are not
constrained to raise the prices of all services they offer uniformly. Banks would not be deterred
from raising the price of one product, such as a small business line of credit, by the possibility
that prospective loan customers would substitute to other products in the cluster, such as a
checking account. Nor would an increase in the price of the loan be defeated by competition
banks face to supply other products in the cluster.

81. On the other hand, others believe that the cluster market approach gives the right answer,
especially if there were strong economies of scope in production, so that all banks supplied all
products in the cluster in the same proportion and if there were strong complementarities in
demand, so that all consumers consumed all products in the cluster in the same proportion. For
example, in analyzing a merger of firms that produce shoes, it probably would not matter much
to the conclusion if the analysis was done in terms of right shoes, or left shoes, or pairs of shoes.

82. In the case of the "commercial banking cluster", some firms in fact compete very
effectively in supplying some, but not all, products in the cluster. In addition, although
consumers and businesses do tend to purchase multiple services from their primary financial
institution, they do unbundle purchases today, and would likely unbundle to a greater extent if
their current bank increased prices of some products in the cluster. The cluster market approach
appears to understate competition in the market by ignoring the role of specialized providers of
some services.

83. The cluster market approach may overstate competition in the market by wrongly
inferring from the existence of abundant competition to supply one product in the cluster that
competition in other product markets is sufficient. For example, suppose that the relevant
geographic market was defined by commuting patterns. This is sensible in the case of consumer
banking products, for which consumers consider services from banks located near their home or
near their work to be good substitutes. But the resulting geographic markets are sometimes far
larger than is appropriate to analyze competition for many small business bank products, for
which proximity of the bank to the place of business is key. In cases in which the structure of
competition is not homogeneous throughout the broad geographic market, the cluster market
approach may miss adverse effects of the merger on local competition.45

U.S. v. Philadelphia National Bank, 374 U.S. 321 (1963); U.S. v. Phillipsburg National Bank & Trust Co., 399 U.S. 350 (1970)
In the U.S., the cluster market approach guides the decisions of the Federal Reserve Bank.
The Australian Competition and Consumer Commission rejected the cluster market approach when analyzing the 1997
Westpac/Bank of Melbourne merger. The ACCC concluded that the geographic market for home loans was national, but that
geographic markets for demand deposits and small business banking products did not extend beyond state boundaries. The
existence of national competitors in the home loan market was correctly understood to be irrelevant to the competitive analysis of
other product markets. Also the EC Commission does not follow a cluster market approach when defining markets in the
competitive analysis of bank mergers.


84. This report has sought to review regulations governing banks in the light of established
principles for good regulation. It raised the question of what, exactly, is the problem (i.e., the
market failure) that (prudential) regulation of banks is designed to address. In particular, while
there are some problems that need a regulatory intervention (protection of smallest depositors,
proper regulation of banks settlements, mandatory information disclosures, risk adjusted
stability concerns), for the rest the sector can be efficiently disciplined by market mechanisms
and by antitrust law. The Report addressed then the importance of switching costs for increasing
market power of each single bank, identifying regulatory solutions to reduce their importance.

85. In terms of recommendations, jurisdictions should:

promote an open, competitive, banking environment without unjustified restrictions

on entry, ownership or exit, resulting either from the rules to be applied or from
enforcement practices;

ensure that there is a proper separation between the enforcement of prudential

regulation and of the general competition rules;

86. In addition agencies should:

whatever the institutional setting, build good working relationships with the
regulatory agencies and coordinate their efforts in reviewing particular matters.

apply in enforcement the usual tools of antitrust analysis, including market

definition, market power/dominance, remedies.

87. Finally, agencies in their competition advocacy functions should consider, as appropriate
when competition concerns are raised, to advocate for:

the elimination of exclusions from competition law for financial institutions;

an environment where banks are informed in a timely and complete manner on the
debt exposure of potential borrowers (in integrated financial markets also on an
international basis), making sure to identify ways and precautions such that
information sharing does not lead to restrictions of competition;

a reduction of switching costs by depositors, for example by asking for disclosure

rules, for example on the costs associated with the closing of an account or paying
off a mortgage;

in countries with a common currency, a reduction of transaction costs on cross

border payments, including the creation of larger than national payment systems,
so as to favor the development of larger markets and greater choices for consumers;

a legal environment where the taking possession of collateral is possible without

especially in developing countries and consistent with maintaining a competitive

market., the creation of a legal environment where financial institutions can reduce
their risk by joint liability lending.

Andersen, Lykke Eg and Osvaldo Nina, "Micro-credit and Group lending: The collateral
effect", Working Paper n.. 1998/18 University of Aarhus, 2000.
Baltensperger, Ernst, "The Economic Theory of Banking Regulation" in Furubotn and Richter,
eds, The Economics and Law of Banking Regulation, p 1, 1989
Benston, George and George Kaufman, "Is the Banking and Payments System Fragile?",
Journal of Financial Services Research, vol 4, p 209-240, 1995
Berger, A.N., Kashyap, A.K., and Scalise, J.M., "The Transformation of the U.S. Banking
Industry: What a Long, Strange Trip Its Been", Brookings Papers on Economic Activity
2:1995, 55-217, 1995
Bryant, J. "A model of Reserves, Bank Runs and Deposit Insurance", Journal of Banking and
Finance, 43, 749-761, 1980
Bank for International Settlements, International Convergence of Capital Measurement and
Capital Standards: A revised Framework, A Report of the Basel Committee on Banking
Supervision, 2004
Claessens, Stijn and Laeven, Luc, "Financial Dependence, Banking Sector Competition and
Economic Growth", World Bank Policy Research Working Paper 3481, January 2005
Crockett, Andrew, Harris, Trevor, Mishkin, Frederic S and Eugene N White, Conflicts of
Interest in the Financial Service Industry: What Should We Do About Them? Geneva Reports
on the World Economy 5, ICMB and CEPR, 2003.
Dewatripont, Mathias and Tirole, Jean, The Prudential Regulation of Banks, MIT Press:
Cambridge, Mass.
Diamond, D. and Dybvig, P.H., "Bank Runs, Deposit Insurance and Liquidity", Journal of
Political Economy, 91, 401-419, 1983.
Dymski, Gary, A. , "Banking on Transformation: Financing Development, Overcoming
Poverty", presented at Seminario Brazil em Desenvolvimento, Universidade Federale do Rio
de Janeiro, September 2003.
Edey, Malcolm and Hviding, Ketil, "An Assessment of Financial Reform in OECD Countries",
OECD Economics Department, Working Papers, No. 154, 1995
Feldstein, Martin, "The Risk of Economic Crisis: Introduction" in Feldstein, Martin, ed., The
Risk of Economic Crisis, Chicago, 1991
Financial Times, Wednesday, November 19, 1997
Gual, J., and Neven, Damien, "Deregulation of the European Banking Industry (1980-1991)",
Discussion Paper No. 703, Center for Economic Policy Research, August, 1992.
Kaufman, George, "Bank Failures, Systemic Risk and Bank Regulation ", Cato Journal, vol 16,
no. 1, p 17, 1996
Krugman, Paul, "Balance Sheet, the Transfer Problem and Financial Crisis" 1999
Mishkin, F., "Asymmetric Information and Financial Crises" in R.G. Hubbard ed., Financial
Markets and Financial Crises, pp 69-108, 1991
OECD, "Competition, Consumers, and Regulatory Reform", chapter 3 in The OECD Report On
Regulatory Reform: Thematic Studies, 1997a
OECD, "Regulatory Reform In The Financial Services Industry", The OECD Report on
Regulatory Reform: Volume I, 1997b
OECD, "Contribution from Pakistan", Global Forum on Competition, Session I, February 2005
Oppenheim, S., "Show Me the Money, Not the Bank, Businesses Say", Am. Banker 162, May
19, 1997
Oppenheim, S., "Wells Small-Business Lending via Mail Pays Off", Am. Banker 161,
December 23, 1996, 1

Parry, Robert T., "A Central Bankers Perspective on Bank Reform", Business Economics, July
Rajan, R.G. and L. Zingales, Saving Capitalism from the Capitalists Princeton: Princeton
University Press, 2003
Rhoades S.A., "Bank Mergers and Industrywide Structure, 1980 - 94", Staff Studies 169, Board
of Governors of the Federal Reserve System, January, 1996a.
Rhoades, S.A., "Competition and Bank Mergers: Directions for Analysis from Available
Evidence", The Antitrust Bulletin, Summer 1996b, 339-363.
Rhoades, S.A., "Research on IO Topics in Banking: An Introduction and Overview", Review of
Industrial Organization 12, 1-8, 1997
Rozanski, George and Dan Rubinfeld, "Mergers and Other Competition Policy Issues in
Banking", an Appendix to the United States Submission to the OECD Report on Enhancing
the Role of Competition in the Regulation of Banks, 1997.
Santos, Nino, Financing small, medium and micro enterprises in post conflict situations.
Microfinance opportunities in the Democratic Republic of Congo OECD Development Center
working paper, 2003.
Scott, K., "Never Again: The S&L Bailout Bill" in The Economics And Law of Banking
Regulation, 1989
Seibel, Hans Dieter, "Agricultural Development Banks: Close Them Or Reform Them?,
Finance & Development, 2000
World Bank, Finance for Growth, Policy Choices in a Volatile World, Washington, D.C., 2001
World Bank, Rural and Financial Services, Washington, D.C., 2003

November 2014

Working paper 407

Financial regulation in Kenya:

Balancing inclusive growth with financial stability
Francis M. Mwega

This case study investigates the potential tradeoffs between regulations and
stability of Kenyas financial sector and their implications for inclusive growth.
This is done in the context of six areas: (i) size and growth of the financial sector
relative to LICs and MICs; (ii) implications of a mixture of local banks (some of
which have spread to neighbouring countries), foreign banks and development
finance institutions; (iii) evolution and macroeconomic implications of financial
innovations and inclusion; (iv) cost and access to credit, especially to SMEs; (e)
prudential regulations; and (f) management of capital flows in the context of
large current account deficits, mainly financed by short-term net capital inflows
such that their easy reversibility could potentially generate a currency crisis.

Shaping policy for development


This paper is an output of the Grant "Financial regulation in low-income countries:

Balancing inclusive growth with financial stability" funded by the DFID-ESRC
Growth Research Programme (DEGRP). ODI gratefully acknowledges the support
of DFID/ESRC in the production of this study. The views presented are those of the
author and do not necessarily represent the views of ODI or DFID/ESRC.
Table of contents

Acknowledgements ii

List of figures and tables ii

1 Introduction 1

2 Size and growth of the financial sector 2

2.1 The financial sector in Kenyas Vision 2030 2
2.2 Financial Sector and economic performance 5

3 The roles of foreign banks, state-state owned banks and DFIs in Kenya 9
3.1 Foreign and state-owned commercial banks 9
3.2 Challenges of Regulating Kenya banks in other countries 12
3.3 3.4: Development Finance Institutions 13

4 Financial inclusion in Kenya 15

4.1 Trends and patterns of financial inclusion 15
4.2 Is regulation of M-PESA and other mobile money platforms adequate? 17
4.3 Financial inclusion (innovations) and macroeconomic stability in Kenya 18

5 Access and cost of credit in Kenya 19

5.1 Introduction 19
5.2 Banks lending to SMEs in Kenya 20
5.3 Cost of credit and interest rate spreads in Kenya 21

6 Prudential regulations in Kenya 27

7 Management of capital flows in Kenya 30

7.1 Evolution of current account deficit and net capital inflows in Kenya 30
7.2 ODA Flows 32
7.3 FDI Flows 33
7.4 Capital account regulation to avoid future currency or banking crises 34

8 Summary and conclusions 36

References 42

Financial regulation in Kenya: i

List of figures and tables

Figure 1: M2 as % of GDP in Kenya versus LICs and MICs 6
Figure 2: Domestic credit to the private sector as % of GDP in Kenya versus LICs
and MICs 6
Figure 3: Domestic credit to the private sector (DCP) and nominal GDP in Kenya,
2005:12 2013:12 7
Figure 4: Quarterly Growth in Financial Intermediation and GDP in Kenya, 2001Q1-
2013Q3 7
Figure 5: The money multiplier and income velocity in Kenya, December 2005 to
December 2013 18
Figure 6: The interest rates spread in Kenya (ex post lending minus deposit rate) 23
Figure 7: The interest rates spread in Kenya (ex post lending rates minus the 91
days TBR) 23
Figure 8: The Performance of the banking Sector, 2002-2012 24
Figure 9: Spreads in Low-Income and Middle-Income Countries 25
Figure 10: Selected prudential and financial stability indicators for the banking
sector 2011 - 2013 28
Figure 11: 12-months cumulative current account deficit as % of GDP, December
2005-November 2013 30
Figure 12: Net capital flows to Kenya, US$ million, December 2005-November
2013 31
Figure 13: Foreign Currency advances and deposits in Kenya, January 2007
December 2013 35

Table 1: Ganger-causality between quarterly growth in financial intermediation
(QGFI) and growth in GDP (QGGDP) 8
Table 2: The performance of commercial banks in Kenya by ownership 10
Table 3: Simulated cost of banking services to SMES 12
Table 4: Financial inclusion and exclusion in Kenya, % 15
Table 5: Overall Use of financial services, % 16
Table 6: Financial inclusion in Kenya by gender 16
Table 7: Financial inclusion in Kenya by location 17
Table 8: Comparative Analysis of Commercial Banks Ex Post Spreads in Kenya
and Selected Countries (%) 23
Table 9: Ex post Spread Decomposition in Kenya, % 26
Table 10: Net Foreign Purchases as % Share of Equity Turnover in Kenya, January
2009-December 2013 31
Table 11: Net ODA to Kenya, 2002-2011 32
Table 12: Public Debt Sustainability in Kenya 32
Table 13: Net FDI inflows to Kenya, 2002-2011 33

Financial regulation in Kenya: ii

1 Introduction 1

In the wake of the global financial crisis (GFC), many countries are prioritizing
stability by strengthening financial regulation. Although important, this might be at
the expense of inclusive growth, especially in poor countries. Without effective
regulation, financial systems can become unstable, triggering crises that can
devastate the real economy as evidenced by the recent GFC that began in 2007
(Spratt 2013). Given the primary purpose of finance is to facilitate productive
economic activity, the aim of regulation is to maintain financial stability and to
promote economic growth. This is a delicate balancing act, as too great a focus on
stability could stifle growth, while a dash for growth is likely to sow the seeds of
future crises.

There are two different ways that regulation could impact on growth and stability
(Spratt 2013). The first is by influencing the day-to-day behaviour of financial
market actors so that financial regulation has direct effects, for example, on how
much a bank chooses to lend to small and medium enterprises (SMEs). The second
is by influencing how the financial system evolves structurally, thereby creating
indirect effects. The diversity of the banking system, for example, will influence the
pattern of lending by sectors.

This case study investigates the potential tradeoff between regulation and stability
in Kenya, a small open economy which is highly vulnerable to domestic and
external shocks, but with a lightly regulated financial system and a fairly open
capital account. The study adopts an empirical approach, entailing quantitative
work and focused policy analysis. The specific objectives of the Kenya case study
are therefore to identify and analyze (i) key national risks to financial stability as
well as obstacles or gaps in financial sector for funding inclusive growth; (ii)
domestic regulatory measures that have been implemented, future options to
support financial stability and the advantages and problems of different
mechanisms for such regulation, given the country characteristics (e.g. weak
institutions, governance and law enforcement, and information problems); and (iii)
the management of capital account to support financial stability prior, during and
after the recent global financial crisis.

To make the research manageable, the study mainly focuses on the banking sector,
although capital markets, pension funds and other financial institutions may
facilitate more long term finance if banks do not provide sufficiently. The Terms of
Reference for the research project identify a number of issues that require
investigation. The paper is therefore organized around these issues. Section 2
analyzes the size and growth of the financial sector and its linkages to economic
performance; Section 3 investigates the role of foreign banks, state-owned banks
and development finance institutions (DFIs); Section 4 examines the evolution of
financial inclusion in the country; Section 5 discusses access and cost of credit;
Section 6 explains prudential regulations; while Section 7 analyzes the management
of capital flows in the country. The paper is concluded in Section 8.

This draws on the studys Terms of Reference.

Financial regulation in Kenya: 1

2 Size and growth of the
financial sector

2.1 The financial sector in Kenyas Vision 2030

The starting point of the study is an analysis of the features and vision of
development of the country in the medium term for example as articulated in Kenya
Vision 2030 and the Medium Term Plans (MTPs), given the countrys main
opportunities (such as the recent discovery of commercially viable oil deposits and
of rare minerals in the country) and challenges (such as continued lack of access
and high cost of credit, especially for SMEs).

Kenya Vision 2030 is the countrys development blueprint which was launched in
2008 (Kenya 2007). It aims to transform Kenya into a newly industrializing,
middle-income country providing a high quality life to its citizens by the year
2030. Its overarching objective is to make Kenya a globally competitive and
prosperous nation with a high quality of life by 2030. The Vision is based on
three pillars: the economic, the social and the political. The economic pillar aims
to improve the countrys prosperity through an ambitious economic development
programme that would achieve an inclusive average GDP growth rate of at least
10% per annum over a period of 25 years. The social pillar seeks to build a just
and cohesive society with social equity in a clean and secure environment. The
political pillar aims to realize a democratic political system founded on issue-
based politics that respects the rule of law, and protects the rights and freedoms of
every individual in Kenyan society. These three pillars are anchored on
macroeconomic stability; continuity in governance reforms; enhanced equity and
wealth creation opportunities for the poor; and investment in infrastructure; energy;
science, technology and innovation; land reforms; human resources development;
security; and public sector reforms.

The Vision identifies financial services as one of six sectors that are the key drivers
of the economy. The others are tourism; agriculture and livestock; wholesale and
retail trade; manufacturing; and business process outsourcing as well as other IT
enabled services. Subsequently, oil and mineral resources sector was added in the
second MTP after the discovery of commercially viable oil deposits and of rare
minerals in the country in 2012. The Vision aims to create a vibrant and globally
competitive financial sector that will create jobs and also promote high levels of
savings to finance Kenyas overall investment needs. It envisages a dynamic
financial sector comprised of banks, the capital market, insurance, pensions,
development finance and financial co-operatives (SACCOs). The Vision therefore
aims to revamp Kenyas fairly diversified financial sector which currently includes
the following institutions:

The capital market, with the stock market the 5th largest by market
capitalization in Africa after South Africa, Egypt, Nigeria and

Financial regulation in Kenya: 2

38 insurance companies.
5 Development Finance Institutions (DFIs) that provide medium and
long-term finance.
1 mortgage company.
7 representative offices of foreign banks.
101 foreign exchange bureaus, first licensed in January 1995.
A Post Office Savings Bank, supported by 890 post offices spread
throughout the country.
About 2700 Savings and Credit Co-operative Organizations
(SACCOs) in both rural and urban areas.
2 credit reference bureau.
9 deposit-taking microfinance institutions, and so on.

The envisaged policy actions and targets of the financial sector under Vision 2030

Raise savings rates from 17% to 30% of GDP. This would be

achieved, for example, by increasing bank deposits from 44% to 80%
of GDP and by lowering borrowing costs. With an average loans
deposits ratio of 76% (over 1978-2012), this implies an increase in the
bank loans from 33% to 61% of GDP. These targets indicate what the
Vision envisages as the desirable scale of banking sector to achieve
middle income status.
Enhance financial inclusion by decreasing the share of population
without access to finance by about 20%.
Increase stock market capitalisation from 50% to 90% of GDP.
Source foreign savings for investment by up to 10% of GDP from
foreign direct investments (FDI), overseas development assistance
(ODA) and sovereign bonds.
Undertake reforms of the banking sector to facilitate the
transformation of the large number of small banks in Kenya to few
larger and stronger ones.
Introduce credit reference bureau.
Streamline informal finance, SACCOs and microfinance institutions.
Deepen financial markets by raising institutional capital through
pension fund reforms and expanding bond and equity markets.
Introduce legal and institutional reforms that would enhance
transparency in all transactions, build trust and make enforcement of
justice more efficient.
Create a critical mass of skills in financial management.
The Kenya Vision 2030 was to be implemented in successive five-year Medium-
Term Plans, with the first MTP covering the period 2008 2012 (recently
completed), and the current second MTP covering the period 2013-2017 (Kenya

The flagship projects and policies that were to be implemented during the First
MTP (2008-2012) included (i) transformation of the banking sector to bring in
fewer stronger, larger scale banks; (ii) development and execution of a
comprehensive model for pension reform; (iii) pursuance of a comprehensive
remittances strategy; (iv) formulation of a policy for the issuing of benchmark
sovereign bonds; and (v) implementation of legal and institutional reforms required
for a regional financial centre.

Financial regulation in Kenya: 3

According to the Second MTP (2013-2017), some of these projects and policies
have not been implemented at all or have been implemented only partially. The
MTP attributes this to a number of factors which include (i) the post-election
violence of 2007/2008; (ii) adverse weather impacting the agricultural sector and
the economy; and (iii) the GFC of 2007/08 and the subsequent worldwide
economic slowdown.

As a result:

Gross national savings as percent of GDP actually decreased from

15.4% in 2007/08 to 10.4% in 2011/12, well below the First MTP set
target of 24.4%.
Total investments as a percentage of GDP rose marginally to 21.9% in
2012/2013 compared to 20.1% in 2010/2011 against a set target of 30-
Credit extended to the private sector amounted to 36.8% of GDP in
2012 compared to 28.3% in 2007.
Nevertheless, the Second MTP identifies some of the following key achievements
under the First MTP:

Increased efficiency of financial services that directly supports

improved credit access by reducing transaction costs. A number of
interventions, including in the payments system, capital markets
infrastructure and credit referencing contributed to efficiency gains
during the period.
Introduction of credit reference bureau. The Banking (Credit
Reference Bureau) Regulations, 2008, were first rolled out in July
2010 and by December 2013, the two licensed credit referenced
bureaus (Credit Reference Bureau Africa Limited and Metropol Credit
Reference Bureau Limited) had received a total of 3.5 million credit
requests from banks, more than 53,000 requests from individual
customers and 12,851 customer inquiries prompted by adverse actions
by institutions. Revised regulations allowing for sharing of positive
and negative credit information by banks and deposit-taking
microfinance institutions were gazetted in January 20142.
Implementation of policies to enhance the stability of the financial
system. Attention has been focused on the deposit-taking institutions,
which account for the largest proportion of the assets in the system.
Oversight of insurance, pension and other investment funds had also
been strengthened with all the regulators adopting a risk-based
approach to the supervision of institutions/entities under their

According to the CBK, Credit Reference Bureaus (CRBs) complement the central role played by banks and other
financial institutions in extending financial services within an economy. CRBs help lenders make faster and more
accurate credit decisions. They collect, manage and disseminate customer information to lenders within a provided
regulatory framework. Credit histories not only provide necessary input for credit underwriting, but also allow
borrowers to take their credit history from one financial institution to another, thereby making lending markets
more competitive and, in the end, more affordable. CRBs assist in making credit accessible to more people, and
enabling lenders and businesses reduce risk and fraud. Sharing of information between financial institutions in
respect of customer credit behaviour, therefore, has a positive economic impact.
Other achievements were (i) progress towards the formation of the Nairobi International Financial Centre
(NIFC); and (ii) the enactment of the Anti-Money Laundering and Combating Financing of Terrorism Act
(AML/CFT Act) in 2009.

Financial regulation in Kenya: 4

2.2 Financial Sector and economic performance

A lot of work has been done on the relationship between the size of the financial
sector and economic performance. Many studies find a close linkage between
financial deepening, productivity and economic growth. It is for example estimated
that policies that would raise the M2/GDP ratio by 10% would increase the long-
term per capita growth rate by 0.20.4% points (Easterly and Levine 1997, Ndulu
and OConnell 2008). According to Levine (1997), there are five functions of the
financial system through which it enhances economic growth: reducing risk;
allocating resources; monitoring managers and exerting corporate controls;
mobilizing savings; and facilitating exchange of goods and services. The impact of
these factors on growth depends, among others, on the level of financial
intermediation; the efficiency of financial intermediation; and the composition of
financial intermediation. In the simple AK model, the financial sector promotes the
growth of the economy by raising the saving rate; the marginal productivity of
capital, and the proportion of savings that is channeled to investment. However,
while low income countries need to increase the size of their financial sectors, there
are limits to this (Spratt 2013). Beyond a certain level, estimated at around 80-
100% of private credit to GDP, financial sector development becomes negative for
economic growth, both through heightened financial instability and the
misallocation of financial resources. The same applies to a too rapid growth of
private sector credit which might lead to output volatility and adverse growth
effects (Griffith-Jones with Ewa Karwowski 2013).

Kenya has a well developed financial system for a country of its income level
(Beck and Fuchs 2004). Kenyas level of financial development is not too far off
from the predicted level in a global cross-country model (Allen et al. 2012).
Christensen (2010) classifies Kenya as a frontier market economy whose financial
market is advanced, but not to the same extent as emerging markets e.g. S. Africa,
given that its M3/GDP ratio was about 34% compared to an average of 63% for
emerging market economies in 2008-10 although these indicators have improved
over time. It is therefore unlikely the size of the Kenyas financial sector is beyond
the threshold to negatively impact on economic growth. Griffith-Jones and
Karwowski (2013) also show that credit expansion in Kenya has been relatively
modest in the last decade (at 19.5% over 2000-10) compared to other selected SSA
countries (for example Angola 1545.5%, Malawi 215.6%, Mali 286.7%, Niger
174.4%, Nigeria 173.0%, Sao Tome and Principe 709.8%, Sierra Leone 384.2%,
Sudan 505.6%, Tanzania 274.4 and Uganda 152.8%).

Two measures of the depth and coverage of financial systems is the M2/GDP and
private credit/ GDP ratios. As seen in Figure 1, while the M2/GDP ratio in Kenya
closely tracks that of low-income countries (LICs), it is far below that of middle-
income countries (MICs), with a clear divergence over time. Between 1980 and
2011, their respective ratios increased from 29.9% to 49.9% for Kenya, 16.8% to
47.2% for LICs and 32.2% to 101.6% for MICs. Figure 2 also shows a similar
pattern with respect to credit to the private sector GDP ratio, with the Kenya ratio
tending to decline from the early 1990s. Between 1980 and 2011, their respective
ratios increased from 29.5% to 37.4% for Kenya, 10.5% to 29.9% for LICs and
31.3% to 76.1% for MICs.

With the country aspiring to MIC status by 2030, it apparently has a long way to go
in building its financial sector. In its monetary programming, the CBK endeavours
to keep the path of private sector credit growth rate close to the projected nominal
GDP path. As seen in Figure 3, domestic credit to the private sector (DCP) closely
tracked the nominal GDP over 2005-2009, with acceleration in 2010-2011, which
was broadly reversed in 2012, with another acceleration in the second half of 2013.

Financial regulation in Kenya: 5

Private sector credit growth picked-up during the first half of 2013 in response to
the gradual easing of the monetary policy stance, pick-up in economic activity and,
improved investor confidence in the economy after the March 2013 elections. The
CBK reduced the Central Bank Rate (CBR) from 9.50% to 8.50% in May 2013 and
retained it at this level in the rest of 2013. Consequently, the annual growth in the
overall private sector credit rose from 12.69% in June 2013 to 21% in December
2013, above the projected growth path of 16.2% in the year to December 2013.

Figure 1: M2 as % of GDP in Kenya versus LICs and MICs







1980 1985 1990 1995 2000 2005 2010


Source: World Bank, World Development Indicators

Figure 2: Domestic credit to the private sector as % of GDP in

Kenya versus LICs and MICs










1980 1985 1990 1995 2000 2005 2010


Source: World Bank, World Development Indicators

Financial regulation in Kenya: 6

Figure 3: Domestic credit to the private sector (DCP) and
nominal GDP in Kenya, 2005:12 2013:12

1,600 4,500

1,400 4,000

1,200 3,500

1,000 3,000

800 2,500

600 2,000

400 1,500

200 1,000
2006 2007 2008 2009 2010 2011 2012 2013


Source: Central Bank of Kenya

The Kenya National Bureau of Statistics (KNBS) provides quarterly GDP and
growth data since 2000. Figure 4 shows four-period moving average growth rates in
financial intermediation and GDP in Kenya over 2001Q1-2013Q3. There is clearly
some correlation (0.24) between the two series during the study period, with the
moving average quarterly GDP growth rate generally less volatile than growth in
financial intermediation (standard deviation of 0.660 versus 1.465, respectively).
Granger causality tests show significant causality from financial intermediation to
growth at 3 and 4 lags at the 5% level, with the other lags non-significant (Table 1),
supporting Kenya Vision 2030 designation of the financial sector as one of the
drivers of growth in Kenya, at least in the short-run4. On an annual basis, the
financial sector growth has consistently outpaced the real GDP growth since 2009.

Figure 4: Quarterly Growth in Financial Intermediation and GDP

in Kenya, 2001Q1-2013Q3




2000 2002 2004 2006 2008 2010 2012

Quarterly growth in Financial Intermediation

Quarterly growth in GDP

Source: Kenya National Bureau of Statistics (

In contrast, the KNBS reports growth data on a quarter-on-quarter basis to remove the seasonal effects. By
ignoring the intermediate values, none of the Granger causality tests are significant, although there is more
correlation in the two series (0.28).

Financial regulation in Kenya: 7

Table 1: Ganger-causality between quarterly growth in financial
intermediation (QGFI) and growth in GDP (QGGDP)

3 lags 4 lags

F-Statistic Prob. F-Statistic Prob.

QGGDP does not Granger Cause QGFI 0.867 0.466 1.426 0.244

QGFI does not Granger Cause QGGDP 2.809 0.050 2.751 0.042

In Kenya, the Second MTP identifies the following emerging issues and challenges:
(i) inadequate access to finance for SMEs; (ii) high bank lending rates and wide
interest rate spreads; (iii) high level of exclusion from financial services; and (iv)
low insurance penetration and pension coverage. We address the first three
challenges later in the paper.

Financial regulation in Kenya: 8

3 The roles of foreign
banks, state-state owned
banks and DFIs in Kenya

3.1 Foreign and state-owned commercial banks

According to the framework papers for the project (Spratt 2013, Griffith-Jones with
Ewa Karwowski 2013), opinion on the merits of foreign banks and state-owned
banks has shifted considerably since the 2007-8 GFC. Foreign banks can have both
positive and negative effects. While they can bring valuable skills, technology and
capital, they can also bring risks. Evidence from the recent financial crisis shows
that countries where foreign banks dominate the market could suffer negative
lending shocks, as turmoil in the home markets cause parent banks to withdraw
capital from the developing countries where they operate. They can have negative
impacts, particularly by bypassing the supply of credit to the less lucrative sections
of the country. Critics of foreign bank participation therefore argue that foreign
banks may have an overall negative effect on financial deepening and inclusion
(Beck 2013). Distance constraints and informational disadvantages may prevent
foreign banks from lending to SMEs. The competitive advantage of foreign banks
can result in domestic banks being crowded out of the market and foreign banks
focusing on the top-end of the market, thus leaving SMEs and poorer households
without access to financial services. Specifically, the greater reliance of foreign
banks on hard information about borrowers as opposed to soft information can have
negative repercussions for riskier borrowers if foreign banks crowd-out domestic
banks. The existing empirical literature has not provided unambiguous findings on
the repercussions of foreign banks for financial development and inclusion and
neither has the African experience (Beck 2013).

Similarly, there has been a change in the negative perception of state-owned

commercial banks, with the some studies finding that these banks performed a
valuable counter-cyclical role in some countries; while others find them to be
associated with higher rates of economic growth (Spratt 2013). The challenge
therefore is to design and regulate them so that they can successfully fulfill their
development mandate, while avoiding the well-documented failures of the past.

Kenya currently (in December 2013) has 43 banks, with 1,313 branches and 34,064
employees, accounting for about two thirds of the financial systems assets. In
terms of shareholding, the Central Bank identifies 14 banks with foreign
ownership, accounting for 32.2% of net assets in 2012. The Central Bank also
identifies 6 banks with state ownership accounting for 24.8.2% of net assets in
2012, with the government having majority ownership in three of these, which
account for 4.2% of net total assets (Consolidated Bank; Development Bank of
Kenya; and the National Bank of Kenya) 5. The remaining 23 are local private

The other three banks are CFC Stanbic, Housing Finance; and Kenya Commercial Bank.

Financial regulation in Kenya: 9

banks, accounting for 43.0% of the banking sectors net assets. Hence Kenyas
banking system is dominated by local private banks and foreign banks.

We therefore study the relative performance of the 14 foreign banks and the 6
banks with state ownership versus the local private banks in the country.
Specifically, this section addresses the following research issues:

How well have foreign banks and banks with state ownership
performed, for example, in terms of financial indicators, such as
ROAs, NPLs, etc, but also in terms of economic indicators, such as
providing access to credit to SMEs, as well as other parts of the private
What are the key challenges of regulating Kenya banks in other
countries? Foreign banks in Kenya are treated symmetrically with the
other banks in the country.
Oloo (2013) proposes a number of indicators to identify the different strengths and
weaknesses of Kenyan banks and provides data on individual banks, which we
aggregate into the various ownership components, weighted by the value of assets
in 2012. These include the rates of return on assets and capital; cost of funds,
efficiency ratio and the ratio of non-performing loans (see Table 2).

Table 2: The performance of commercial banks in Kenya by


Foreign banks Banks with state- Banks with majority state Local private All banks
ownership ownership banks

Return on assets, %6
2009 3.6 2.8 3.7 3.8 3.6
2010 4.7 3.7 4.2 4.8 4.6
2011 4.7 4.1 3.1 4.8 4.7
2012 5.2 4.1 1.4 4.8 4.9
Return on capital, %7
2009 36.7 30.0 27.2 30.3 32.3
2010 46.1 23.4 30.8 46.6 40.7
2011 50.6 44.9 27.6 50.4 49.1
2012 51.9 38.0 12.7 50.9 48.0
Average cost of funds, %
2009 3.0 2.7 3.5 4.0 3.4
2010 2.2 2.1 2.9 3.4 2.7
2011 2.5 2.3 3.8 3.8 3.0
2012 4.9 5.3 7.6 7.0 6.0

Return on assets (ROA) is the ratio of profits before tax to average total assets (at beginning and end of the year).
A higher ratio is desirable.
Return on capital (ROC) is measured as the return to the average core capital (at the beginning and end of the
year). A higher ratio is desirable.

The ability of a bank to acquire external funding cheaply to boost its investments is a critical measure. There are
two main sources of funds for the bank: (a) deposits from customers; and (b) borrowed funds. This ratio therefore
is a measure of how cheaply, or expensively these funds have been acquired: it reflects the ease with which a bank
is able to secure such funds. A lower rate is desirable.

Financial regulation in Kenya: 10

Efficiency ratio, %9
2009 53.1 66.4 64.4 58.8 60.0
2010 47.1 61.4 58.0 51.6 53.6
2011 45.8 56.8 63.1 51.6 52.0
2012 50.7 57.0 74.8 52.0 53.9
Non-performing loans to advances ratio, %
2009 4.5 9.7 10.1 6.4 6.7
2010 3.6 6.4 6.6 5.1 5.0
2011 2.7 4.4 6.5 3.7 3.6
2012 2.4 5.2 8.8 3.6 3.7
The foreign banks have done as well as local private banks with both having an
average rate of return on assets of 4.6% over 2009-2012, ahead of banks with state
ownership (3.7%) and state-owned banks (3.1%). The poor performance of the
latter is attributed to poor legacy in the past of poor governance and massive
interference by the state in their management.

The same pattern is repeated in the other indicators. Foreign banks have on average
done slightly better on the rate of return on core capital (46.3%) over 2009-2012
when compared to local private banks (44.6%), ahead of banks with state
ownership (34.1% and 24.6%, respectively). They also have the lowest cost of
funds (index of 3.2%) together with banks with state ownership (index of 3.1% and
4.5%, respectively) and local private banks (index 4.6%). Foreign banks are also
the most efficient (with an average score of 49.1%) slightly ahead of local private
banks (score of 53.5%), with banks with state ownership the least efficient (scores
of 60.4% and 65.1%, respectively). Finally, foreign banks have the least non-
performing loans ratio (average 3.3% over 2009-2012), followed by local private
banks (4.7%) and banks with state ownership (6.4% and 8.0%, respectively).

It is therefore apparent that foreign banks largely behave like local private banks,
except that they have cheaper sources of finance due to their reputation capital.
They are also very diverse so that it is difficult to generalize their behavior. They
include for example (i) the traditional multinational banks from Europe and USA
(Barclays, Citibank, Habib A.Z. Zurich and Standard)11; (ii) banks from Asia and
the Middle East (Bank of Baroda, Bank of India, Gulf African Bank, Habib Bank
and Diamond Trust Bank, the last two from Pakistan and owned by the Aga Khan
Fund for Economic Development); (iii) pan-African banks (Bank of Africa, United
Bank of Africa; and Ecobank); and (iv) Islamic banks (First Community Bank
licensed in 2007 with some shareholding from Tanzania and Gulf African Bank
licensed in 2008). K-Rep Bank was incorporated as a commercial bank in 1999,
from microfinance NGO and has largely maintained the microfinance banking

According to World Bank (2013), most foreign banks have dedicated units serving
SMEs. There are however a few exceptions such as Citibank and, to a less extent,
Standard that focus on corporate and high-end clients, and hence do not lend to
SMEs. Oloo (2013) simulates the cost of provision of banking services to SMEs
from customers perspective. In the first scenario, he considers a small business

The efficiency ratio is measured by taking the total operating expenses, which include the banks overheads and
weighting them against the total operating income. A lower ratio is desirable.
Non-performing loans is the single most important threat that a bank can face. To assess its magnitude, it is
weighted against the total portfolio of all loans and advances that the bank has extended. A high ratio is a reflection
of imprudent lending practice and poor credit management. A low ratio is therefore desirable.

Barclays and Standard have been in the country for more than 90 years.

Financial regulation in Kenya: 11

firm, with a turnover of about US$ 60,000 (at the exchange rate on Ksh 84.5 per US
dollar in 2012). He assumes the annual cost of opening and maintaining a business
current account to require 6 50-leaf cheque books, 48 customer withdraws, 48
bankers cheques, 24 standing orders, charges for 600 transactions and 12 ledger
fees. In the second scenario, he considers a medium-sized business enterprise with
a turnover of about US $ 6 million per year. He assumes the annual cost of opening
and maintaining a business current account requires 12 50-leaf cheque books, 96
customers withdraws, 96 bankers cheques, 96 standing orders, charges for 6,000
transactions and 12 ledger fees.

He derives the following total costs of operating the accounts by type of bank
ownership. The results show that local private banks have the lowest costs to
SMEs, followed by foreign banks and then banks with state ownership.

Table 3: Simulated cost of banking services to SMES

Small firm, US$ Medium-sized firm, US$

Foreign banks 795.5 1751.0

Banks with state ownership 825.3 1902.0

Banks with majority state ownership 800.6 1835.5

Local private banks 733.0 1667.7

3.2 Challenges of Regulating Kenya banks in other countries

In Kenya, some banks have expanded their branch networks in the region. By
December 2012, Kenyan banks had established 282 branches in neighbouring
countries (Uganda 125, Tanzania 70, Rwanda 51, Burundi 5, and South Sudan 31).
Such banks pose an increasing challenge for regulators across Africa (Beck 2013).
Financial integration implies that the negative externality costs of bank failure go
beyond national borders that are not taken into account by national regulators and
supervisors. Close cooperation that can help internalize these cross-border
externalities, although the institutional extent of such cooperation should be a
function of the strength of externalities but also the heterogeneity of countries
legal and regulatory frameworks.

Central banks in Eastern African countries have, for example, signed a

Memorandum of Understanding (MOU) to facilitate information sharing and
supervisory co-operation for regional banking groups. The CBK has developed and
implemented a consolidated supervision program for the effective oversight of
banking groups. As part of efforts aimed at implementing consolidated supervision,
it launched Prudential Guidelines on Consolidated Supervision and convened two
Supervisory College meetings in 2012 and 2013 bringing together all Central
Banks of the East African countries where Kenyan banks currently have operations.
The introduction of guidelines on Country and Transfer Risk, Risk-based
Supervision and Consolidated Supervision is timely given the increasing cross
border risks faced by the Kenyan banks as they expand regionally 12. The East
African Central Banks are also currently working to harmonise their banking sector
supervisory rules and practices as a prerequisite for the envisaged East African
Monetary Union (EAMU). The recently established Committee of African Bank
Interview with CBK Governor in Oloo (2013).

Financial regulation in Kenya: 12

Supervisors as part of the African Association of Central Banks can give this
cooperation further impetus, by enabling informal exchange of information and
experiences and networking possibilities (Beck 2013).

Two issues appear critical in this increasing regulatory cooperation (Beck 2013).
First, based on the experience of European countries, there should be a focus on
proper preparation for resolution. Non-binding MOUs and Colleges of Supervisors
limited to information exchange are of limited use in times of bank failure. Second,
it is important not to ignore development benefits of foreign banks when
considering them as potential source of fragility. Financial stability is not an
objective in itself, but rather a necessary condition for sustainable financial
deepening, with the main goals of economic development and poverty alleviation.

3.3 3.4: Development Finance Institutions13

It has long known that commercial banks will under-supply long-term finance, and
under-serve key sectors, such as agriculture or small and medium enterprises
(SMEs), and that these market failures are more acute in LICs (Spratt 2013).
Although DFIs are an obvious solution, they were widely seen as inefficient,
ineffective and corrupt so that the cure was thought worse than the disease. This
perception has shifted significantly since the recent financial crisis, where some
countries with significant DFIs saw them fill the gap left by the commercial banks.
The success of DFIs in countries as diverse as Brazil, South Africa and Germany
has shown it is possible to avoid many pitfalls.

Is there a need for a greater role for DFIs in Kenya, to cover gaps in financing in
key sectors, essential for inclusive growth, as in Asia (Hosono 2013)? What are
experiences of DFIs in Kenya? How can good DFIs be expanded /created, taking
into account issues of incentives and governance?

There is no doubt that DFIs in Kenya could play a significant role in the financial
sector by providing long-term finance (CBK 2013). Targeted interventions for
specific sectors or groups like SMEs, youth, women, and so on would best be
served by DFIs. This is recognized under Vision 2030, where DFIs are expected to
contribute towards enhanced financial access and investment goals. For DFIs to
play this role and fulfill market expectations, they require enhanced capacity with
clear ground rules and enhanced finance allocation. In Kenya, DFIs are under the
purview of the National Treasury. But the sector remains small. The five existing
DFIs account for less that 1% of the assets of the banking sector and had lent only
Ksh.6.8 billion (approximately USD80.73 million) as of June 2012 when compared
to Ksh 1,224.11 billion (approximately USD 14.53 billion) of credit to the private
sector from the countys banking sector (CBK 2013). Hence these DFIs supplied
only about 0.56% of the banking sector credit to the private sector.

According to CBK (2013), some Kenyan DFIs converted to commercial banks in

the1990s (e.g. DFCK to DBK Ltd) in order to mobilize deposits. But the journey
was not smooth due to their inability to comply with the prudential. As a result,
they experienced high non-performing loans and high concentration risk due to
dependence on a few borrowers. The DFIs-turned-banks non-compliance with the
prudential requirements could have mainly been driven by the conflict of their
primary mandate of long term lending and the banking regulatory framework which
is applicable to all banks irrespective of their circumstances. They were therefore
unsuccessful in mobilizing long term local deposits to match their assets profile.

The five existing DFIs service industry and commerce (IDB Capital, Kenya Industrial Estates and Industrial and
Commercial Development Corporation); agriculture (Agricultural Finance Corporation); and tourism (Kenya
Tourist Development Corporation).

Financial regulation in Kenya: 13

The failure of DFIs to customize their policies and practices towards commercial
bank orientation was compounded by weak corporate governance structures arising
from operating for a long period without prudential guidelines.

According to a Presidential Task Force on Parastatals Reform (2013), the role of

DFIs has atrophied since the mid-1980s which the Task Force attributes to DFIs
inability to respond successfully to the change to a liberal policy regime in the
1980s and 1990s; narrow credit focus and limited sources of financing from donors
and government; as well as poor governance in part due to state interference,
coupled with ineffective management and low staff morale. The Task Force
therefore advocates consolidating DFIs under a Kenya Development Bank (KDB)
with sufficient scale, scope and resources to place a catalytic role in Kenyas
economic development by providing long-term finance and other financial and
advisory, investment and advisory services. CBK (2013) as well calls for
introduction of prudential regulation and supervision consistent with their mandate
(for example the AADFI standards of the Association of African DFIs) as done in
several countries including Tanzania, Nigeria, China, Swaziland and Korea which
already regulate and supervise DFIs. As a result, Kenya would only customize the
regulatory and supervisory frameworks to local circumstances. An effective
regulatory and supervisory framework should adequately address the potential risks
faced by DFIs by tailoring them to suit their unique features, especially the tradeoff
between the focus on economic development orientation and long term structure of
assets. Regulation and supervision must also continuously evolve to keep pace with

Financial regulation in Kenya: 14

4 Financial inclusion in

4.1 Trends and patterns of financial inclusion

The envisaged targets of the financial sector under Vision 2030 included enhancing
financial inclusion by decreasing the share of population without access to financial
services by about 20%. Financial inclusion in Kenya has been monitored through
financial access surveys of which three so far have been conducted: in 2006, 2009
and 2013. These surveys reveal that Kenyas financial inclusion landscape has
undergone considerable change. The proportion of the adult population using
different forms of formal financial services has increased from 27.4% in 2006, to
41.3% in 2009 and stood at 66.7% in 2013, amongst the highest in Africa
(Table 4)14. In addition, the proportion of those accessing informal financial
services has declined substantially from 33.3% in 2006 to 27.2% in 2009 and to
only 7.8% in 201315. Overall, the proportion of the adult population totally excluded
from financial services has declined from 39.3% in 2006 to 31.4% in 2009 and to
25.4% in 2013. With a decline of 35% between 2006 and 2013, this has
substantially exceeded Vision 2030s expectations.

Table 4: Financial inclusion and exclusion in Kenya, %

2006 2009 2013

Formal 27.4 41.3 66.7

Informal 33.3 27.2 7.8

Excluded 39.3 31.4 25.4

Source: Financial Sector Deepening Kenya (2013)

The last half decade has therefore seen a massive increase in access to financial
services in the country. Deposit accounts have, for example, increased from about 2
million to 18 million while loan accounts have increased from 1 to 3 million since
200716. This is reflected in Table 5 which shows a substantial increase in the use of
bank services, from 13.5% in 2006, to 17.1% in 2009 and to 29.2% in 2013.
However, the most dramatic increase is usage of mobile money services from
virtually 0% in 2006 to 28.4% in 2009 to 61.6% in 2013. The rapid growth of

Formal financial institutions are defined broadly to include commercial banks, deposit-taking microfinance
institutions (DTMs), foreign exchange bureau, capital markets, insurance providers, deposit-taking SACCOs
(DTSs), mobile phone financial service providers (MFSP), Postbank, NSSF, NHIF, credit-only MFIs, credit-only
SACCOS, hire purchase companies and the government.

The informal financial sector includes informal groups, shopkeepers and merchants, employers, and money
lenders who are all unregulated under structured law provisions.
Interview with the Governor, Central Bank of Kenya. EastAfrican, August 24-30, 2013.

Financial regulation in Kenya: 15

mobile money banking services shows its ability to overcome problems of physical
access and high relative costs (Spratt 2013). Mobile banking has introduced
alternative channels at financial service provision to conventional banking and has
provided clear, quick and convenient platforms to conduct a range of financial
transactions. The adoption of mobile money service M-PESA in 2007 far exceeded
expectations. Currently, the four mobile money services (M-PESA, Airtel Money,
YuCash and Orange Money) have close to 20 million customers, handling over
US$ 54.4 million worth of transactions per day. M-PESA however remains
dominant with 82% of market share, Airtel Money 15%, YuCash 2% and Orange
Money 1%.

Table 5: Overall Use of financial services, %

Usage of: 2006 2009 2013

Banks 13.5 17.1 29.2

SACCOs 13.5 9.3 9.1

Microfinance 1.8 3.5 3.5


Informal groups 39.1 29.5 27.7

Mobile money 0.0 28.4 61.6

financial services
Source: ibid.

Financial inclusion has varied with the socio-economic statues of the population.
According to FSDK (2013), financial exclusion in 2013 varied from 55.3% for the
poorest 20% of the population to 5.7% for the wealthiest 20% of the population. As
well, financial exclusion was highest for those without any education (60.7%) and
lowest for those with tertiary education (1.8%). Table 6 shows that women use of
formal financial services has lagged behind that of men, but the gap substantially
reduced between 2009 and 2013, while exclusive use of informal financial services
have declined for both men and women. Similarly, Table 7 show that rural areas
have lagged behind urban areas in access to financial services

Table 6: Financial inclusion in Kenya by gender

2006 2009 2013

Male Female Male Female Male Female

Formal 34.3 21.0 49.3 34.4 71.1 62.7


Informal 27.0 39.2 19.8 33.9 4.7 10.8

Excluded 38.7 39.8 31.0 31.8 24.2 26.6

Source: ibid.

Financial regulation in Kenya: 16

Table 7: Financial inclusion in Kenya by location

2006 2009 2013

Rural Urban Rural Urban Rural Urban

Formal 24.6 35.7 35.5 63.1 59.6 80


Informal 37.0 22.2 30.0 16.8 9.8 4.4

Excluded 38.4 42.0 34.5 20.1 30.6 15.6

Source: ibid.

On use of financial services by business owners, 36.8% used banks; 4.2%

SACCOs, 7.8% microfinance institutions, 33.5% informal groups and 72.6%
mobile money financial services, exhibiting the same pattern as for the wider

4.2 Is regulation of M-PESA and other mobile money platforms


The success of M-PESA in Kenya is often used to argue for a light-touch approach,
where mobile banking was allowed to flourish (Spratt 2013). Possible systemic and
individual users risks seem to require careful evaluation, however. It is clearly
important to enable, rather than stifle, innovation but it is also clear that regulation
should be comprehensive in the longer term. How to strike the right balance here is
an important area of research.

In responding to this question, the CBK admits that the technology used to deliver
the mobile money services carries inherent threats, the main ones being operational
risk, financial fraud and money laundering 17. However, prior to the launch of
mobile banking services by the various companies, the CBK requires them to
provide a detailed risk assessment, outlining all potential risks and satisfactory
mitigating measures they have put in place. In the case of M-PESA, Safaricom
sought authorization from CBK to undertake the money transfer business. In
evaluating the proposal, the CBK considered the request on the basis of safety,
reliability and efficiency of the service. In addition, precautionary measures were
put in place to ensure that the service did not infringe upon the banking services
regulatory framework as provided for in the Banking Act. Following the enactment
of the National Payments System Act in 2011, the CBK now has the oversight
mandate of the National Payments System. All payment service providers including
mobile phone service providers offering money transfer services fall under the
CBKs regulatory framework18.

The Kenya Bankers Association (KBA) has however complained that the Mobile
Network Operators (MNOs) offer services similar to those offered by banks, yet

Interview with the CBK Governor in Oloo (2013). This section draws on this interview.
According to the December 2013 Monetary Policy Statement, The CBK will continue to support development
of new products and innovations towards enhancing financial access in order to encourage economic growth. In
this regard it will continue to propose suitable legislation aimed at ensuring that such innovations are regulated
accordingly to enhance market confidence. The Bank will also continue to monitor any new financial derivatives
and /or innovations in the market that could have adverse effects on market stability.

Financial regulation in Kenya: 17

they are not subject to similar regulations19. KBA argues that there is a blurred line
between what constitutes taking deposits from customers as done by MNOs and
taking deposits for savings as done by banks. The e-float for example which is kept
in special accounts in banks by MNOs is not subject to banking regulations such as
subjecting them to deposit insurance, undermining the security of such deposits in
case of a bank failure or financial crisis. The response by MNOs is that the
possibility of such risks making a huge impact on clients is very rare as the e-float
is relatively small, and it is distributed across several banks.

4.3 Financial inclusion (innovations) and macroeconomic

stability in Kenya

Increased financial inclusion through financial innovations does not seem to have
compromised financial stability. First, the stock of e-money is backed 100% by
accounts held at commercial banks. The mobile money e-float is also a small
proportion of the other monetary aggregates in terms of size for it to matter much
for monetary policy. Weil et al. (2011) estimate the outstanding stock of M-PESA
e-float at 1.6% of M0 and 0.4% of M1.
Second, while there has been increased instability in monetary relationships post-
2007, reflected in a decline in the income velocity of circulation and an increase in
the money multiplier undermining the conduct of monetary policy which assumes
stable monetary relationships, stability seems to have been re-established since
2010. The instability was therefore a temporary phenomenon. Velocity which is the
ratio of nominal GDP to money supply (M3X) declined significantly from a
monthly average of 2.50 in 2006 to 2.09 in 2010 and stabilized at that level
thereafter. Similarly, the money multiplier increased from a monthly average of
5.49 in 2006 to 5.96 in 2010 and stabilized at that level. The demand for money
also shows stability post-2010 (Weil et al. 2011).

Figure 5: The money multiplier and income velocity in Kenya,

December 2005 to December 2013

7.6 2.6

7.2 2.5

6.8 2.4

6.4 2.3

6.0 2.2

5.6 2.1

5.2 2.0

4.8 1.9
2006 2007 2008 2009 2010 2011 2012 2013


Source: Central Bank of Kenya

See the Daily Nation, January 26, 2014, Banks revive battle with money service providers

Financial regulation in Kenya: 18

5 Access and cost of
credit in Kenya

5.1 Introduction

This section looks at access to finance, where the key problem is how to provide
financial access that is both affordable and suited to the needs of poor people
(Spratt 2013). On this, the costs of providing basic banking services are often
prohibitive, and credit is either unavailable or too expensive. The reasons are well
understood: providing physical access in rural areas is inherently expensive, and
providing financial services for people with few financial resources entails high
relative costs; a lack of credit history and collateral is a key constraint on extending
credit, and small loan sizes also mean high transaction costs. Extending financial
access thus tends to be unattractive for banks in LICs. Although microfinance
institutions (MFIs) have partially filled this gap, their record is mixed.
Kenyas financial sector has undergone reforms since the late 1980s aimed at
achieving (i) stability so as to ensure that banks and other financial institutions
taking deposits can safely handle the publics savings and ensure that the chances
of a financial crisis are kept to a minimum; (ii) efficiency in the delivery of credit
and other financial services to ensure that the costs of services become increasingly
affordable and that the range and quality of services better caters to the needs of
both savers and investing businesses; and (iii) improved access to financial
services and products for a much larger number of Kenyan households (Nyaoma
2006). The country formally adopted financial sector forms in 1989, supported by a
$170 million World Bank adjustment credit. Financial reform proposals were first
incorporated in the 198690 structural adjustment program. The main features of
the program included: (i) interest rate liberalization which was achieved in July
1991; (ii) liberalization of the treasury bills market in November 1990 which was
accompanied by introduction of the treasury bonds of long-term maturities - one,
two and five-year maturities; (iii) setting up a Capital Markets Authority in 1989 to
oversee the development of the equities market; (iv) abolition of credit guidelines
in December 1993 (which were in existence since 1975 in favour of agriculture);
and (v) improving and rationalizing the operations and finances of the DFIs.

Financial sector reforms have undoubtedly strengthened Kenyas banking sector in

the last decade or so, in terms of product offerings and service quality, stability and
profitability (Kamau 2009). Major indices show an improvement, including: (a) the
capital adequacy ratio; (b) rates of return on assets (ROA); (c) non-performing
loans; (d) growth and composition of credit to the private sector; and (e)
composition of banks assets and liabilities20.

Assets of the banking system in Kenya are dominated by loans and advances, government securities and cash
reserves at CBK. Kenya commercial banks hold minimal derivatives or asset-based securities in their portfolios.
They mainly hold risk-free government securities.

Financial regulation in Kenya: 19

5.2 Banks lending to SMEs in Kenya

The World Bank (2013) devotes itself to this issue. The report notes that although
retail banking has improved markedly in Kenya in the last decade, access to credit
for SMEs is still limited, with SMEs accounting for about 90% of all enterprises in
the country, according to the Kenya Private Sector Development Strategy 2006-10.
SMEs are provided with financial services by a range of institutions, including
banks, non-bank financial institutions, savings and credit cooperatives (SACCOs),
and microfinance institutions. The report cites an analysis of firms that made it to
the 2013 Top 100 mid-sized companies survey that showed that the number of
SMEs that turned to lenders for credit lines and overdrafts increased to 67%
compared to 57% in 2012. Most of the surveyed entrepreneurs cited the high cost of
credit as the reason for cash flow challenges they face, leaving them with no
recourse but to dig deeper into their personal savings or turn to family friends to
raise funds for day to day operations.

The report notes there is some evidence that Kenyan banks are actually ahead of
their counterparts in Nigeria and South Africa in lending to SMEs. From field
surveys, about 17.4% of total bank lending goes to SMEs in Kenya, compared to
only 5% in Nigeria, and 8% in South Africa. Kenyas ratio is comparable to that of
Rwanda, which is a smaller market with a relatively small presence of large-scale
firms (Aziz and Berg 2012). These numbers are supported by the innovations in the
banking sector that suggest a strong appetite for SME lending.

According to a survey reported in the World Bank (2013), involvement of Kenyan

banks in the SME segment is growing, both in terms of size and the diversity of
their approaches to the SME client relationship. This has been driven by
innovations specifically targeting this market. These innovations started through
microfinance-rooted institutions scaling up to becoming commercial banks and now
include innovations with lending models and technology in the retail banking
segment by other institutions, most notably Equity Bank. In addition, in Kenya
there are active markets for hire-purchase and invoice-discounting mainly provided
by banks to SMEs which deliver to government and larger enterprises with
reputable payment histories. Adoption of these instruments has facilitated entry by
some mid-sized Kenyan banks into the SME financing.
According to this survey, Kenyan banks tend to provide more working capital than
investment loans. Demand factors play a role with Kenyan firms citing working
capital shortages as the primary reason for approaching banks. The distribution of
loans may also reflect an assessment by banks that long-term loans are too risky.
According to the banks interviewed, on average it takes 190 days to recover bad
loans in Kenya, with a rate of recovery of about 80%; the cost is about 40% of the
amount of the loan. The situation seems somewhat better in Rwanda, where on
average it takes 135 days to recover loans, the rate of recovery is about 85%, and
the cost is about 10% of the loan. Nigerian banks operate in the most difficult
environment: on average they need 246 days to recover a loan and are able to
recover on 30% of the loan.
According to the World Bank (2013), most Kenyan banks have dedicated units
serving SMEs. At most institutions, however, the unit is a subunit of the retail
banking unit rather than a division. Products are largely standardized, although the
number of banks such as Equity Bank, Cooperative Bank and K-Rep Bank are
producing customized loan products for the SME sector. Some banks provide
training to their clients to improve their management skills and financial reporting.
Lending remains based on collateral. Risk management is increasingly automated,

Financial regulation in Kenya: 20

although domestically owned banks have not yet embraced the use of scoring and
risk-rating technologies on a large scale.

The report notes that banks prefer to engage with formal firms rather than with
informal or semi-informal firms. As part of the loan application, they require SMEs
to provide a variety of documents certifying their compliance with government
regulations and providing details about their finances. The most common
documents required include the registration certificate from the Business Registrar
(Attorney Generals Office); the Single Business Permit, obtained from the City
Councils: and sometimes the certificate of compliance from the Kenya Revenue
Authority. These filing requirements are quite onerous and often discourage SMEs
from seeking bank financing.

According to World Bank (2013), donors have been encouraging banks to engage
in SME financing, providing bank-specific lines of credit and partial credit
guarantees. Donors prefer this bank-specific approach to establishing schemes that
are open to all qualified institutions, although a more open approach would be
better suited to encouraging competition. In markets where SME financing is in its
infancy, schemes can augment banks willingness to push the frontier and
demonstrate that lending to SMEs can be a viable and profitable business line.
U5AID reportedly operates the largest credit guarantee scheme in Kenya, a US$70
million program. ARIZ, a risk-sharing program funded by the African
Development Bank, guarantees 50% of all loans in the portfolio. Other donors that
are encouraging lending to SMEs include the European Investment Bank, Proparco,
FMO, DEG, SIDA, KfW, Norlund, and the China Development Bank.

On policy, the report recommends that tapping the full growth and job-creating
potential of the SME sector will entail a move towards providing growth capital
and not just working capital. A growing number of private equity providers are
active in East Africa in general and in Kenya in particular. Most of them are not
interested in SMEs. A number of new entrants cite lack of information and
expertise as a deterrent to venturing into this market. Technical assistance could
help bridge the distance between the demand for and the supply of private equity.

Improving the listability of SMEs as well could increase their access to equity
finance. Kenyan SMEs have shown some interest in tapping equity financing to
grow, by turning to the growing number of private equity funds or by issuing shares
on the stock market. In fact, about 28% of firms surveyed in the Top 100 Mid-Sized
Companies said they were considering listing on the Nairobi exchange, which now
has a special segment, the Growth Enterprise Market Segment (GEMS) for SMEs.

5.3 Cost of credit and interest rate spreads in Kenya

One of the key criticisms of the Kenyan banking sector is that the cost of credit and
the interest rate spread remains high. This has raised concerns from government,
regulators and parliament, with the latter trying severally to introduce legislation to
control them. As seen in Figure 6, the interest rate spread was fairly stable,
although gradually increasing, between January 2005 and October 2011, averaging
9.56%. It jumped to a peak of 13.05% in December 2011 following a decision by
the Central Bank of Kenya to raise the policy Central Bank Rate (CBR) from 11%
to 16.5% in November 2011 and to 18% in December 2011 where it stayed until
June 2012. As a consequence, both deposit and lending rates rose sharply as the
CBK attempted to control inflation and stem currency depreciation. As seen in the
figure, the increase in the spread was because banks raised the lending rate more
than the deposit rate. The spread subsequently gradually decreased as the central

Financial regulation in Kenya: 21

bank has relaxed monetary policy, lowering the CBR from 18% to 9.5% during
January-April 2013 and to 8.50% since May 2013.

At an average of 10.02% over 2005-13, the interest rate spread has therefore
remained high despite improved economic conditions in the country. According to
the critics of commercial banks, there have been many developments that have
taken place in the country that should have significantly reduced the spread (Oloo
2013). These include (i) improvements in technology (ATMs, mobile phones, etc)
that have reduced the cost of doing business, and the need for human resource
requirements; (ii) agency banking, with 16,000 agents that are now available to
banks at nominal cost; and (iii) introduction of credit reference bureau to reduce
information asymmetries and risk. As well, the opening of Currency Centres across
the country has reduced costs associated with transporting cash for the banks.

The spread between the lending rate and the risk free 91-days Treasury bill rate is
also high and more volatile at an average of 7.43% over 2005-13 (Figure 7). This
spread can be taken as a measure of the risk premium faced by banks. It captures
perceived risk by lenders of borrowers ability to pay; as well as inefficiency in the
banking system. It has however declined since the mid-2011 denoting a decline in
the risk premium. The collapse of the 91-days TBR in 2005 was due to a reduction
of the required cash ratio from 10% to 6% in 2003 which injected a lot of liquidity
into the economy, drastically lowing interest rates.

Table 8 compares interest rate spreads in Kenya vis a vis a few selected comparator
countries over 2000-2012. The spreads are on average relatively higher in Kenya
than in Malaysia, Botswana, South Africa, Nigeria and Tanzania. They are only on
average higher in Uganda. The high spread in Kenya may reflect the comparably
higher lending by Kenyan banks to SMEs that are perceived to have a higher risk

Alongside high lending interest rates and wide spreads, the banking sector profits
have increased over time. Profits before tax increased from about US$ 70 million in
2002 to US$ 1,256 million in 2012, an average growth rate of 38.7%. The major
sources of income were interest on loans and advances (average of 49.6% of total
income during the period) which increased over time reflecting an increase in the
spread; and fees and commissions (14.6%), and government securities (19.8%)
which declined during the period (Figure 8). As also seen earlier in Table 2, the
banking sector experienced a general improvement in performance indices over
2009-2012, although there were some setbacks in 2012 with respect to the average
return in core capital, average cost of funds, the efficiency ratio and non-
performing loans ratio due to an adverse macroeconomic environment in late-2011.

Financial regulation in Kenya: 22

Figure 6: The interest rates spread in Kenya (ex post lending
minus deposit rate)





05 06 07 08 09 10 11 12 13


Source: Central Bank of Kenya

Figure 7: The interest rates spread in Kenya (ex post lending

rates minus the 91 days TBR)





05 06 07 08 09 10 11 12 13


Source: Central Bank of Kenya

Table 8: Comparative Analysis of Commercial Banks Ex Post

Spreads in Kenya and Selected Countries (%)

Year Malaysia Botswana South Nigeria Kenya Tanzania Uganda


2000 4.31 6.06 5.30 9.58 14.24 14.19 13.08

2001 3.75 5.66 4.40 8.18 13.03 15.25 14.19

2002 3.32 5.75 4.98 8.10 12.97 13.11 13.53

2003 3.23 6.45 5.20 6.50 12.44 11.47 9.09

Financial regulation in Kenya: 23

2004 3.05 5.90 4.74 5.48 10.10 9.94 12.86

2005 2.95 6.48 4.58 7.42 7.80 10.52 10.85

2006 3.34 7.59 4.03 7.16 8.50 8.93 9.61

2007 3.24 7.60 4.01 6.65 8.18 7.39 9.84

2008 2.95 7.87 3.51 3.51 8.71 6.73 9.78

2009 3.00 6.29 3.17 5.07 8.84 7.06 11.20

2010 2.50 5.86 3.37 11.06 9.81 7.98 12.49

2011 2.00 5.85 3.33 10.32 9.42 8.18 8.81

2012 1.81 7.39 3.31 8.39 8.15 5.95 10.08

Average 3.04 6.52 4.15 7.49 10.17 9.75 11.19

Source: World Bank, World Development Indicators

The persistently high spreads and growing profitability of the industry have left it
open to repeated criticisms of collusive price-setting behaviour (World Bank 2013,
Oloo 2013). In the popular press and elsewhere, Kenyan banks have repeatedly
been portrayed as using their market power to extract high interest rates from
businesses, especially SMEs. The larger banks have been particularly subject to this
criticism, based on the perception that they use their reputational advantage to
charge higher rates on loans and advances, while not having to pay high interest
rates to attract deposits. This perception of high spreads at big banks is reinforced
by data showing them to be the most profitable segment of the industry. The
competition Commission has launched an investigation into the price-setting
behaviour of commercial banks, based largely on the concerns of consumers
regarding interest rate spreads.

Figure 8: The Performance of the banking Sector, 2002-2012

1,400 80

1,200 70

1,000 60

800 50

600 40

400 30

200 20

0 10
02 03 04 05 06 07 08 09 10 11 12


Source: Central Bank of Kenya

There have been several studies of interest rate spreads in Kenya (Abdul et al.
2013, Were and Wambua 2013, World Bank 2013). The World Bank (2013)
provides a good summary of these studies, first noting that that, while no hard rules
prescribe the optimal interest spreads that correspond to specific market conditions,
market lending rates are typically a mark-up over the risk-free (government paper)
interest rate, the magnitude of the mark-up depending on a host of factors,

Financial regulation in Kenya: 24

including industry structure, tenor, overhead costs, and risk. Determining this mark-
up when information markets are incomplete is especially challenging.

According to the Kenya Bankers Association (Oloo 2013), interest rate spreads
reflect the macroeconomic, regulatory and institutional environment under which
banks operate such that the determinants of the spread are in four categories:
macroeconomic factors and the state of financial sector development; industry-
specific factors; and bank-specific factors. We discuss these factors below.
(a)Macroeconomic environment. The size of the spread will depend on the
macroeconomic environment and the countrys monetary policy stance. There is a
high correlation between the spread and the CBR. The Central Bank of Kenya, for
example, raised the benchmark interest rate by nearly 300% (from 6.25% to 18%)
in less than three months in late-2011. As a result, banks raised their lending and
deposit rates. After August 2012, when the central bank started to lower the policy
rate as inflation moderated, bank lending rates were not as responsive. Although
banks did eventually lower their lending rates, the interest rate spread remained
high. According to the Kenya Bankers Association (Oloo 2013), the banks best
interests are served when interest rates remain low and stable, arising from a stable
macroeconomic environment. Further, a low interest rate regime has a direct
relationship with the quality of the banks' loan books, with expectations that non-
performing loans will increase in a regime of high interest rates.

(b) Financial sector development. Cross-country studies show large interest

spreads are associated with low levels of financial sector development. In general,
spreads in East Asia and Pacific are lower than in Sub-Saharan Africa. And within
Sub-Saharan Africa, the most advanced market (South Africa) exhibits small
spread. As the financial sector grows, spreads narrow. The spreads in Low Income
Countries (LICs) averaged 11.4% compared to 7.4% in Middle Income Countries
(MICs) over 1990-2012 (Figure 9). Compared to the 1990s, spreads have also
declined in Kenya.

Figure 9: Spreads in Low-Income and Middle-Income Countries





90 92 94 96 98 00 02 04 06 08 10 12


Source: World Bank, World Development Indicator. Missing LIC spreads were extrapolated.

(c) Industry-specific factors especially overhead costs. Kenya banks justify the high
spreads as due to the difficult business environment they operate in (Oloo 2013).
The main argument is that dispute resolutions take too long and is costly; while
national infrastructure services (e.g. electricity) are expensive and unreliable. They
also cite the high cost of attracting, training and maintaining human resources.
Salaries and other forms of labour compensation make up a large part of their

Financial regulation in Kenya: 25

overhead, as the scarcity of skilled financial sector workers leads to high turnover
and compensation packages geared to retain scarce skills (World Bank 2013). Most
banks estimate that salaries make up 50% of their overhead cost despite the fact tact
that Kenya has a fairly well-developed pool of banking skills. Nevertheless, the
largest portion of spreads is explained by profits in recent times (Table 9).

Table 9: Ex post Spread Decomposition in Kenya, %

2009 2010 2011 2012

Profit 41.6 47.6 51.4 47.9

Bad loans provisions 7.9 8.6 3.6 4.3

Overhead costs 44.6 38.1 38.7 40.2

Reserves 5.9 5.7 6.3 7.7

Total 100 100 100 100

Source: World Bank (2013)

As seen in Table 9, overhead costs do play an important role in explaining the

spreads in recent years in Kenya. Overhead costs have actually been going up since
the macroeconomic disruptions in 2011. Given the large share of salaries in the
overhead costs of the banking sector, increasing the supply of skilled labor to this
sector should be a priority (World Bank 2013).

(d) Bank-specific factors: Market structure. Large banks have higher spreads than
medium-size and small banks. The difference can be attributed to differences in the
cost of raising capital. Small and poorly capitalized banks find it more difficult to
raise funds. They have to offer higher deposit rates to attract funds and compensate
for the perception that they are riskier than large, more liquid, better-capitalized
banks, which are perceived to be too big to fail. Consequently, big banks are able
to mobilize more deposits even at relatively low or near zero deposit rates while at
the same time attracting large loan applications despite charging higher rates
(World Bank 2013). In Kenya, the banking sector is characterized by an
oligopolistic structure and market segmentation, in which the largest four banks
control about two-fifths of the market, partly as a result of their reputation and
customer loyalty, hence the need for increased competition and breaking the market
dominance by a few players (Mwega 2011).

(e) Bank specific factors: Lending risk premium. The difference between market
lending rates and short-term T-bill rates (Figure 7) can be interpreted as the risk
premium, and reflect the markets perception of risk. Over and above the actual risk
perception, where information gaps on credit history or market conditions and other
deficiencies in the financial infrastructure persist, banks are likely to price these
deficiencies through a higher risk premium (World Bank 2013).

Financial regulation in Kenya: 26

6 Prudential regulations
in Kenya

In 1988, the Basel Committee issued the Basel I Accord which assesses banks
capital adequacy requirements in the context of the credit risk they face and
advocates risk-based supervision. Basel I therefore emphasized a set of minimum
capital requirements for banks in order to address credit risk. In 2004, the
Committee issued the Basel II Accord which contained further recommendations
on banking laws and regulations. The Committee attempted to accomplish this by
setting up rigorous risk and capital management requirements designed to ensure
that a bank holds capital reserves appropriate to the risk the bank exposes itself to
through its lending and investment practices. The Accord was to be implemented
from 2007 by G10 countries, with more time given to developing countries, as they
were yet to satisfy the prerequisites for the new accord. Basel II has three pillars:
Pillar I on minimum capital requirements; Pillar II on the supervisory review
process; and Pillar III on market discipline. In December 2010, the Committee
announced proposals dubbed Basel III which are currently being reviewed for
regulatory and supervisory suitability to financial systems (Kasekende et al. 2011).
These proposals include the strengthening of capital adequacy and liquidity
requirements as well as countercyclical macroprudential measures.

The CBK continues to regulate banks mainly based on Basel I but was in the
process of formulating a policy position on Basel II implementation (KPMG 2012).
New guidelines that came into force in January 2013 contain some features of Basel
II and Basel III on capital adequacy requirements (Oloo 2013). Overall, Kenya has
endeavoured to implement the Basel accords for ensuring financial stability of the
countrys financial sector. The Kenyan banking system has continued to record
compliance with the minimum capital and liquidity prudential requirements. The
prudential and financial stability indicators have shown that the financial sector is
sound (Figure 10). All the banks have in the recent past met the four minimum
capital requirements with respect to the (i) Minimum core capital of Ksh 250
million which was raised to Ksh 1 billion over 2008-12; (ii) Core Capital/Total
Deposit Liabilities ratio (Minimum 8%); (iii) Core Capital / Total Risk Weighted
Assets ratio (Minimum 8%) and Total Capital/ Total Risk Weighted Assets
(Minimum 12%). In addition, the NPL/Assets ratio has decreased from a high of
22.6% in 2001 to a low of 4.3% in 2007, and of December 2013 averaged 5%, an
indication that the banking systems asset quality has generally improved over time.
As well, the ROA and ROE have generally shown an upward trend since 2002.

Based on the unaudited financial statements for 2012, almost all banks had met the
enhanced minimum core capital requirement of Ksh 1 billion, according to CBK21.
However, the final capital positions of the Kenyan banks will be determined once

Interview with CBK Governor in Oloo (2013).

Financial regulation in Kenya: 27

the audited financial statements are submitted at the end of March 2013. This is
however a minimum threshold and several banks already hold capital way above
the minimum of Ksh 1 billion. The key determinant of capital for an institution is
the needs of the market niche it serves.

One theory is that increased capital base is important for financial sector stability
and may lead to cost reduction from economies of scale which may lead to lower
lending rates. On the other hand, a further increase the capital requirement will
only create more concentration, making the banking sector more oligopolistic.
Gudmundsson et al. (2013) conclude that capital regulation improves the
competition, performance and financial stability of Kenyan banks22.
Implementation of the CBKs capital requirements for banks to build their core
capital can therefore be expected to enhance financial sector stability and lead to
cost reduction from economies of scale and ultimately lowering lending rates.

Figure 10: Selected prudential and financial stability indicators

for the banking sector 2011 - 2013

Dec-11 Dec-12 Dec-13 Statutory


Core capital to Total Risk 18.0% 18.9% 19.5% 8.0%

Weighted Assets Ratio

Total Capital to Total Risk 21.0% 21.9% 23.2% 12.0%

Weighted Assets Ratio

Core capital/Deposits 15.6% 16.3% 17.3% 8.0%

Liquidity Ratio 37.0% 41.9% 38.6% 20.0%

Gross Non-Performing 4.4% 4.5% 5.0% N/A

loans to Gross Loans Ratio

Return on Assets (ROA) 3.4% 3.8% 3.6% NA

Return on Equity (ROE) 30.3% 34.2% 28.9% NA

Source: Central Bank of Kenya

CBK has focused more on microprudential regulation which relates to factors that
affect the stability of individual banks and less so on macroprudential regulation
which relates to factors which affect the stability of the financial system as a whole.
In the latter case, changes in the business cycles may influence the performance of
banks, hence the Basel III proposal for countercyclical capital changes to provide
the way forward for future macroprudential regulation, which should take into
account the growth of credit and leverage as well as the mismatch in the maturity of
assets and liabilities. Murinde (2012) however argues that review of
macroprudential regulations should encompass the broader aspects of financial

They estimate the Lerner index and the Panzar and Rosse H-statistic as measures of competition and relate them
to core capital. The panel estimates show the log of core capital is positive and significant while squared log of
core capital is negative and significant. This implies that an increase in core capital reduces competition up to a
point and then increases competition so that the benefits of increasing capital requirements on competitiveness are
realized once consolidation in the banking sector takes place. They then use return on equity to capture bank
performance and stability and the estimation results confirm a positive relationship supporting the evidence that
capital regulation improves the performance of banks and financial stability.

Financial regulation in Kenya: 28

services regulation, such as depositor protection or deposit insurance and the safety
of the payments system which have received attention from CBK.

The regulatory toolkit in Kenya has also relied substantially on other variables such
as structure of banking assets and liabilities such as restrictions on banks large loan
concentrations and foreign exchange exposure limits (Kasekende et al. 2011). As
well, according to KPMG (2012), Kenya has a highly skilled workforce and the
banking sector is able to secure banking staff with relevant training, and finance-
related profession certification. In addition, the country has returning citizens with
international professional experience to add to an already diverse talent pool.
Capacity for implementing different regulations and supervision, such as lack of
information and insufficient staff do not seem to be a major constraint. In a group
of 11 SSA countries, Gottschalk (2013) finds Kenya to have the second largest
number of supervisors (60), largest number of supervisors with more than ten years
of experience (30) and the largest percentage of supervisors with a postgraduate
degree (80), although the number of onsite supervisors by banks in the previous
five years was comparatively low at 1.

Among other regulatory issues, Kenya has increasingly moved into universal
banking reflected in increasing share of net commissions and fees in the banks' total
income. The country now has banks that own insurance companies, others have set
up insurance agencies to push forward their concept of bank-assurance; while
others own stock brokerage firms. Hence there have been increased synergies
between the banking, insurance and securities sectors with removal of regulatory
barriers between the different segments of the financial sector. This poses
regulatory challenges as different financial sector entities are subject to different
regulatory regimes. Given the convergence and consolidation of the financial
services, some players have called for the established of an overall services
regulatory authority, as in UK (Mutuku 2008, Presidential Task Force on Parastatal
Reforms in Kenya 2013).

According to the Central Bank, the convergence of financial services is a global

phenomenon, with among its key drivers being the customer demands for a one
stop financial services super markets and competition. This poses regulatory
challenges as different financial sector entities are subject to different regulatory
regimes. The Central Bank has adopted a consolidated supervision approach, which
requires information sharing and coordination amongst the various regulators in the
financial sector. This is consistent with Spratt (2013) who advocates for (i) a
unified approach to supervision, with the centra