Social Science & Medicine 66 (2008) 2076e2087

www.elsevier.com/locate/socscimed

Markets, information asymmetry and health care:

Towards new social contracts
Gerald Bloom a,*, Hilary Standing a, Robert Lloyd b
a
The Institute of Development Studies, Brighton, UK
b
One World Trust, London, UK
Available online 7 March 2008

Abstract

This paper explores the implications of the increasing role of informal as well as formal markets in the health systems of many
low and middle-income countries. It focuses on institutional arrangements for making the benefits of expert medical knowledge
widely available in the face of the information asymmetries that characterise health care. It argues that social arrangements can
be understood as a social contract between actors, underpinned by shared behavioural norms, and embedded in a broader political
economy. This contract is expressed through a variety of actors and institutions, not just through the formal personnel and arrange-
ments of a health sector. Such an understanding implies that new institutional arrangements, such as the spread of reputation-based
trust mechanisms can emerge or be adapted from other parts of the society and economy. The paper examines three relational
aspects of health systems: the encounter between patient and provider; mechanisms for generating trust in goods and services in
the context of highly marketised systems; and the establishment of socially legitimated regulatory regimes. This analysis is
used to review experiences of health system innovation and change from a number of low income and transition countries.
Ó 2008 Elsevier Ltd. All rights reserved.

Keywords: Health systems; Trust; Social contracts; Reputation; Markets; Low/middle-income countries

Introduction in significant numbers of low income and transition

countries in Asia and parts of sub-Saharan Africa. As
In this paper, we examine health systems from the a result, there is now a considerable gap between nor-
perspective of how access to competent health care mative accounts of how health systems operate and re-
can be secured in environments characterised by high alities on the ground in the context of the increasing
levels of unorganised markets in health services and growth of markets in health care goods and services,
commodities, porousness of boundaries between public particularly informal and unregulated ones (Bloom &
and private health care sectors and lack of state regula- Standing, 2001).1 We argue that health systems are
tory capacity. These environments increasingly prevail
1
The concepts of ‘‘informal’’ and ‘‘unregulated’’ derive from the
* Corresponding author. The Institute of Development Studies, literature on the development of markets in low income countries.
University of Sussex, Knowledge, Technology and Society Team, The former refers to the degree to which the activity is recognised
Brighton BN1 9RE, UK. Tel.: þ44 1273 678733. in law or by legally recognised regulatory agencies and the latter
E-mail addresses: g.bloom@ids.ac.uk (G. Bloom), h.standing@ to the reach of the state in enforcing laws and regulations. We use
ids.ac.uk (H. Standing), rlloyd@oneworldtrust.org (R. Lloyd). the term ‘‘unorganised’’ to encompass both concepts.

0277-9536/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.socscimed.2008.01.034
 G. Bloom et al. / Social Science & Medicine 66 (2008) 2076e2087 2077

relatively path dependent and that institutional arrange- Second, some health care goods, such as immunisation,
ments, which have evolved over a long time in one have positive externalities in that an individual’s con-
context, do not necessarily transplant well to different sumption confers benefits on others. Third, markets
political economy contexts. Yet much of the textbook will lead to under-insurance against risks of major
advice to developing and transition countries has rec- health expenditure. Fourth, markets cannot compensate
ommended precisely that. In the introduction to this for inequalities in access to health resources (and may
special issue, we noted that in many contexts, various exacerbate them). We consider these arguments both
sets of actors are finding their own way, introducing in- convincing and largely uncontroversial. They all point
novations involving different arrangements that better to the need for regulatory regimes ‘‘beyond the market’’
fit the realities they face. We, therefore, argue for and incorporating some political or normative consen-
a much stronger focus on understanding the institu- sus on what health systems should deliver. These re-
tional context and associated adaptations within which gimes have generally been associated with the nation
health systems operate in low income and transition state, although substantial global flows of resources ear-
countries. marked for health services in low and middle-income
In this paper, we employ the frame of the social countries raise regulatory challenges as well.
contract to understand the transactions that underpin Here we focus more on a fifth market failure argu-
health systems. We focus particularly on the problem ment e that of information asymmetry. Such asymme-
of information asymmetry and associated power rela- tries are seen to pervade health care markets, which are
tionships in the context of the growth of markets in characterised by high levels of uncertainty (Arrow,
health goods and services, as information asymmetries 1963; Haas-Wilson, 2001). For instance, patients may
are seen particularly to characterise transactions in be able to describe their symptoms, but they have inad-
health care. We examine three relational aspects of equate information to relate their condition to a partic-
health systems: the encounter between patient and pro- ular type of treatment or course of medication. This
vider; mechanisms for generating trust in goods and creates an unequal power relationship between experts
services in the context of highly marketised systems; and clients which the former may exploit in their own
and the establishment of socially legitimated regula- interest. Non-market institutions such as professions,
tory regimes. We use this analysis to review experi- regulatory frameworks, standard setting and public
ences of health system innovation and change from health bureaucracies, have evolved to mediate pa-
a number of countries. tienteprovider interactions (Haas-Wilson, 2001).
Where these arrangements are poorly developed or
Conceptualising markets and institutions have broken down, access to competent health-related
in health in the 21st century knowledge is compromised, particularly for the poor.
In the light of the changes we outlined in the intro-
Our starting point is an understanding of health sys- duction, including the greatly increased availability of
tems as knowledge economies which produce and me- information and services through pluralistic and unme-
diate access to health knowledge embedded in people, diated channels and the growing significance of mar-
services and commodities and which can potentially be kets in health goods and services, it is worth looking
organised in different ways. These ways encompass again at the question of information asymmetry and
context-dependent factors, such as types of governance the different sources of trust which underpin health
and relative strengths of states, and other factors such transactions. Is information asymmetry less or more
as changes in technologies and health needs. There is of a problem now, and for whom? Are new institutional
a long history of debate about the degree to which forms evolving in the health sector to deal with it and if
the health sector is different from other economic so, what kinds? What ways are emerging to cope with
sectors. In particular, economists argue that health the phenomenon of unorganised markets?
care markets have inherent structural features that The concept of the social contract is a useful fram-
lead to market failures (Arrow, 1963; Bennett, ing device. Arrangements for making the benefits of
McPake, & Mills, 1997). There are several powerful expert medical knowledge widely available can be
arguments for why markets alone cannot produce the understood as social contracts between actors, under-
most efficient or equitable allocation of scarce re- pinned by formal and informal rules and shared under-
sources in health and require state intervention. First, standings and behavioural norms. Contracts, in the
health care includes ‘public goods’, such as sanitation, form of reciprocal relationships that are social, politi-
which would be undersupplied if left to the market. cal, organisational and psychological enable us to ask
2078 G. Bloom et al. / Social Science & Medicine 66 (2008) 2076e2087

broader questions about how the social rules and asso- reward or punishment, which become socially embed-
ciated behaviours governing relationships between the ded through imposition by a central body, or through
different sets of actors develop historically, are sanc- reputational mechanisms where individuals sanction
tioned, challenged and changed. Markets are under- each other, or through self-policing. Platteau (1994b)
pinned by such contracts, expressed through the places the greatest emphasis on the last of these as
social relations and norms that constrain opportunistic a powerful way for internalising norms into a self-
behaviour and mediate interactions (Granovetter, 1985; sanctioning system that maintains attitudes, values
Moore, 1994; Platteau, 1994a, 1994b). Central to these and patterns of behaviour in the absence of external
social relations is the functioning of trust in governing rewards or punishment (Coleman, 1987). Social norms
all kinds of market transactions, including in health. thus impose a self-enforced order on a market, compel-
The concept of trust in the context of health care has ling agents to behave fairly and constrain individual
received attention in recent years (Gilson, 2003; interest while engaging in exchange transactions.
Gilson, Palmer, & Schneider, 2005). Gilson (2003) ar- Granovetter (1985) argues that reputation built
gues that in highlighting the essentially relational and through concrete personal relations and networks is
co-operative nature of health systems, trust offers a cor- the key to trust. The repetition of contact that arises
rective to economic individualism in understanding from agents operating within the same social networks
critical challenges such as staff behaviour and motiva- builds a stock of information shared by all members on
tion and building institutional and ethical legitimacy. the reputation of others. This serves to discipline the
In this sense it is linked conceptually to recent discus- behaviour of agents, forcing them to weigh the long-
sions of social capital as the ‘‘glue’’ which creates func- term costs of a bad reputation against the short term
tioning institutions (Woolcock, 2001; World Bank, benefits of dishonesty (Grabowski, 1998). The result
1999). Gilson notes that theories of trust lie on a spec- is an accumulation of reputation-based trust amongst
trum which partly reflects disciplinary approaches. members.
Strategic and calculative notions of trust derive more Where these different analyses converge is in their
from approaches that focus on individual behaviours, understanding of markets as social phenomena where
such as rational choice theory (Gilson, 2003; Lyon, order is created and enforced through social relations.
2000). These involve a calculation by an individual of Examples of both generalised norms (such as internal-
whether a person’s future behaviour will be beneficial ised values of probity and service to others) and repu-
or harmful, and whether co-operating with them will tational mechanisms operating to construct trust and
produce benefits that outweigh the costs of trust. manage information asymmetries can be found in
From this perspective trust can be viewed as a means health care markets. In highly marketised settings these
of reducing the complexity and risks that come from different mechanisms are increasingly being drawn
the autonomy and freedom of others. Trust is, therefore, on and used in innovative ways to manage health
a mechanism to make calculations easier. At the other transactions.
end of the spectrum are affective understandings of
trust which are emotionally based and rooted in as- The changing role of states and
sumptions of shared moral values. Both types of trust markets in the health sector
play a role in facilitating the construction of trust-based
economic institutions, including in health. The pluralistic health systems of many low and mid-
Theoretical positions on how trust is created within dle-income countries reflect a historical legacy of the
markets and other forms of associational life fall construction and subsequent decay of particular institu-
broadly into two kinds: trust based on generalised tional arrangements. During the third quarter of the
moral norms, and trust built on reputational mecha- 20th century many African and Asian countries estab-
nisms. Both are relevant to the management of infor- lished regimes that made a radical break from a colonial
mation asymmetry. The understanding of trust as or pre-revolutionary past. These newly established
a product of generalised moral norms is associated par- governments were committed to the provision of uni-
ticularly with Platteau (1994b). He sees the creation of versal access to health care in response to expectations
trust as a function of scale. Once a market exceeds the raised during anti-colonial and revolutionary mobilisa-
size of a community, where there is complete informa- tion. This was to be achieved by the rapid creation of
tion and repeated interactions among agents, trust must a government health service. This strategy was based
be based on generalised attributes or norms. These on the prevalent understanding of the role of the state
norms are underpinned by sanctions, in the form of as economic actor and service provider.
 G. Bloom et al. / Social Science & Medicine 66 (2008) 2076e2087 2079

Since the early 1980s, economic and structural cri- reneging on previous commitments.2 This has contrib-
ses have exposed the weak institutional capacity of the uted to a growing gap between reality and the dis-
public sector in many countries. In health as in other courses that underlie policy discussions, particularly
sectors, growing demands and structural reform poli- on the part of government officials and international
cies have left the state overworked and underfunded. donor agencies. This gap has meant that a growing pro-
The result has been the rapid growth of non-state pro- portion of activities of government employees takes
vision to fill gaps in supply. This has been both a formal place outside their formal roles and terms of employ-
process, sanctioned by the state through contracts with ment (Bloom & Standing, 2001). It has encouraged
not-for-profit providers and legislation allowing the other actors to construct institutional arrangements
development of the private health care market, and through various kinds of co-production or outside the
an informal process with burgeoning numbers of unli- ambit of the state (Joshi & Moore, 2004).
censed practitioners, pharmacists and drug pedlars in-
creasingly providing services. In addition, public The changing context of encounters between
sector providers have emerged to sell their services clients, consumers and providers
in an unorganised market. The balance between state
and market has undergone a radical shift with inter- Trust is central to the social contract of the medical
penetration between market-type transactions and state encounter and affects numerous dimensions of the
institutions, operating to a large extent through infor- relationship (Arrow, 1963; Davies & Rundall, 2000;
mal arrangements. Hall, 2001; Hall, Dugan, Zheng, & Mishra, 2001;
Broader trends have affected this shift of balance in Mechanic, 1998). These writers argue that trust serves
health-related transactions. The ideological climate, as a means of managing the problem of information
reflected in the so-called ‘‘Washington consensus’’ of asymmetry. It both diminishes the transaction costs
policies of international institutions, encouraged a re- of large amounts of external monitoring and is essen-
duction in the size of the state (Stiglitz, 2002). Invest- tial to the types of transaction that are less amenable
ments in infrastructure and education reduced to management by explicit contracts, such as interper-
constraints to market activities. In a number of Asian sonal clienteprovider relationships (Davies & Dibben,
countries, for instance, India and China, economic 2001). Trust also facilitates the disclosure of sensitive
growth and the expansion of information flows have information; ensures the co-operation of patients dur-
contributed to rising demand for health goods and ing treatment; and increases the likelihood that patients
services and high levels of private health expenditure will recommend treatment to others (Gilson, 2003;
(NHA, 2007). In many other countries economic crisis Mechanic, 1998). The role that trust plays points to
and weaknesses in public sector management and ways in which the medical encounter is both like and
governance contributed to the growth of formal and in- unlike other types of relationship between providers
formal markets. In the former command economies, and users of expertise. One particularity of trust in
the rapid growth of markets was a response to the the medical encounter lies in the way it mediates a re-
adoption of a new economic model. All these factors lationship focused on the intimacy of the human body
led to a dramatic increase in the importance of market and the risks and vulnerabilities this entails for the pa-
relationships, particularly unorganised ones. Many tient. Trust also mediates medical outcomes (Hall,
governments could not adapt and have limited capacity 2005; Hall et al., 2001; Mechanic, 1998). For example,
to influence markets (Fukuyama, 2004). This has left medical anthropology has explored the significance of
households in many low and middle-income countries trust in the healerepatient relationship for outcomes
relying heavily on community-level institutions to across cultures and in accounting for the differential
‘‘regulate’’ markets and organise ‘‘social protection’’ effects of treatment among patients (Hall et al., 2001;
(Wood, 2004). Helman, 2007).
The legitimacy of many of the regimes that emerged Behind the medical encounter lies power. Although
in the mid-20th century has been linked to post-colonial institutional structures exist to signal a provider’s com-
and post-revolutionary understandings and political petence and thus offer patients a rational basis for trust,
commitments (Mackintosh, 2006). Politically weak
governments have been unable or unwilling to engage 2
We use politically weak to refer to states where public authority is
in public debates about how best to use their limited re- poorly legitimated and/or where government authority is overridden
sources and regulatory powers, fearing that their legiti- by alternative forms of control such as patronage and other forms of
macy would be diminished further by explicitly non-state interest.
2080 G. Bloom et al. / Social Science & Medicine 66 (2008) 2076e2087

Hall et al. (2001) argue that trust in the medical en- ‘‘informed patient’’ model where people suffering
counter is predominantly non-rational and affective, from a particular condition are provided with ‘‘best
deriving from the inherent vulnerability of the care- available evidence’’ to enable them to play an active
seeker. This makes it a powerful relationship, open to role in their treatment (Detmer et al., 2003). These
abuse as well as to fulfilment of the psychological con- approaches provide systematic evidence to facilitate
tract. However, providerepatient or client relationships the creation of instrumental forms of trust to manage
are also deeply influenced by the social context in information asymmetry.
which they are embedded. Where there are high levels As noted, these debates have largely been con-
of poverty and inequality, the social and economic sta- ducted in the context of the OECD countries. They
tus differences between providers and service users are have largely focused on interactions with the medical
often considerable. These status and other markers of profession. Their relevance to environments where un-
difference have been shown repeatedly to affect the regulated markets dominate and much health care is
capacity of poor users to obtain good quality and re- delivered or purchased outside the formal encounter
spectful services (Gilson et al., 2005). Gender and between health professionals and users has been
other markers of identity can also profoundly influence much less explored. In these contexts, empirical stud-
interactions (Sen, George, & Östlin, 2002). ies suggest considerable complexity in the kinds of so-
The degree of uniqueness of the medical encounter cial contracts that providers have with users, the types
is partly at issue in recent debates in OECD countries of trust they embody and the extent of information
on the desirability of greater patient autonomy. A com- asymmetries (Leach, Fairhead, Millimouno, & Diallo,
bination of broader social movements, declining trust 2008; Leonard, 2002, 2005) For instance, informal pro-
in experts and the rise of consumerism have challenged viders, such as some indigenous healers or local
what is seen as old-style paternalism in medicine, asso- ‘‘quacks’’ with a high stock of social capital may at-
ciated with unconditional trust and the disempower- tract high levels of affectively based trust which can
ment of patients (Quill & Brody, 1996; Tauber, survive an adverse outcome, while others may be based
2003). This has led to different responses. Advocacy on relatively fragile instrumental trust which evapo-
for institutional measures to standardise procedures rates when a poor outcome occurs (George, 2007).
and protocols to reduce provider discretion and the Similarly, formal providers and facilities may or may
need for emotionally based trust challenges the view not be trusted in different ways depending on their be-
that health care providers are different from other pro- haviour towards users or reputations acquired through
viders of skilled services. ‘‘Collaborative’’ models, in social networks (Davies & Mannion, 2000;
which providers and clients engage in dialogue and ne- Tibandebage & Mackintosh, 2005).
gotiation over the best way forward (Quill & Brody, Perhaps most significantly, evidence from surveys
1996), attempt to broker between experience and of health expenditure across developing countries indi-
expertise and thus diminish what is seen as an informa- cates that very high percentages of health care transac-
tion gap on both sides. Kuhlmann (2003) draws on tions now take place through ‘‘self-treatment,’’
empirical work into the nature of trust in the pa- meaning usually resort to pharmacies and informal
tientedoctor relationship in Germany to suggest that drug sellers (Cederlof & Tomson, 1995; Reynolds
evaluations and audits can be a new resource for build- Whyte, Van der Geest, & Hardon, 2003). In these set-
ing a more rationally based trust between providers and tings, issues of trust may also be related more instru-
patients as they reduce information asymmetries. ‘‘In- mentally to concerns as to whether the supplier is
creasingly, justified trust in health care is founded on able to provide affordable treatment or whether the
information, and the ‘carriers’ are the new tools from drugs being supplied are genuine or counterfeit.
the realms of management’’ (Kuhlmann, 2003: 11). Central to understanding these issues is the question
The rise of discourses of the patient as consumer re- of where people get information, how they make
flects the growing importance of markets in health care judgements about quality and how these vary across
more globally (Kickbusch, 2004). Rothman (2001) social groups. Recent work by Leonard (2007) in rural
notes the extension of consumer influence in the Tanzania has questioned common assumptions about
United States over an increasing range of treatments, the relative ignorance of poor households in making
aided by the greater availability of information through health care choices. He found that on a range of ac-
a variety of media as well as the stimulus from cepted quality indicators, households exhibited social
patients’ rights movements. One recognition of the learning over time in converging towards ‘‘good’’ phy-
challenge this represents is the development of the sicians and bypassing less good ones. He also noted
 G. Bloom et al. / Social Science & Medicine 66 (2008) 2076e2087 2081

that the common assumption by health staff that users decrease less than might be anticipated if and when
sought out unnecessary or harmful interventions such formal services improve. An important question raised
as injections was not borne out in this context. In by these trends is whether this signals a more general
fact they sought out the consultation and avoided the shift towards instrumentally based forms of trust at
injections. Such findings suggest that both mechanisms the level of basic service provision. This would point
of learning and types and levels of information asym- towards more use of self-managed and routine proto-
metry vary in important ways across populations. col-based health care for routine curative and preven-
Rather than a standard call for more health education, tive services. The emerging shape and modes of
they point to the need for greater empirical specificity delivery of basic services to take account of this unme-
in our understanding of user choices and the informa- diated consumption of health knowledge and its impli-
tion they need. cations for usereprovider interactions require more
A number of emergent approaches have practical research.
implications for improving access to competent health
goods and services through reducing information Mechanisms for generating trust in health
asymmetries in the clienteprovider relationship. These care in highly marketised environments
include the introduction of report cards on facilities
and the certification of providers eligible to see patients Unorganised markets for health services are sub-
holding cards or vouchers issued by government for optimal in terms of efficiency, equity and quality
specific health goods and services, such as maternity (Bennett et al., 1997; Hsiao, 2000); yet they are a reality
care (Montagu, 2003). in many countries, delivering a major share of health-
Where the management capacity and governance ar- related goods and services. Such contexts are character-
rangements of health systems are weak there are argu- ised by a blurring of boundaries between public and
ments for focusing on making reliable information private and organised and unorganised health sectors;
available in order to limit the need for affective forms for instance, government health workers supplement
of trust in specific providers and suppliers. This is par- low salaries with market-like activities (Bloom &
ticularly important where people depend heavily on the Standing, 2001; Mackintosh & Tibandebage, 2002).
informal sector. Relevant initiatives focus on increas- Unorganised markets and blurred boundaries pose sub-
ing the availability of both mediated and unmediated stantial challenges in bringing the informal sector into
information directly to users. In Bangladesh, low cost an overall public policy net (Mills, Brugha, Hanson,
Internet technology through the mobile phone network & McPake, 2002). In this section, we note emerging ap-
has facilitated the introduction of a 24-h direct health proaches to managing these challenges through mecha-
line staffed by health professionals, which users any- nisms to build trust through norms and/or reputationally
where in the country can call for consultation and in- based ways of delivering competent health care.
formation on pharmaceuticals and services (Barton, Norms-based ways of influencing performance are
2006). Lucas (2008) notes other developments of this predominant in the growing use of contracts between
kind of technologically mediated information which governments or donor agencies and health service sup-
have information asymmetry reducing effects. Other pliers. Contracting approaches are used particularly to
examples include making information directly avail- align incentives for performance more closely with pol-
able to users on sources of reliable goods and services icy objectives in contexts where the public sector has
through branding and franchising (see next section), on serious problems, employees face strongly perverse in-
prices charged for services or on provider performance centives, both financial and non-financial and where
(Reynolds Whyte et al., 2003). formal management systems have limited leverage
We may be seeing a more fundamental global shift over health worker performance (Pritchett & Woolcock,
in the boundaries between mediated and unmediated 2004). Evidence on their use in low or middle-income
access to medical knowledge which echoes trends in countries suggests they work best for services that re-
some OECD countries but has different origins. The in- quire little discretion on the part of providers and where
creasing level of resort to self-treatment in low and it is possible to define clear outcome measures (Liu,
middle-income countries is perhaps more a function Hotchkiss, Bose, Bitran, & Giedion, 2004; Loevinsohn
of the high cost of good quality mediated care. How- & Harding, 2005; Meesen, Musango, Kashala, &
ever, given the spread of markets, multiple sources of Lemlin, 2006).
information and growing availability of, for example, Many other factors influence the performance of
low cost diagnostic technologies, self-treatment may health workers (Dieleman & Harnmeijer, 2006). These
2082 G. Bloom et al. / Social Science & Medicine 66 (2008) 2076e2087

influences include behavioural norms, such as moral transaction costs of monitoring quality through localis-
rules and professional ethics. Concerns have been ing incentives and aligning the goals of the franchisee
raised about the potential influence on these norms of with those of the organisation (Montagu, 2002). A suc-
the widespread use of financial incentives to influence cessful organisation then translates a reputation for
health worker performance (Gilson, 2003; Segall, quality into formal and informal norms that can power-
2000). However, one can also anticipate contexts fully influence the performance of employees.
where measures that align formal and informal incen- The important role of international and large na-
tives could increase a sense of service and profession- tional NGOs in providing health services is an example
alism and where contracts can be vehicles for of a branded service. Hospitals associated with reli-
reconstructing norms and ethics. This underlines the gious organisations also display aspects of branding,
need to understand the nature of the existing social particularly if an aspect of their reputation derives
contracts onto which formal contracting is laid, as from their religious affiliation.
this will have a major influence on outcomes. Examples of franchising occur, particularly in re-
Recent work by Mackintosh and Tibandebage productive health (Bishai, 2002; Prata, Montagu, &
(2002) and Tibandebage and Mackintosh (2005) rein- Jeffeys, 2005). Socially franchised health care in
forces this point. They found that within Tanzania’s Kenya (KIMET), Pakistan (Green Star) and Bihar State
highly marketised health system, where self-interested in India (Janani) trains health workers in basic repro-
behaviour predominates, there, nevertheless, exist ductive health services and identifies them to clients
pockets of institutional resistance where norms of pro- as providers of quality services. Bishai points out
bity and charity are in operation, including in private that providers have not earned enough revenue from
practice. In such a context, they argue for starting participating in these initiatives to pay for the co-
with what works in terms of strengthening valued yet ordinating and reputation-building role themselves
fragile norms. These might include support for self- and argues that the future lies in government subsidies
regulatory processes, as well as tackling perverse in- for this key trust-building function, since improved
centives through methods such as contracting private quality of basic services is a public good.
sector providers. Examples from middle-income countries may pres-
In pluralistic health systems, reputation is critical age future developments in the provision of basic
to providers’ livelihoods and is built and maintained health services by for-profit companies in low income
in different ways, both informal and formal. One infor- countries with growing private markets. One is the
mal source of reputations is through local knowledge Farmacias Similares chain of retail pharmacies in
networks (Fang, 2006; Leonard, 2007). Provider associ- Mexico, which provides a consultation with a doctor
ations can enhance and protect reputations. One exam- and generic medications at a relatively low cost
ple is the patent medicine vendors’ associations in (Knowledge@Wharton, 2005). Another is a chain of
Nigeria. Amongst other things, they organise informa- primary health care providers in South Africa, which
tion meetings for members and penalise members for provides care based on treatment algorithms at a com-
bad practices, including the sale of sub-standard drugs petitive price (Palmer, Mills, Wadee, Gilson, &
(Oladepo et al., 2007). This type of self-regulation Schneider, 2003). The latter is an example of the use
is likely to play an increasingly important role in of information technology to reduce the health work-
marketised environments, as part of a suite of reputa- er’s discretion and, thereby the influence of informa-
tion-based ways of managing informal markets which tion asymmetry.
also include reinforcement from ‘‘voice’’ mechanisms The situation is different for the complex services
such as consumer organisations and pressure groups that specialised hospitals provide, with a growing seg-
(Tibandebage & Mackintosh, 2005). mentation of services for different social groups.
Organised reputational mechanisms, such as those Branded hospital chains play a growing role in the pro-
embodied in brands, franchises and accreditation vision of services to the better off. For example, Chee
schemes, are beginning to play an increasing role in (2008) shows how local and foreign-owned chains
basic health care. Built into a brand is an institutional have become an important source of care for local elite
reputation that can be damaged if its products or ser- groups and medical tourists in Malaysia. International
vices are low quality. This provides producers with or local accreditation has become another marker of
an interest in maintaining quality and consumers with superior quality. There are a few examples of the use
the sanction of withdrawing their custom. This same of accreditation to improve the quality of hospital
dynamic is at work in franchising, which reduces the care for the poor (Montagu, 2003), but accreditation
 G. Bloom et al. / Social Science & Medicine 66 (2008) 2076e2087 2083

requires a high level of technical capability which is a similar point in its discussion of the so-called
less likely to be present in contexts where informal ‘‘long route’’ of political accountability for making
health care markets dominate (Zwi, Brugha, & Smith, services work for poor people. This is particularly
2001). challenging in contexts of deep structural inequalities,
These approaches rely heavily on the social context where governments face conflicting pressures while at-
within which they are embedded. Many examples are lo- tempting to build consensus around rules understood to
cal and tentative and it is clear that much of the work in be legitimate, fair and enforceable and which contrib-
constructing reputations for quality and trustworthiness ute to the construction of shared norms of behaviour
must be local, will involve a combination of informal (Bloom, 2001). The outcome depends on how actors
and formal mechanisms and will depend on a locally are involved in the building of consensus. Cornwall
specific configuration of state, market and civil society and Shankland (2008) illustrate how the creation of
partnerships (Mackintosh & Tibandebage, 2002). Nev- mechanisms for citizen participation in planning local
ertheless, these offer some basic modalities for socialis- government health systems in Brazil contributed to
ing unorganised markets through different combinations the translation of policy statements into a broad na-
of norms and reputation-based mechanisms. tional consensus on government’s responsibilities in
health, which transcends party politics. The creation
New regulatory regimes of a consensus around a new understanding of the re-
sponsibilities of government and other relevant health
Initiatives in marketised environments to build trust sector actors can change the context within which the
and improve performance require a supporting regula- inevitable contestation between interest groups takes
tory environment if they are to go beyond the local. place.
Government strategies for creating such an environ- The institution-building process differs between
ment have changed as their direct influence over pro- countries with strong and weak states. In countries
viders has diminished. One response has been the with weak states, the reliance on partnerships reflects
creation of partnerships between states, market players a failure to consolidate a government bureaucracy
and civil society organisations to ‘‘co-produce’’ regula- that acts in the public interest. Pritchett and Woolcock
tory arrangements (Centre for the Future State, 2005). (2004) argue that existing rules and norms of behaviour
For these approaches to be effective, the rules and the often do not provide sufficient disincentives to discour-
agents tasked with enforcing them have to build trust age service providers from acting against the interests
through social and political legitimacy. This section of their clients, particularly for activities that involve
focuses on initiatives to create legitimated regulatory a lot of interaction between providers and users of ser-
arrangements in contexts of both weaker and stronger vices and require technical discretion. They attribute
states. the increasing interest in arrangements for giving citi-
Several writers have argued for a broader understand- zens and users of services more influence to the failure
ing of the meaning of regulation. Ensor and Weinzierl of organisational models from advanced market econ-
(2006) note the limited capacity of governments of omies to function as expected. Joshi and Moore
many low and middle-income countries to control qual- (2004) point out that the blurring of boundaries be-
ity or prices through administrative and bureaucratic tween the roles of the state and of other bodies in ser-
controls. They emphasise the need for complementary vice provision and/or regulation is quite common
measures such as contracting, self-regulation and where capacity to fund and regulate services is weak.
accreditation and advocate strengthening the role of in- They propose the term institutionalised co-production
formed consumers and citizens and of quasi-government to describe the way that public service provision, in-
or independent bodies in assisting with market-enhanc- cluding regulation, becomes a joint production through
ing controls. Londono and Frenk (1997) see the role of regular, long-term relationships between state agencies
government as the provision of strategic direction to and organised groups of citizens, to which both con-
the health system. The World Health Organization and tribute substantial resources. Examples in the health
some analysts use the term ‘‘stewardship’’ to reflect gov- sector are partnerships between government and local
ernment’s role in representing the overall social good NGOs and/or religious mission health facilities to de-
(Saltman & Ferroussier-Davis, 2000). liver services and/or organise community health
This thinking draws attention to the long-term, po- insurance.
litical nature of the construction of regulatory arrange- Loewenson (2003) notes the significance of institu-
ments. A recent World Bank (2003) report makes tionalised co-production in the health sector between
2084 G. Bloom et al. / Social Science & Medicine 66 (2008) 2076e2087

the state and civil society organisations and other financial collapse of health facilities in the face of rad-
health sector actors to deliver services, plan resource ical changes to a myriad of economic and institutional
allocation or regulate provider performance. State ac- arrangements (Bloom, 2005). The partially regulated
tions include the development of legal frameworks markets that emerged provided powerful incentives
and taxation policies as well as service subsidies. But for health providers to practice an increasingly costly
a common problem is weak state capacity to undertake style of care (Eggleston, Li, Meng, Lindelow, &
many of these tasks. She also notes considerable areas Wagstaff, 2006). The government began a reform pro-
of tension in the relationship due, for instance, to dif- cess in the mid-1990s, which is gradually changing
fering expectations between collaborating actors and institutional arrangements and creating new social
mistrust by governments of civil society motivations. contracts (Bloom et al., 2006). It is using top-down
But there are also examples of other institutional ac- bureaucratic systems to identify and punish corrupt be-
tors, such as international agencies, playing facilitating haviour. It is establishing systems to license pharma-
roles which can mediate these potential areas of ceuticals and health workers. It is gradually
tension. establishing minimum quality standards, often in re-
One area of increasing interest is the potential sponse to serious regulatory failure. Meanwhile, it is
for partnerships between governments of low and mid- encouraging health facilities to build trust with the
dle-income countries and organisations subject to regu- community and one finds a variety of initiatives to
lation in a country with a strong state, including increase accountability (Fang, 2006). It has recently
international NGOs and a variety of for-profit corpora- begun to earmark money for local insurance schemes
tions. This provides opportunities for building supra-na- that could potentially become strategic purchasers of
tional trust-based institutional arrangements in contexts medical care and it is considering reforms to alter
where unorganised markets predominate. However, it the incentives that health facilities face. Although the
raises complex issues concerning the mediation of con- health system still performs much less well than eco-
flicts of interest and the construction of sustainable nomic sectors that rely less on trust-based institutions,
national institutions. the picture that emerges is of a combination of local
Finally, despite weaknesses, national governments innovations to build trust-based arrangements and the
also show some capacity to produce focused regulatory gradual creation of a basic regulatory framework.
responses. One example is the Nigerian Government’s These arrangements are embryonic and new social
response to public outcries against deaths caused by contracts are evolving.
the supply of counterfeit pharmaceuticals. Established China’s approach to the construction of trust-based
in 1994, the National Agency for Food and Drug institutional arrangements through a combination of
Administration and Control (NAFDAC) was greatly the creation of minimum enforceable regulations to
strengthened in 2001 as part of a government-wide deal with the worst practices and the encouragement
anti-corruption campaign. NAFDAC has widely publi- of local adaptations is relevant to other countries grap-
cised the seizure of sub-standard products and in the pling with the consequences of unorganised markets. It
process, challenged powerful vested interests. In doing is particularly applicable where the state is relatively
so, the government has linked its legitimacy to its strong, but even in countries not known for the en-
capacity to end these corrupt practices. If it succeeds, forcement of rules-based regulations, governments
it will contribute greatly to the construction of a legiti- can respond effectively to practices widely acknowl-
mate and effective regulatory framework for health that edged to be harmful and corrupt. The attempt by the
should in turn make it easier to construct more com- Nigerian government to remove low quality pharma-
plex arrangements. ceuticals from the market is an important example.
We noted earlier in this paper the importance of
trust and shared expectations and behavioural norms Conclusion
to the performance of the health sector. This explains,
to some degree, its path dependent nature and the time In this paper, we have examined some of the impli-
it takes to construct and de-construct particular ar- cations of the increasing role played by markets, par-
rangements (Pierson, 2000, 2003). China’s approach ticularly informal ones, in the health systems of low
to transition management illustrates how a strong state and middle-income countries. We have argued that in
is attempting to build institutions to govern an increas- dealing with the consequences of these, it is necessary
ingly complex economy (Yang, 2004). For many years, to take into account the path dependent nature of health
the priority in the health sector was to prevent the systems and move away from policy prescriptions
 G. Bloom et al. / Social Science & Medicine 66 (2008) 2076e2087 2085

based on universalistic prescriptions implicitly derived 20th century institutional arrangements. The combina-
from the experiences of the OECD countries. tion of the spread of markets and the development of
Health systems should be understood dynamically new technologies, including ICTs and new diagnostic
as knowledge economies, not just as assemblages of technologies, will lead to very different arrangements.
technical services, goods and personnel. This means This will have major implications for the roles of
that they can potentially be organised in ways which expert intermediaries and for bureaucracies. In partic-
draw upon other aspects of the economy and society. ular, there are likely to be further changes in the ex-
We illustrated this with examples of different kinds tent to which access to the health knowledge
of trust-based mechanisms and regulatory responses economy is mediated by organised professions and
which have emerged as local responses to manage bureaucracies. This, in turn, has implications for the
highly marketised environments. We have thus tried construction of knowledgeable citizens/consumers
to go beyond simple state-market dichotomies and de- and for the role of governments in the creation of reg-
bates about privatisation in the health sector. Markets, ulatory partnerships and in facilitating improved
themselves, are socially embedded and their institu- access to information. Finally, this will also have con-
tional arrangements reflect local social relationships sequences for the political debates about how to build
and accommodations with a variety of social, eco- socially inclusive future health systems to reverse the
nomic and political forces. It is more productive to inequalities in access to reliable knowledge and re-
explore approaches to constructing social contracts sources that tend to characterise highly marketised
for health care which build on existing areas of compe- health systems.
tence and good practice, whether mediated by states,
markets or other institutional actors. We noted areas Acknowledgements
where this is happening. We hypothesise that we will
see an increasing range of institutional innovations in The authors would like to acknowledge the valuable
the health sector, which involve different co-produc- comments and suggestions by participants at a work-
tion arrangements and contribute to rewriting the social shop on Future Health Systems at the Institute of De-
contracts between states and citizens on access to ex- velopment Studies, University of Sussex in October
pert health knowledge. These are likely to include 2004 and also the thoughtful comments by Damian
new forms of market arrangements, such as franchising Walker, Maureen Mackintosh, John Abraham and three
and forms of vertical integration; the increasing use of anonymous reviewers. They thank Chris Pell for his
different types of knowledge mediators that can facili- excellent work as a research assistant. This paper is
tate access to reliable information and services; new an output of the Future Health Systems Research Pro-
types of regulatory approaches drawn from experience gramme Consortium (www.futurehealthsystems.org).
in other sectors; and different kinds of political engage- The authors express their appreciation for the financial
ment for policy influence coming from an expanded support (Grant # H050474) provided by the UK De-
range of social actors. partment for International Development (DFID). The
Trust, in its different dimensions, is essential to views expressed are not necessarily those of DFID.
functioning relationships at different levels of health
systems through establishing shared norms and values,
References
reputation and legitimacy. We have focused particu-
larly on trust in the context of the problem of informa- Arrow, K. J. (1963). Uncertainty and the welfare economics of
tion asymmetry as that lies at the core of many of the medical care. American Economic Review, 53(3), 941e973.
institutional arrangements found in the health sector. In Barton, A. (2006). Grameen launches all-day health-line in
the light of increasing marketisation, consumerism and Bangladesh. [ICT for Emerging Markets]. Developing Tele-
coms. <http://developingtelecoms.com/content/view/671/26/>
technological change in the health sector, we speculate
(Accessed 27.11.06).
that the role played by information asymmetries is Bennett, S., McPake, B., & Mills, A. (1997). Private health providers
changing in low and middle-income countries as well in developing countries. London: Zed Press.
as in rich ones and may be decreasing in some types Bishai, D. (2002). Improving quality and access in private sector
of transaction with changes in the ways in which ac- primary health care e the role of business models. Carolina Pop-
cess to the health knowledge economy is mediated. ulation Center Alternative Business Models Project. Available
from. <http://www.cpc.unc.edu/projects/abm/papers/bishai.pdf>.
This is not just a narrative about failed states or Bloom, G. (2001). Equity in health in unequal societies: meeting
about linear stages of development of health systems. health needs in contexts of social change. Health Policy, 57(3),
It is unlikely that future health systems will reconstruct 205e224.
2086 G. Bloom et al. / Social Science & Medicine 66 (2008) 2076e2087

Bloom, G. (2005). Health policy during China’s transition to a market Grabowski, R. (1998). Development, markets and trust. Journal of
economy. In R. Guald (Ed.), Comparative health policy in Asia- Development Studies, 10, 357e371.
Pacific. Maidenhead: Open University Press. Granovetter, M. (1985). Economic action and the social structure: the
Bloom, G., Fang, J., Fang, L., Ren, J., & Wu, H. (2006). China’s problem of embeddedness. American Journal of Sociology, 91(3),
rural hospitals in the transition to a market economy. [Unpub- 481e510.
lished paper]. Brighton: Institute of Development Studies. Haas-Wilson, D. (2001). Arrow and the information market failure
Bloom, G., & Standing, H. (2001). Pluralism and marketisation in in health care: the changing content and sources of health care
the health sector: Meeting needs in contexts of social change in information. Journal of Health Politics and Law, 26(5),
low and middle-income countries. [IDS Working Paper No. 1031e1044.
136]. Brighton: Institute of Development Studies. Hall, M. (2005). The importance of trust for ethics, law, and public
Cederlof, C., & Tomson, G. (1995). Private pharmacies and the policy. Cambridge Quarterly of Healthcare Ethics, 14, 156e167.
health sector reform in development countries: professional and Hall, M. A. (2001). Arrow on trust. Journal of Health Politics and
commercial highlights. Journal of Social and Administrative Law, 26(5), 1131e1144.
Pharmacy, 12, 101e111. Hall, M. A., Dugan, E., Zheng, B., & Mishra, N. K. (2001). Trust in
Centre for the Future State. (2005). Signposts to more effective states. physicians and medical institutions: what is it, can it be
Brighton: Institute of Development Studies. measured, and does it matter? The Milbank Quarterly, 79(4),
Chee, H. L. (2008). Ownership, control and contention: challenges 613e639.
for the future of health care in Malaysia. Social Science & Helman, C. G. (2007). Culture, health and illness, (5th ed.). London:
Medicine, 66(10), 2145e2156. Hodder Arnold.
Coleman, J. S. (1987). Norms as social capital. In G. Radnitzky, & Hsiao, W. (2000). What should macroeconomists know about health
P. Bernholz (Eds.), Economic imperialism: The economic care policy? A primer. [IMF Working Paper Series WP/00/136].
approach applied outside the field of economics. New York: Washington: International Monetary Fund.
Paragon House Publishers. Joshi, A., & Moore, M. (2004). Institutionalised co-production:
Cornwall, A., & Shankland, A. (2008). Engaging citizens: lessons unorthodox public service delivery in challenging environments.
from building Brazil’s National Health System. Social Science Journal of Development Studies, 40(4), 31e49.
& Medicine, 66(10), 2173e2184. Kickbusch, I. (2004). Health and citizenship: the characteristics of
Davies, H., & Rundall, T. G. (2000). Managing patient trust in 21st century health. World Hospitals and Health Services,
managed care. The Milbank Quarterly, 78(4), 609e623. 40(4), 12e14.
Davies, H. T. O., & Dibben, M. (2001). Exploring trust in the public Knowledge@Wharton. (2005). The frenetic expansion of phar-
sector. Social Policy Focus, 1(4), 10. macy chains in Latin America. Wharton School at the Univer-
Davies, H., & Mannion, R. (2000). Clinical governance: striking sity of Pennsylvania. <http://www.wharton.universia.net/index.
a balance between checking and trusting. In P. S. Smith (Ed.), Re- cfm?fa¼viewfeature&id¼943&language¼english> (Accessed
forming markets in health care reform: An economic perspective. 10.12.06).
Milton Keynes: Open University Press. Kuhlmann, E. Traces of doubt and sources of trust e health
Detmer, D., Singleton, P., MacLeod, A., Wait, S., Taylor, M., & professions in an uncertain society. Paper presented at the 6th
Ridgwell, J. (2003). The informed patient. [Report]. Cambridge: conference of the European Sociological Association, Research
Cambridge University Health. Network Professions, September 23e26, 2003, Murcia, Spain.
Dieleman, M., & Harnmeijer, J. (2006). Improving health worker Leach, M., Fairhead, J., Millimouno, D., & Diallo, A. (2008). New
performance: In search of promising practices, evidence and therapeutic landscapes in Africa: parental categories and prac-
information for policy. Geneva: Department of Human Resources tices in seeking infant health care in the Republic of Guinea.
for Health, World Health Organisation. Social Science & Medicine, 66(10), 2157e2167.
Eggleston, K., Li, L., Meng, Q., Lindelow, M., & Wagstaff, A. Leonard, K. (2007). Learning in health care: evidence of learning
(2006). Health service delivery in China: A literature review. about clinician quality in Tanzania. Economic Development and
[World Bank Policy Research Working Paper 3978]. World Bank. Cultural Change, 55(3), 533e555.
Ensor, T., & Weinzierl, S. (2006). A review of regulation in the health Leonard, K. L. (2002). When both states and markets fail:
sector in low and middle income countries. [Oxford Policy asymmetric information and the role of NGOs in African health
Management Working Paper]. Oxford: Oxford Policy Management. care. International Review of Law and Economics, 22(1), 61e80.
Fang, J. (2006) Tailoring a western coat for a Chinese body: Leonard, L. (2005). Where there is no state: household strategies for
understanding the accountability of a health system in transition. the management of illness in Chad. Social Science & Medicine,
PhD thesis, University of Sussex, UK. 61, 229e243.
Fukuyama, F. (2004). State building: Governance and world order in Liu, X., Hotchkiss, D., Bose, S., Bitran, R., & Giedion, U. (2004).
the twenty-first century. Cornell: Cornell University Press. Contracting for primary health services: Evidence on its effects
George, A. S. (2007) The outrageous as ordinary: health workers’ and a framework for evaluation. [Partners for Health Reform
perspectives on accountability in primary health care in Koppal Plus]. Bethesda, MD: Abt Associates.
district, Karnataka State, India. DPhil thesis, University of Loevinsohn, B., & Harding, A. (2005). Buying results? Contracting for
Sussex, UK. service delivery in developing countries. Lancet, 366, 676e681.
Gilson, L. (2003). Trust and the development of health care as a social Loewenson, R. (2003). Annotated bibliography on civil society and
institution. Social Science & Medicine, 56(7), 1453e1468. health. [CSI/2003/BI2]. Geneva: World Health Organization
Gilson, L., Palmer, N., & Schneider, H. (2005). Trust and health Civil Society Initiative.
worker performance: exploring a conceptual framework using Londono, J., & Frenk, J. (1997). Structured pluralism: towards an
South African evidence. Social Science & Medicine, 61, innovative model for health system reform in Latin America.
1418e1429. Health Policy, 41, 1e36.
 G. Bloom et al. / Social Science & Medicine 66 (2008) 2076e2087 2087

Lucas, H. (2008). Information and communications technology for Platteau, J. P. (1994b). Behind the market stage where real markets
future health systems in developing countries. Social Science & exist e Part II: The role of moral norms. Journal of Development
Medicine, 66(10), 2122e2132. Studies, 30(4), 753e814.
Lyon, F. (2000). Trust, networks and norms: the creation of social Prata, N., Montagu, D., & Jeffeys, E. (2005). Private sector, human
capital in agricultural economies in Ghana. World Development, resources and health franchising in Africa. Bulletin of the World
28(4), 663e681. Health Organisation, 83, 274e279.
Mackintosh, M. (2006). Commercialisation, inequality and the Pritchett, L., & Woolcock, M. (2004). Solutions when the solution is
limits to transition in health care: a Polanyian framework for the problem: arraying the disarray in development. World
policy analysis. Journal of International Development, 18, Development, 32(2), 191e212.
393e406. Quill, T. E., & Brody, H. (1996). Physician recommendations and
Mackintosh, M., & Tibandebage, P. (2002). Inclusion by design? patient autonomy: finding a balance between physician power
Rethinking health care market regulation in the Tanzanian and patient choice. Annals of Internal Medicine, 125(9), 763e769.
context. Journal of Development Studies, 39(1), 1e20. Drug vendors and their market: the commodification of health. In
Mechanic, D. (1998). The functions and limitations of trust in the S. Reynolds Whyte, S. Van der Geest, & A. Hardon (Eds.), The
provision of medical care. Journal of Health Politics and Law., social lives of medicine (pp. 79e103). CUP.
23(4), 661e685. Rothman, D. J. (2001). The origins and consequences of patient
Meesen, B., Musango, L., Kashala, J., & Lemlin, J. (2006). Review- autonomy: a 25-year retrospective. Health Care Analysis, 9(3),
ing institutions of rural health centres: the Performance Initiative 255e264.
in Butare, Rwands. Tropical Medicine and International Health, Saltman, R., & Ferroussier-Davis, O. (2000). The concept of
11(8), 1303e1317. stewardship in health policy. Bulletin of the World Health
Mills, A., Brugha, R., Hanson, K., & McPake, B. (2002). What can Organisation, 7(6), 732e739.
be done about the private health sector in low-income countries? Segall, M. (2000). From co-operation to competition in national
Bulletin of the World Health Organisation, 80(4), 325e330. health systems e and back? Impact on professional ethics and
Montagu, D. (2002). Franchising of health services in low-income quality care. International Journal of Health Planning and
countries. Health Policy and Planning, 17(2), 121e130. Management, 15, 61e79.
Montagu, D. (2003). Accreditation and other external quality Sen, G., George, A., & Östlin, P. (2002). Engendering international
assessment systems for health care, DFID Health Systems health e The challenge of equity. Cambridge: MIT Press.
Resource Centre Working Paper. Stiglitz, J. (2002). Globalization and its discontents. London:
Moore, M. (1994). How difficult is it to construct market relations? A Penguin Books.
commentary on Platteau. Journal of Development Studies, 30(3), Tauber, A. L. (2003). Sick autonomy. Perspectives in Biology and
818e830. Medicine, 46(4), 484e495.
National Health Accounts. (2007). Summaries of country data. Tibandebage, P., & Mackintosh, M. (2005). The market shaping of
Available from. <http://www.who.int/nha/country/en/>. charges, trust and abuse: health care transactions in Tanzania.
Oladepo, O., Salami, K. K., Adeoye, B. W., Oshiname, F., Ofi, B., & Social Science & Medicine, 61, 1385e1395.
Oladepo, M., et al. (2007). Malaria treatment and policy in Wood, G. (2004). Informal security regimes: the strength of relation-
Nigeria: the role of patent medicine vendors. Future Health ships. In I. Gough, & G. Wood (Eds.), Insecurity and welfare
Systems Working Paper No. 1. Available from. <http://www. regimes in Asia, Africa and Latin America. Social policy in
futurehealthsystems.org/publications/index.htm>. development contexts. Cambridge: Cambridge University Press.
Palmer, N., Mills, A., Wadee, H., Gilson, L., & Schneider, H. (2003). Woolcock, M. (2001). The place of social capital in understanding
A new face for private providers in developing countries: what social and economic outcomes. Isuma: Canadian Journal of
implications for public health? Bulletin of the World Health Policy Research, 2(1), 1e17.
Organization, 81(4), 292e297. World Bank. (1999). What is social capital?. [PovertyNet]. Available
Pierson, P. (2000). Increasing returns, path dependency, and the from World Bank. <http://www.worldbank.org/poverty/scapital/
study of politics. American Political Science Review, 94, whatsc.htm>.
251e267. World Bank. (2003). Making services work for poor people.
Pierson, P. (2003). Big, slow-moving and. Invisible. In J. Mahoney, & Washington: World Bank.
D. Rueschemeyer (Eds.), Comparative historical analysis in the Yang, D. (2004). Remaking the Chinese Leviathan: Market transition
social sciences. Cambridge: Cambridge University Press. and the politics of governance in China. Stanford: Stanford
Platteau, J. P. (1994a). Behind the market stage where real markets University Press.
exist e Part I: The role of public and private order institutions. Zwi, A., Brugha, R., & Smith, E. (2001). Private health care in
Journal of Development Studies, 30(3), 533e577. developing countries. British Medical Journal, 323, 463e464.