SUB NATIONAL GOVERNANCE BETWEEN NETWORKS AND HIERARCHY ... · SUB NATIONAL GOVERNANCE BETWEEN NETWORKS AND HIERARCHY – COMPARING THE ... (cf. Kooiman 1993 & 2000 ... policy networks

SUB NATIONAL GOVERNANCE BETWEEN NETWORKS AND HIERARCHY – COMPARING THE LOCAL

CO-ORDINATION OF HEALTH SERVICES IN BRITAIN AND GERMANY

Viola Burau*

FIRST DRAFT Comments welcome.

Please do not quote without permission. Paper to be presented at the workshop on ’Policy Networks in Sub National Governance: Understanding Power Relations’ at the ECPR Joint Sessions, 13-18 April 2004, Uppsala * Department of Political Science, University of Aarhus, Bartholins Allé, DK-8000 Aarhus C,

Denmark Tel +45 89 42 54 32, fax +45 86 13 98 39, email [email protected]

1

Introduction

The ‘governance school’ (Börzel 1998) describes policy networks as a third ‘logic of co-

ordination’, which is distinct from both markets and hierarchies (cf. Kooiman 1993 & 2000 & 2003,

Kickert 1993 & 1995, Rhodes 1996 & 1997, Mayntz 1993a & 1993b, Mayntz and Scharpf 1995a &

1995b). Governance through networks relies on interdependent relations among autonomous actors,

who, on the basis of shared interests, exchange resources to achieve a common goal. Horizontal co-

ordination among actors is the key characteristic of network-based governance and points to a

considerable capacity of actors for self-organisation. Not surprisingly, network-based governance

emerges as a form of co-ordination, which is outside hierarchical control. Here, Rhodes’ (1996,

1997, 2000) influential characterisation of policy networks as ‘self-organizing’ and as ‘governing

without government’ is indicative.

However, recent empirical studies suggest that network-based governing is closely intertwined

with vertical power relations and steering opportunities of the state. This echoes Pierre and Peters’

(2000, also Peters 2000) critique of society-centred notions of governance and their plea for a more

state-centric understanding of governance. On the basis of their study of old age care policies in

Sweden Johnansson and Borrel (1998), for example, suggest that central government has a variety

of control and steering mechanisms at its disposal. Government control has not disappeared but

taken on different forms. Similarly, Bache’s (2000) study of urban generation in the UK highlights

that government itself can be a pivotal actor in policy networks. In their study of urban partnerships

in the UK Lowndes and Skelcher (1998) go even further and suggest that different forms of

governance prevail at different points of the life cycle of inter-organisational networks.

The paper contributes to this strand of the literature by critically examining the dynamics of

network-based governance through a comparative analysis of the local co-ordination of health

services in Britain and Germany. This is a particularly powerful test case, as the local-co-ordination

of health services lends itself to network-based co-ordination, yet is embedded in multi-level

governance structures, which also differ significantly between the two countries. The analysis

addresses two questions. First, to what extent is the local co-ordination of health services

characterised by network-based governance? Second, in what ways do vertical relations impact on

the (horizontal) co-ordination among local actors? This is closely related to the more general

question as to how best to achieve co-ordination among (local) actors. The analysis looks at the

local co-ordination of health services as a policy outcome and uses evaluation reports, government

documents together with existing studies to assess the relative importance of network-based

2

governance and the opportunities of the centre to steer. The cross-country comparison indicates that

the local co-ordination of health services has some characteristics of network-based governance, but

often depends on steering from the centre. In conceptual terms, this suggests that the analysis needs

to look more closely at the context of inter-organisational relations, at both horizontal and vertical

levels.

Network-based governance and hierarchies

At a basic level, network-based governance can be defined as a form of informal co-ordination,

which involves relations among autonomous yet interdependent actors, who co-ordinate their

interests through the exchange of resources (cf. Börzel 1998, Marin and Mayntz 1991). The

autonomy of actors makes for co-ordination that is primarily non-hierarchical and takes place at

horizontal level, not least because no single actor has sufficient capacity to make policy on its own

(cf. Kenis and Schneider 1991, Kickert 1995). Instead, the network perspective focuses on the

processes and structures of inter-organisational relationships.

In the literature on network-based governance vertical co-ordination and especially the state

tends to be downplayed and remains ambivalent at best. In part, this reflects the genesis of the

notion of network-based governance, which is closely tied to the debate about the failure of

hierarchies as instruments of government (cf. Mayntz 1993a). The crises of public finances and the

welfare state, in particular questioned the effectiveness of instruments, which conceive of governing

as a top-down, one-way process (cf. Pierre and Peters 2000, Flynn 2000). In terms of problem

diagnosis the literature points to the fact that as power becomes increasingly dispersed in highly

functionally differentiated, hierarchical governance looses its steering capacity, while networks of

organised actors offer a new source for governing. (cf. Kenis and Schneider 1991, Mayntz 1993a,

Rhodes 1997). Network-based governance responds to this heightened complexity by combining

elements of both market and hierarchy and as such making for a distinct mode of governing:

network-based governance brings together a broad range of actors (as in markets), who work to

pursue a common goal (as in hierarchies). Seen in this way, network-based governance combines

the capacity of actors to act responsibly (that is collectively) without being forced (Mayntz 1993b).

The nature of network-based governance as a more strongly society-based form of co-ordination is

particularly explicit in the work of Kooiman (1993, 2000, 2003) and his concept of governing as

‘social-political governance’. As such, the governing capacity of networks explicitly emerges from

beyond central authority.

3

Here, Rhodes’ work is indicative (1996, 1997, 2000). His starting-point is the observation that

the state has been hollowed out from below and above as well as from within and that British

politics has gradually moved from the centralised ‘Westminster’ model to a ‘differentiated polity’

model; many centres of government link many levels of government. This reflects both the limits of

central steering and the fact that there are many policy areas and issues government cannot steer.

Instead, the aggregation of interests through networks is a functional necessity in a polity, which

lacks a clear centre. By way of a base line, Rhodes therefore characterises network-based

governance as ‘governing without the state’; policy networks are self-governing and resistant to

central direction and as such enjoy considerable autonomy from the state. However, he

acknowledges that relations between government and networks are asymmetric, but insists that

centralisation co-exists interdependence and introduces the notion of ‘asymmetric interdependence’

(1997:15). One way in which this tension manifests itself is the fact that in the differentiated polity

the state has more control over less; the decentralisation of service delivery co-exists with

centralisation of financial control (Rhodes 1996: 662). Rhodes (1997: 53) also notes that it is the

state, which enables networks and sets the parameters for network actors, although steering is bound

to remain indirect and imperfect.

A similar ambivalence underlines Kickert’s (1995) notion of ‘steering at a distance’, whereby

the state as co-ordinator governs through influencing networks. One the one hand, the concept

points to the existence of a non-hierarchical form of steering. Yet on the other hand, Kickert

emphasises that steering by incentives is almost equivalent to non-steering, reflecting the self-

organising capacity of networks (Kickert 1993). As Hogget (1996) argues the focus on the dispersal

of power means that the concept of ‘steering at a distance’ has difficulties seeing regulation and

autonomy together, being less sensitive to new forms of power.

In contrast, in his recent work Kooiman (2000, 2003) addresses the relationship between

horizontal and vertical forms of co-ordination more explicitly. He argues that what he calls ‘social-

political governance’ consists of different modes of governing including: self-governance, co-

government and hierarchical governance. Kooiman suggest that over recent decades governing has

shifted to the first two modes of governance, although hierarchical forms of co-ordination still play

an important role. Significantly, however, vertical forms of co-ordination have been transformed

(from command to regulation, and from procuring to enabling) and the relationship with network-

based governance is one of residual and quite separate co-existence.

4

The local co-ordination of health services – Britain and Germany compared

The (local) co-ordination of health services is a salient policy issues, reflecting traditional divisions

between different sectors of health care delivery, especially between ambulatory and hospital

services as well as between medical and non-medical care services. At the same time, the (local) co-

ordination of health services has become a buzz word in current health policy debates, as it

promises both higher quality and higher cost efficiency (cf. Burau 2004, Kühn 2001, Kümpers et al.

2002, Peckham and Exworthy 2003, Verdi/PLS Rambøll without year). The increasing

specialisation of health services, together with the growing number of patients with multiple

illnesses (especially older people) has highlighted the quality aspect of service co-ordination. This

coincides with a greater interest in quality management in health policy more generally (cf. Blank

and Burau 2004) and here, service co-ordination is associated with such notions as ‘integrated’ and

‘seamless’ care. The focus on cost containment for its part has highlighted the efficiency aspect of

service co-ordination. The underlying argument is that the growing specialisation of health services

leads to ‘parallel’ service provision (and potential oversupply), not least because chains of service

delivery become longer and more complicated. Here, health service co-ordination offers the

possibility of shifting tasks to their ‘appropriate’ place and freeing funds for additional services.

The co-ordination of health services can occur at different levels, although service co-

ordination at the local level is of particular relevance; it is here, that problems relating to the

interfaces between different services occur (cf. Badura et al. 2000). Problems of co-ordination are

also likely to be locally specific to some extent, reflecting the specific needs of local populations

and the specific features of local service infrastructures. The local level of co-ordination has also

come into focus as part of a strong trend towards decentralisation underpinning health reforms

across Europe over the last two decades, which have been concerned with the fine-tuning of the

planning and delivery mechanisms (cf. Freeman 2000). The debate about the ‘regionalisation’ of

health services in Germany is even directly related to issues about strengthening the (local) co-

ordination of health services (Kühn 2001).

The local co-ordination of health services has many meanings, but broadly describes the co-

ordination of services as well as the collaboration among provider organisations across different

sectors (cf. Glendinning 2002, Kühn 2001). The local co-ordination of health services can take a

variety of forms reflecting different extents of co-ordination among local actors and here Hudson et

al. (1998 as quoted in Peckham and Exworthy, 2003: 181) distinguish between the following:

‘situation and encounter’, where interactions between organisations are infrequent, with

5

organisational goals and interests remaining divergent; ‘communication’, where separate

organisations engage in some formal/informal joint working; ‘collaboration’, where joint working is

a central aspect of the mainstream activities of organisations; and, finally, ‘integration’, where the

boundaries between organisations become blurred. The following analysis uses these types of

service co-ordination to map out the extent of the local co-ordination of health services in Britain

and Germany over the last decade.

Britain: pushing the boundaries of health care1

In Britain, debates about the local co-ordination of health services have often focused on the divide

between health and social care services, which has come to be described as a ‘Berlin Wall’ (cf.

Means et al. 2003). Attempts to co-ordinate services across the health and social care divide have

had a long history and have tended to concerned with either the strategic co-ordination at the level

of organisations (such as joint planning) or operational co-ordination at practitioner level (Rummery

and Coleman 2003). The divide between the two services became even more visible following the

introduction of the purchaser-provider in the early 1990s and its emphasis on provider competition.

split, but also more difficult to address (cf. Kümpers et al. 2003, Mur-Veeman et al. 2003). Against

this background and the perceived failure of earlier initiatives, the 1990 Community Care Act

aimed to redefine the health and social care divide with social services acting as the lead agency and

health authorities having more narrowly defined responsibilities (Means et al. 2003). Organisational

goals and interests remained divergent, tough, typical of the ‘situation and encounter’ type of

service co-ordination.

In practice, the local co-ordination of health and social care services has tended to be strongest

in relation to community nursing and social services and has hardly involved general (medical)

practitioners (GPs) working in ambulatory settings (Glendinning et al. 2001). However, this

changed with the increasing primary care orientation of the British National Health Service (NHS),

which has been strengthened by the parallel but related trend towards a more decentralised

organisation of health services (Moon and North 2000). In the context of local health services, GPs

have traditionally played a key role as gatekeepers to hospital and other specialist services, and the

process of transferral can be seen as a forerunner of service co-ordination. But GPs have

1 Devolution has become a distinct feature of health policy in the Britain over recent years and the following analysis focuses on developments in England. However, as the main focus is on the cross-country comparison with Germany, the text will primarily use the term ‘Britain’ and only refer to ‘England’ organisational features specific to this part of Britain are concerned.

6

traditionally been quite separate organisationally pointing to their marginal position in a health

service dominated by hospitals.

The situation has been changing following the introduction of the purchaser-provider split and

GPs have gradually become the centre of the local health service. This process has helped both to

extend and define more clearly the local co-ordination of health services. Building on their role as

gatekeepers many GPs in England opted to become ‘fundholders’ and to purchase an increasingly

wider range of hospital and other specialist services for their patients. Health service co-ordination

only became an explicit aim of the fundholding scheme in 1996, when ‘partnerships’ emerged as

central theme for health care delivery, marking a turn away from the competitive orientation of the

purchaser-provider split (Peckham and Exworthy, 2003: 187). Nevertheless, service co-ordination

was in some ways implicit in the process of contracting. Contracting requires the definition not only

of services, but also the division of labour between different services, and as providers fundholding

GPs also have a strong interest in pushing the second aspect.

In this way, contracting made for more intense contact between different providers, loosened

some of the boundaries between services (what Peckham and Exworthy refer to as the ‘unfreezing

of entrenched organisational positions’ (2003: 151)) and as such paved the way for more extensive

service co-ordination of the ‘communication’ type. Here, it was crucial that the contractual space of

GPs (initially as fundholders and later as part of so-called ‘multi-funds’) gradually expanded, to

include an increasingly greater share of the budget, community nursing services and finally all

elective surgery and outpatient services (Moon and North 2000). In the mid-1990s this culminated

in so-called ‘total purchasing’ initiatives, where consortia of GPs were responsible for the entire

range of services for the patients on their list. It is also indicative that these developments gradually

transformed ‘ambulatory’ into more broadly defined ‘primary care’ services, while at the same time

allowing for a more strongly locally-oriented health service.

In terms of the local co-ordination of health service, the introduction of primary care groups

(and trusts from 2000) in England in the late 1990s under the new Labour government was

significant in two respects: the trusts have much extended purchasing and planning responsibilities

in relation to health services, and they are also responsible for integrating health and social care

services. This also means that the two streams of co-ordinating local health services have been

brought together. In contrast to the fundholding schemes membership in the new primary care trusts

is compulsory and the new provider organisations are also responsible for a greater number of

patients. In formal terms, service co-ordination thus increasingly resembles the ‘communication’

7

and in some respects even the ‘collaboration’ type of co-ordination. The trusts provide community

nursing and in addition to primary medical services, and commission the entire range of health

services for the population of their locality. Significantly, the trusts are also responsible for the

development of primary and community care services in their area, and for improving the health of

the local population. This responsibility for public health is likely to provide a catalyst for further

service co-ordination and notably with a broader range of local actors, including: local authority

departments other than social services, voluntary agencies as well private organisations. Against

this background, Peckham and Exworthy (2003: 18) suggest that primary care trusts are at the

vanguard of local co-ordination, significantly through both intra and inter-organisational structures

and processes.

In addition, the introduction of primary care trusts co-exists with a range of additional

mechanisms to promote co-ordination among different provider organisations, including: health

improvement programmes, local strategic partnerships, health action zones, and the possibility of

pooling budgets of health and social care services (Hudson and Henwood 2002). The 2000 NHS

Plans is also significant here, as it pushes, at least at the formal level, local service co-ordination

further towards ‘collaboration’ and even ‘integration’. This is reflected in the introduction of a ‘duty

of co-ordination’ and the introduction of the new care trusts, that is joint provider organisations for

health and social care, which also include more specific organisational set-ups, such as rapid

response teams and integrated home care teams (Hudson and Henwood 2002).

Germany: isolated and loose local co-ordination of health services

In comparison, the local co-ordination of health services in Germany is both more isolated and

more loosely defined. Office-based doctors, who compromise both generalists and specialists,

hospitals and nurses working in care and people’s homes all provide health services at local level.

However, neither collectively as a group nor as individual local actors do they have any

responsibility beyond the local provision of their own specific service. At the same time, the

responsibilities of local authorities in health care are defined only vaguely and following the

introduction of the long-term care insurance in the mid 1990s local authorities also lost the

responsibility for ensuring the adequate provision for long-term care (Geiser 1995). Any local co-

ordination of health services therefore emerges in a more ad hoc and informal way, relating to

specific local service issues and focusing on establishing the ‘situation and encounter’ type of

service co-ordination.

8

As in Britain, the problem of the local co-ordination of health services has been highlighted by

the needs of older people, which cut across the divides among different services. In the case of

Germany, this is the divide between acute health care and long-term (nursing) care, which has

become even more apparent with the introduction of a separate insurance for long-term care in the

mid-1990s. The legislation itself in effect defines long-term care as a joint responsibility: the

insurance funds have to work together with other actors (§8 SGB XI), in order secure the provision

of long-term care services (Sicherstellungsauftrag). To this end insurance funds have the possibility

of setting up working groups at local and regional level (§12 SGB XI), further details of which are

regulated by state legislation. Here, the state of Northrhine-Westfalia has the most extensive

legislation (Oldemeyer without year) and it is the responsibility of the local authorities to set up so-

called ‘long-term care conferences’ (Pflegekonferenzen) at local level. The care conferences include

mainly different providers, third party payers and users, but also other actors as and when needed.

The care conferences are forums for exchanging views and identifying problems and as such a

platform for agreements over issues of service co-ordination. Here, strengthening interactions

among organisations is key (typical of the ‘situation and encounter’ type of service co-ordination),

whereas the ‘communication’ type of co-ordination is a more remote goal.

The thrust of the locally-based advisory or clearing centres for long-term care, which have

emerged in many states, is similar, although the focus on service co-ordination is less explicit (for

an overview see Igl et al. 1995). In the case Baden-Wüttemberg, for example, the principal focus of

the local advisory centres is to facilitate the access of patients to suitable care provision, through

providing information about local providers and also advise on a case-by-case basis (Wendt 1995).

This includes the co-ordination of services (for example in relation to weekend coverage), although

not always across different sectors. However, the regular contact among providers in relation to

individual cases may offer a basis for developing more structured inter-organisational co-ordination

(Moscher 1995).

In some states, the long-term care conferences have inspired similar conferences in relation to

health care. A prominent example is again the state of Northrhine-Westfalia, where local and state

level ‘health care conferences’ (Gesundheitskonferenzen) have been established as an instrument of

public health policies (and are embedded in the relevant state legislation on public health). The

conferences take the form of round tables and thematic working groups, and are supported by an

office to facilitate a continuous and stable process of exchange (and co-ordination) among local

actors. The aim of the conferences is to discuss problems relating to health services at local level

9

and to formulate recommendations for action (Landesinstitut für den öffentlichen Gesundheitsdienst

without year). The recommendations are not legally binding and as such the conferences very much

depend on the (voluntary) commitment and compliance of local actors (Knesebeck et al. 2002).

Consequently, the joint nature of the process (of discussing problems, identifying possible solutions

and formulating recommendations) becomes an important condition for the relative success of

conferences as co-ordinating mechanisms. Significantly, an evaluation of a three year pilot project

indicates that the conferences are most successful where the local co-ordination of services is the

explicit aim and where services are concerned, which are at the margins of the responsibilities of

the self-administration, such as health promotion and mental health (Badura et al. 2000).

The local co-ordination of ambulatory and hospital care more specifically, has also been

supported by the fact that the boundaries between the two sectors have become weaker over recent

years. For example, hospitals are now allowed to provide pre admission diagnostic procedures, post

discharge treatment as well as one-day surgery (Wasem 1997). Building on this, some local service

co-ordination occurs around specific diseases reflecting the growing importance of fees based on

diagnosis-related groups (Verdi/PLS Rambøll without year). Other instances of local service co-

ordination centre around consortia of office-based doctors, which sometimes also include hospitals.

The consortia can have a variety of aims, from avoiding hospital stays by co-ordinating the work of

different specialists working in ambulatory settings to ensuring the seamless care of patients once

they are in hospitals by setting up an ambulatory care out-of-hours service (for an overview see

Verdi/PLS Rambøll without year). Legislative changes in 2000 further support the development of

such consortia, as it is now possible to complete contracts, which are separate from the normal (that

is collective) contract (Rosenbrock and Gerlinger 2004). This is significant as service co-ordination

becomes a more explicit focus of inter organisational relations, which also become more formal

(and as such pushing towards the ‘communication’ type of service co-ordination). Nevertheless,

compared to England, the scale of such collaborative pilots remains small: the report by Verdi/PLS

Rambøll (without year) estimates that there exist about 3000 consortia and that in 2002 only 26 of

these were based on a specific contract.

Horizontal relations in the local co-ordination of health services

As illustrated by the comparative analysis of Britain and Germany, any local co-ordination of health

services is a complex task, and involves a wide range of actors (reflecting the complexity of modern

health care) drawn from across the public-private private (reflecting the increasing welfare-mix in

10

many health systems). By definition, the local co-ordination of health services is also locally

specific. As such, the local co-ordination of health services is likely to lend itself to informal, non-

hierarchical forms of governing in form of network-based governance (cf. Hudson and Henwood

2002, Rummery and Coleman 2003). This would suggest that the greater extent of local co-

ordination of health services in Britain reflects particularly strong horizontal co-ordination among

local actors, whereas network-based governing is weaker in Germany. However, the situation is far

from clear cut.

In Britain, the local co-ordination of health services has become increasingly formalised and

characterised by clear asymmetries of power. Service co-ordination occurs in different situations: as

part of the formal duty of providers to work in partnerships; as part of the formal responsibilities of

primary care trusts to improve the health of their populations; and as part of mergers of provider

organisations (as in the case of community care and in future social care). As such, service co-

ordination is based on a clear sense organisationally embedded obligation, with primary care trusts

being in the driving seat. Here it is also significant, that the extent of co-ordination increased (from

‘situation and encounter’ towards ‘communication’, some ‘collaboration’ and even ‘integration’) as

the hierarchical elements became stronger.

In Germany, by contrast, the local co-ordination of health services, as and when it exists, tends

to be informal and underpinned by the autonomy of local actors; none of the local actors can force

the other to join any efforts of service co-ordination. Service co-ordination occurs as part of local

‘conferences’ or within service consortia, bringing together local actors to identify problems and

possibly work towards common solutions. The underlying expectation is that the joint nature of this

process makes for joint working in practice and as such the conferences and consortia in effect only

act a springboard for service co-ordination.

In sum, the comparison of Britain and Germany seems to suggest that there is more local co-

ordination where there is more hierarchy and formalisation, and less local co-ordination where there

is more autonomy and more informality. How can this be explained? Further, what does this say

about network-based governance? In the following, I argue that the comparison of Britain and

Germany points to the context of inter-organisational relations and more specifically to the specific

factors enabling or/and prohibiting co-ordination among local actors. Significantly, such factors

exist not only in relation to horizontal relations, but also vertical relations.

The literature suggests that shared interests together with resource dependencies among actors

are the central building blocks for the formation of inter-organisational relations and as such an

11

important condition for the emergence (and existence) of network-based governance. However, the

comparative analysis of the local co-ordination of health services in Britain and Germany indicates

that shared interests and resource dependencies among local actors are not always strong enough to

provide a firm basis for co-ordination among actors. The interests of local actors in health service

co-ordination are often diffuse and existing resource dependencies are clouded over by

organisational and professional divisions.

Diffused interests

The interests of local actors in the (local) co-ordination of health services fall into two categories.

The first, normative set of interests in service co-ordination relates to ‘higher quality health

services’, and is shared widely among local actors reflecting the positive connotations of ‘quality’

as well as overlaps with professional service ethics (cf. Hudson 2002b, Rummery and Coleman

2003). Yet, higher quality services do not necessarily yield any specific benefits for individual local

actors, and local actors are unlikely to be motivated to co-ordinate services by the promise of higher

quality alone as the salience of the debate about the local co-ordination of health services

demonstrates. It is also indicative, for example, that in Britain it is with the advent of the primary

care trusts that there is a local actor whose normative interest in service co-ordination is

underpinned by formal responsibilities, which in turn are firmly integrated into the quality

management structure of the health service as a whole.

The second, pragmatic set of interests in service co-ordination relates to the possibility of cost

savings through avoiding overlaps among different services. In comparison to higher quality

services this has a more direct appeal to local actors, especially under circumstances of tight

financial resources. In this case, service co-ordination is no longer the primary aim of local actors,

but a means to achieve something else, and if funding is highly constrained, strengthening the local

co-ordination of health services is in danger of focusing primarily on ‘getting rid of tasks’ rather

than co-ordinating services as such (Rummery and Coleman 2002). In the context of Britain this is

illustrated by the now classical debate as to whether a bath is a health or a social care activity,

reflecting the economic pressures the purchaser-provider split put on providers of social and

community health care (Hudson and Henwood 2002).

Tight financial constraints also sharpen the sensitivity of local actors for any additional costs

associated with service co-ordination. In their evaluation of local health care conferences in

Germany Badura et al. (2000), for example, found that following the introduction of budgets in

12

ambulatory care office-based doctors were more interested in experimenting with new ways of

delivering medical care than to invest in resources in cross-sector initiatives. Similarly, in

discussions about local health service co-ordination some hospitals showed considerable reluctance

reflecting economic pressures to reduce the number of beds and the threat of hospital closure. As

such, cost savings through greater service co-ordination can also be an explicit threat to local actors,

as they may loose funding as well as organisational territory. The last aspect is exacerbated by fact

that the provision of health services relies to a great extent on the work of professionals such as

doctors and nurses, who define their identity through boundaries and for whom service co-

ordination therefore easily becomes a threat. In his study of working relations among GPs,

community nurses and social workers in Britain Hudson (2002b), for example, found that such

professionally-rooted obstacles persisted despite the widespread insight into the benefits of service

co-ordination.

Clouded over resource dependencies

As discussed above the local co-ordination of health services requires, by definition, the exchange

of resources among local actors as modern health care is highly specialised. Individual local actors

cover different yet complimentary parts of the local division of health care and as such strong

resource dependencies exist among the local actors (cf. Hudson and Henwood 2002, Rummery and

Coleman 2003). Services by GPs and Social Service Departments in the UK and home care

providers and to some extent office-based doctors in Germany are a particularly interesting resource

for exchange, as health policy over recent years has emphasised the importance of ‘basic’ over more

specialised health care services (cf. Burau 1999). In the case of medical services, this applies

specifically to GPs, as they have a strong gate-keeping position referring patients to both specialist

and hospital services, and less to office-based doctors in Germany, who include both generalists and

specialists. The services offered by Social Service Departments and long-term care providers is an

even more attractive resource for exchange, since services are more low-tech and inexpensive

reflecting their Cindarella status in relation to medical and especially hospital services.

However, existing resource dependencies are often clouded over by the distinctive

organisational and professional orientations of different services, which as legacies continue to

shape the co-ordination among local actors (Glendinning et al. 2001, Mur-Veeman et al. 2003). As

mentioned above, in the context of the UK the divide between health and social care services has

been a salient issue (cf. Hudson 2002a, Hudson and Henwood 2002, Means et al. 2003).

13

Traditionally, health services have been organised hierarchically around medical expertise in form

of a National Health Service and have been at the centre of modern welfare states. As such, health

services are provided on the basis of need and are funded through general taxation. In contrast,

social care services have traditionally been organised by local authorities and focus on people’s

social problems in situations when all other forms of support have failed. This has put social

services at the margins of the welfare state and significantly services are needs-tested and are also

funded by co-payments.

The clouding over of resource dependencies is more pronounced in Germany. This reflects a

health service, which is divided into several distinct sectors, based on a plurality of funding sources

(Rosenbrock and Gerlinger 2004, Sachverständigenrat 2003). At a basic level, any service co-

ordination is made difficult by the fact that the different insurance funds use different definitions of

illness: the long-term insurance defines illness as chronic, the health insurance as acute and

therefore treatable, whereas the old age pension insurance focuses on rehabilitation. There are

additional fractions at service level, such as between services provided by office-based doctors and

nurses working in care homes/people’s homes. The two groups have traditionally been

characterised by highly gendered (that is unequal) working relations, which are now cross cut by

two separate insurance schemes, both of which are also under pressure to contain contribution rates.

Over recent years, both insurance schemes have tried to tighten up their commitment to long-term

care services leading, often sidelining issues of service co-ordination (cf. Bundesministerium für

Gesundheit und Soziale Sicherung 2001).

Vertical relations and the local co-ordination of health services

The analysis of horizontal relations in Britain and Germany demonstrates that there are limits to

network-based governing, reflecting the fact that the interests of local actors in health service co-

ordination are often diffuse and that resource dependencies among local actors are clouded over by

the existence of divisions in terms of organisational and professional orientations. However, the

comparative analysis also indicates that the formalisation of inter-organisational relations in Britain

seems to have strengthened health service co-ordination. This challenges the view in the literature

on network-based governance that co-ordination among actors occurs ‘naturally’ and instead points

to the importance of vertical relations of governing. This raises a number of questions: What are the

vertical relations in which the local co-ordination of health services in the two countries is

embedded? In what ways do vertical relations impact on the local co-ordination of health services?

14

Britain: structuring and steering from close-up

In England, a small set of actors spread over two levels makes up the vertical relations in which the

local co-ordination of health services is embedded (Department of Health 2004). At the top is the

Department of Health, which decides on the overall direction of health services, sets standards and

secures resources. Below are strategic health authorities, which are responsible for co-ordinating

local service planning and holding local providers to account. Here, the key mechanisms are the

strategic development of health services at local level and performance management.

Vertical relations are highly concentrated and integrated not only because of the small set of

actors, but also because the different levels are connected by strong hierarchical relations. The

strategic health authorities are creatures of statute and are directly accountable to the Department of

Health and as such enjoy only limited autonomy. This also points to the concentration of extensive

governing resources at the centre (cf. Blank and Burau 2004). The Department of Health has

considerable control over finances and also heads the ‘administration’ of the health service. This

reflects the fact that the health service is mainly funded out of general taxation as well as the

predominantly public nature of service provision. The health service is also embedded in a centralist

political system, which make for a high degree of institutional integration and the Department heads

the ‘administration’ of the health service. Strong institutional integration and the concentration of

governing powers mean that there are considerable resources for steering the co-ordination health

services at local level, especially of the hierarchical nature (Kümpers et al. 2002, Mur-Veeman et al.

2003). Since the late 1980s steering from the centre has moved towards increasingly closer steering

in form of mandating service co-ordination and crucially has also involved the restructuring of

inter-organisational relations itself (Hudson 2002, Hudson and Henwood 2002).

The introduction of the fundholding scheme meant that GPs became more strongly integrated

into the mainstream organisation of the NHS, making GP services more receptive to future reforms.

The contractual relations also helped to blur the boundaries between different services to some

extent and GP services came to be defined as primary care rather than medical care services. As

such, central government acted as an ‘architect of political order’ (Döhler 1995) and restructured

the inter-organisational relations at local level itself by repositioning GPs. However, it was left to

local actors to make use the newly created spaces for service co-ordination.

This changed when the government become more explicitly interested in the local co-

ordination of health services as part of its focus on ‘joined up’ governing, which consists of both

joined-up government at the centre and joined up governance (for example through public sector

15

partnerships) at the local level (cf. Newman 2001). The new primary care trusts have explicit and

now also mandatory responsibilities in relation to the co-ordination of health services. Relations

among local actors have become more formalised as a result, not least as relations are more strongly

integrated into policies of outcome management. The 2000 NHS Plan, for example, includes

performance measure for local partnerships, such as to reduce the number of delayed discharges

from hospitals. The restructuring of inter-organisational relations at local level has also become

bolder and has taken the form of structural integration, whereby community nursing services and

eventually also social services become subsumed under (primary) care trusts.

However, there are limits to this highly hierarchical approach to steering the local co-

ordination of health services. Although (primary) care trusts are about local service co-ordination,

they have to work in a prescribed organisational framework and towards specified goals in some

cases. In relation to care trusts, for example, central guidelines specify the terms, speed and

parameters of their creation (Hudson and Henwood 2002). In the face of the continuous flow of

changes it is also difficult to sustain existing or emerging inter-organisational relations at local

level. Rumery and Coleman (2003) therefore suggest that structural integration and force are not

sufficient to make the local co-ordination of health services happen. Hudson (2002a) goes even

further arguing that the imposition of structural change even undermines inter organisational co-

ordination by limiting the autonomy of local actors. This echoes Peckham and Exworthy’s (1999 as

quoted in Peckham and Exworthy, 2003: 193) findings that commitment and trusts are key

conditions for joint working.

Germany: stimulating at a distance

In Germany, by contrast, vertical relations consist of a wide range of different actors, which are

spread not only over several levels but also sector-specific arenas of governing (cf. Burau 1997,

2001). Here, the federal Ministry of Health is responsible for setting the framework of health care

services, whereas the specification and implementation of policies is in the hands of the joint self-

administration of insurance funds and providers, that is office-based doctors and hospitals. Federal

actors also share governing responsibility with the state level. Federalism together with self-

administration makes for fragmented and decentralised vertical relations and the centre has fewer

resources for steering the local co-ordination of health services. Steering therefore is more indirect

and takes the form of ‘stimulation at a distance’, leaving it to the self-administration and local

16

actors to react to these stimuli (for similar observations in relation to the Netherlands see Mur-

Veeman et al. 1999).

Significantly, the self-administration exists only at federal and state level, but not at local level

and changes over recent years have led to a further centralisation of governing at federal level (cf.

Burau 1997). This means that the local level enjoys very limited legitimacy in procedural terms.

The Basic Law also does not explicitly define the role of local authorities in health care, the only

exception being public health. Beyond that local authorities are allowed to regulate all issues

affecting their communities, but without interfering with the responsibilities of other actors in health

care (Badura et al. 2000).

A typical example of steering by ‘stimulating at a distance’ is the long-term care insurance. As

discussed earlier, the initial legislation itself stipulated the setting-up of cross-sectoral working

groups, but left the formulation of specific guidelines to the state governments. As a result, the

existence and form of such local working groups varies considerably across states (Oldemeier

without year). In a similar vein, the federal government initiated a series of pilot projects on

developing the provision of long-term care, a key aspect of which was to strengthen the regional co-

ordination across different providers and third party payers (Bundesministerium für Gesundheit und

soziale Sicherung 2001). In relation to both measures it is significant, that they only provide a

framework for (non-public) actors at lower levels to take the initiative and work towards the

strengthening service co-ordination, which above all is informal in nature.

However, as an evaluation of local health conferences in Northrhine-Westfalia shows, this

strategy is clearly limited in scope (Badura et al 2000). The local co-ordination focuses on ‘niche

services’, which are at the margins of the responsibilities of the health and long-term care insurance.

In contrast, it is next to impossible to change the structure of the local provision of health services

itself, as this tends to affect the responsibilities non-local actors. This also reflects the fact that

health governance is strongly legalistic in nature and as such rather inflexible.

In this respect, more recent federal initiatives indicate a slight change in strategy and focus on

more formal forms of service co-ordination (Rosenbrock and Gerlinger 2004, Verdi/PLS Rambøll

without year). The 1997 health reform included the possibility of pilot projects and so-called

‘structural contracts’, which are broadly aimed at increasing the cost efficiency and quality of health

services, but not specifically with service co-ordination itself. The importance of the two measures

lies in the fact that they create formal, contractual spaces for the insurance funds to (locally)

experiment with new forms of providing and funding services. Opening-up the legalistic framework

17

is important, as the structure of collective contracts has proved to be very inflexible. Building on

this, the 2000 health reform included provisions for a new type of contract, which explicitly aims to

develop a more co-ordinated service provision. The direct contracts between insurance funds and

providers may involve the transfer of budgets to provider organisations and some financial

incentives, putting activities concerned with service co-ordination on a more formal footing.

Nevertheless, the local co-ordination of health services remaims voluntary and loosely coupled. The

same applies to the measures of the 2004 health reform (Bundesministerium für Gesundheit und

Soziale Sicherung 2004), which offer insurance funds the possibility of establishing health centres,

allowing for the closer co-ordination of medical and non-medical services, and to contract hospitals

to offer care for certain diagnoses in ambulatory settings.

In this vacuum between ‘stimulation’ from the federal government and more or less informal

service co-ordination at local level states assume a potentially important intermediary role as the

example of Northrhine-Westafalia illustrates. The state has been something of a pioneer in

stimulating the (local) co-ordination of health services. In 1995 the state government initiated an

extensive pilot project on the ‘local co-ordination of health service’ (Ortsnahe Koordinierung der

Gesundheitsversorgung) (Badura et al. 2000). Even before the completion of the pilot project new

legislation on public health established local (together with state) health care conferences as a

regular component of state level (public) health policy. The state has also played a pioneering role

in putting long-term conferences on a firm legal footing (Oldemeier without year).

Sub national governance between networks and hierarchies

The literature suggests that network-based governance offers a particularly effective way of co-

ordination among organisational actors, as it combines elements of markets (involving a wide range

of actors) and hierarchies (pursuing a common goal). As such, network-based governance is

particularly well suited for complex issues, also referred as ‘wicked’ or ‘cross-cutting’ issues, which

require a ‘holistic approach’ combining re-integration and continued differentiation (Kümpers et al.

2002). The local co-ordination of such an issue: the complexity of health care means that service

co-ordination involves a wide range of actors, many of which enjoy considerable autonomy as

professionals; and the co-ordination of health services also has a strong horizontal component

reflecting its local specificity. This suggests that network-based governance is likely to facilitate the

local co-ordination of health services.

18

Table 1 The local co-ordination of health services in Britain and Germany

LOCAL CO-ORDINATION OF HEALTH SERVICES

MODES OF GOVERNING

high

low

horizontal relations

network-based

hierarchy-based

Britain primary care trusts

care trusts

Germany long-term/health care

conferences practice consortia

vertical relations

network-based

hierarchy-based

Britain close-up steering

structuring

Germany stimulating at a distance

However, the comparative analysis of Britain and Germany (summarised in Table 1 above)

presents a more ambivalent picture. The local co-ordination of health services seems to be stronger

where horizontal relations (among local actors) are based more on hierarchy than on networks, and

weaker where horizontal relations are based more on networks than on hierarchy. This points to a

19

number of factors, within horizontal but also vertical relations, inhibiting or/and promoting co-

ordination among organisations.

In the context of horizontal relations pragmatic but ultimately unstable interests in service co-

ordination often prevail over normative interests, making for rather diffused interests of local actors

in the co-ordination of health services. Similarly, existing resource dependencies among local actors

are often ‘clouded over’ by organisational and especially professional divisions. It is against this

background that the importance of vertical relations and steering from the centre becomes apparent.

The (increasingly) hierarchical approach taken by the central government in Britain helps to

counterbalance the inhibiting factors that exist at the horizontal (local) level by restructuring inter-

organisational relations (and especially repositioning individual actors) and backing up service co-

ordination by formal obligation. In contrast, the approach taken by the federal and state

governments in Germany is more hands-off and consists of ‘stimulation at a distance’. This

approach consists of creating communicative and increasingly contractual spaces, which local

actors can decide to use to work towards the strengthening of the local co-ordination of health

services. However, as the example of Germany demonstrates this approach often seems to less

successful at counterbalancing the inhibiting factors that exist at the horizontal level.

What are the implications for understanding governance at sub national levels? The

comparative perspective in particular highlights in the importance of the context of inter-

organisational relations at sub national levels. Looking at such contexts helps to identify the specific

factors promoting or/and inhibiting co-ordination among organisational actors, and to move beyond

the notion that co-ordination occurs naturally through sharing interests and exchanging resources.

Crucially, the analysis of the contexts of inter-organisational relations also broadens the perspective

to include vertical in addition to horizontal relations.

Further, the analysis of the local co-ordination of health services in the two countries

demonstrates not only that governments can still steer (cf. Johansson and Borrel 1998), but also that

the co-ordination among organisational actors explicitly requires some steering from the centre.

This can take a variety of forms, from steering up close and restructuring to stimulation at a

distance, and different types of steering make for different types of inter organisational relations at

local level. However, the analysis also indicates that more tightly focused inter-organisational

relations (reflecting more hierarchical steering) seem to be better positioned to co-ordinate health

services at the local level, although there are limits to this. In its very strong version hierarchical

steering is in danger of in effect weakening inter organisational relations, as recent developments in

20

Britain illustrate. In sum, co-ordination among organisational actors at sub national level emerges as

a combination of both network and hierarchy-based elements of governing and, as Newman (2001)

argues, governance emerges from the dynamics between different, seemingly conflicting modes of

governing.

21

Bibliography

Bach, I (2000) ‘Government within governance: network steering in Yorkshire and the Humber’, Public Administration, 78 (3), 575-92.

Badura, B, E Münch, M Schnabel, P Wilking, P Zamora, J Siegrist, J Büchner, H Fuchs, L Joksimovic, O v d Knesebeck and R Rugulies (2000) Ortsnahe Koordinierung der gesundheitlichen und sozialen Versorgung. Abschlussbericht der wissenschaftlichen Begleitforschung zu einem Modellvorhaben, Düsseldorf: Ministerium für Frauen, Jugend, Familie und Gesundheits des Landes Nordrhein-Westfalen.

Blank, R H and V Burau (2004) Comparative health policy, Basingstoke: Palgrave. Börzel, T A (1998) ‘Organizing Babylon – on the different conceptions of policy networks’, Public

Administration, 76, 253-73. Bundesministerium für Gesundheit und Soziale Sicherung (2001) Entwicklung der

Pflegeversicherung. Zweiter Bericht, Bonn: Bundesministerium für Gesundheit und Soziale Sicherung.

Bundesministerium für Gesundheit und Soziale Sicherung (2004) Die Gesundheitsreform: eine gesunde Entscheidung für alle!, Berlin: Bundesministerium für Gesundheit und Soziale Sicherung.

Burau, V (1997) ‘Decentramento e “governo” dell’assistenza sanitaria: il caso della Germania’ [Decentralisation and “government” in the health service: the case of Germany], L’Assistenza Sociale, (1-2), 61-71.

Burau, V (1999) ‘Challenging boundaries in nursing: some preliminary observations on the reform of community care in Britain and Germany’, in M Dent, M O’Neill and C Bagley (eds) Professions, new public management and the European welfare state, Stoke-on-Trent: Staffordshire University Press.

Burau, V (2001) ‘Medical reform in Germany: the 1993 health care legislation as an impromptu success’. in M A P Bovens, P ‘t Hart and B G Peters (eds) Success and failure in public governance, Cheltenham: Edward Elgar.

Burau, V (2004) ‘Die Koordination der Gesundheitsversorgung auf kommunaler Ebene – das Beispiel Grossbritannien’, Stadtforschung und Statistik, (1), 56-61.

Department of Health (2004) ‘The NHS explained. How the NHS works’.. www.nhs.uk/thenhsexpalined/HowTheNHSWorks.asp

Döhler, M (1995) ‘The state as architect of political order: policy dynamics in German health care’, Governance, 8 (3), 380-404.

Flynn, N (2000) ’Managerialism and public services: some international trends’, in J Clarke, S Gewirtz and E McLaughlin (eds) New managerialism, new welfare?, Milton Keynes: Open University Press.

Geiser (1995) ’Kommunale Altenpolitik und Pflegeversicherung: zur Zukunft der Altenplanung aud örtlicher Ebener’, in G Igl, S Kühnert and G Naegle (eds) SGB XI als Herausforderung für die Kommunen, Hanover: Vincentz.

Gitschmann, P (1995) ‚Vernetzung und Qualitätssicherung in der Altenhilfe der Stadt Hamburg – Perspektiven im Pflegeversicherungssystem’, in G Igl, S Kühnert and G Naegle (eds) SGB XI als Herausforderung für die Kommunen, Hanover: Vincentz.

Glendinning, C (2002) ’Partnerships between health and social services: developing a framework for evaluation’, Policy & Politics, 30 (1), 115-27.

Glendinning, C, S Abbott and A Coleman (2001) ‘”Bridging the gap”: new relationships between primary care groups and local authorities’, Social Policy & Administration, 35 (4), 411-25.

Hudson, B (2002a) ‘Integrated care and structural change in England: the case of care trusts’, Policy Studies, 23 (2), 77-95.

22

Hudson, B (2002b) ‘Interprofessionality in health and social care: the Achilles’ heel of partnership’, Journal of Interprofessional Care, 16 (1), 7-17.

Hudson, B and M Henwood (2002) ‘The NHS and social care: the final countdown?’, Policy & Politics, 30 (2), 153-66.

Igl, G, S Kühnert and G Naegle (eds) (1995) SGB XI als Herausforderung für die Kommunen, Hanover: Vincentz.

Johansson, R and K Borrel (1998) ‘Central steering and local networks: old-age care in Sweden’, Public Administration, 77 (3), 585-98.

Kenis, P and Schneider, V (1991) ‘Policy network and policy analysis: scrutinizing a new analytical toolbox’, in B Marin and R Mayntz (eds) Policy networks: empiricial evidence and theoretical considerations, Frankfurt am Main: Campus.

Kickert, W (1993) ‘Complexity, governance and dynamics: conceptual explorations of public network management’, in J Kooiman (ed) Modern governance. New government-society interactions, London: Sage.

Kickert, W (1995) ‘Steering at a distance: a new paradigm of public governance in Dutch higher education’, Governance, 8 (1), 135-57.

Knesebeck, O v d, L Joksimovic, B Badura and J Siegrist (2002) ‘Evaluation of a community-level health policy intervention’, Health Policy, 61, 111-22.

Kooiman, J (1993) ‘Governance and governability: using complexity, dynamics and diversity’ in J Kooiman (ed) Modern governance. New government-society interactions, London: Sage.

Kooiman, J (2000) ‘Societal governance: levels, models, and orders of social-political interaction’, in J Pierre (ed) Debating governance. Authority, steering, and democracy, Oxford: Oxford University Press.

Kooiman, J (2003) Governing as governance, London: Sage. Kühn, H (2001) Integration der medizinischen Versorgung in regionaler Perspektive. Dimensionen

und Leitbild eines politisch-ökonomischen, sozialen und kulturellen Prozesses, (WZB discussion papers, P01-202), Berlin: Wissenschaftszentrum Berlin für Sozialforschung.

Kümpers, S, A van Raak, B Hardy and I Mur (2002) ’The influence of institutions and culture on health policies: different approaches to integrated care in England and the Netherlands’, Public Administration, 80 (2), 339-58.

Landesinstitut für den öffentlichen Gesundheitssdienst (2004) ‘Ortsnahe Koordinierung und kommunale Gesundheitskonferenzen’. www.loeg.nrw.de/gesundheitspolitik/kommunen/geko_oko/geko_oko.html

Lowdnes, V and C Skelcher (1998) ‘The dynamics of multi-organizational partnerships: an analysis of changing modes of governance’, Public Administration, 76, 313-33.

Marin, B and Mayntz, R (1991) ‘Introduction: studying policy networks’ in B Marin and R Mayntz (eds) Policy networks: empiricial evidence and theoretical considerations, Frankfurt am Main: Campus.

Mayntz, R (1993a) ‘Governing failures and the problem of governability: some comments on a theoretical paradigm’, in J Kooiman (ed) Modern governance. New government-society interactions, London: Sage.

Mayntz, R (1993b) ‘Modernization and the logic of interorganizational networks’, in J Child, M Crozier and R Mayntz (eds) Societal change between market and organization, Aldershot: Avebury.

Mayntz, R and F W Scharpf (1995) ‘Der Ansatz des akteurszentrierten Institutionalismus’, in R Mayntz and F W Scharpf (eds) Gesellschaftliche Selbststeuerung und politische Steuerung, Frankfurt am Main: Campus.

23

Mayntz, R and F W Scharpf (1995) ‘Steuerung und Selbstorganisation in staatsnahen Sektoren’, in R Mayntz and F W Scharpf (eds) Gesellschaftliche Selbststeuerung und politische Steuerung, Frankfurt am Main: Campus.

Means, R, S Richards and R Smith (2003) Community care. Policy and practice, (third edition), Basingstoke: Palgrave.

Moon, G and N North (2000) Policy and place. General medical practice in the UK, Basingstoke: Macmillan.

Moschner (1995) ‘Erfahrungen mit der Einrichtung eines kommunalen Verbundssystems pflegerischer Einrichtungen und Dienste in der Stadt Münster’ in G Igl, S Kühnert and G Naegle (eds) SGB XI als Herausforderung für die Kommunen, Hanover: Vincentz.

Mur-Veeman, I, A van Raak and A Paulus (1999) ’Integrated care: the impact of governmental behaviour on collaborative networks’, Health Policy, 49, 149-59.

Mur-Veeman, I, B Hardy, M Steenbergen and G Wistow (2003) ‘Development of integrated care in England and the Netherlands. Managing across public-private boundaries’, Health Policy, 65 (3), 227-41.

Newman, J (2001) Moderninsing governance. New Labour, policy and society, London: Sage. Oldemeier, U (without year) Einrichtung von regionalen Pflegekonferenzen zur Sicherung und

qualitativen Weiterentwicklung der pflegerischen Angebotsstruktur, Facharbeit im Rahmen der Ausbildung zur Pflegedienstleistung am Fortbildungsinstitut für akademische und soziale Berufe der Deutschen Angestellten Akademie in Hannover. www.pflegenet.com/beruf/beruf/politik/pflegekonferenzen.html

Peckham, S and M Exworthy (2003) Primary care in the UK. Policy, organisation and management, Basingstoke: Palgrave.

Peters, B G (2000) ‘Governance and comparative politics’ in J Pierre (ed) Debating governance. Authority, steering, and democracy, Oxford: Oxford University Press.

Pierre, J and B G Peters (2000) Governance, politics and the state, Basingstoke: Macmillan. Rhodes, R A W (1996) ‘The new governance: governing without government’, Political Studies, 44

(4), 652-67. Rhodes, R A W (1997) Understanding governance. Policy networks, governance, reflexivity and

accountability, Buckingham: Open University Press. Rhodes, R A W (2000) ‘Governance and public administration’ in J Pierre (ed) Debating

governance. Authority, steering, and democracy, Oxford: Oxford University Press. Rosenbrock, R and T Gerlinger (2004) Gesundheitspolitik. Eine systematische Einführung, Bern:

Hans Huber. Rummery, K and A Coleman (2003) ‘Primary health and social care services in the UK: progress

towards partnerships’, Social Science & Medicine, 56, 1773-82. Sachverständigenrat für die Konzertierte Aktion für das Gesundheitswesen (2003) Gutachten 2003

des Sachverständigenrates für die Konzertierte Aktion im Gesundheitswesen. Finanzierung, Nutzerorientierung und Qualität, (Drucksache 15/530), Berlin: Deutscher Bundestag.

Verdi/PLS Rambøll (without year) Von einem fragmentierten zu einem integrierten medizinischen Versorgungssystem – Überblick über den aktuellen Stand der Neuausrichtung des deutschen Versorgungssystems, (Transparenzstudie gefördert durch die Hans-Böckler-Stiftung), without place: Verdi.

Wasem, J (1997) ‘A study on decentralization from acute care to home care settings in Germany’, Health Policy, 41 (supplement), S109-29.

Wendt, W R (1995) ‘Kooperation im Vorfeld und Umfeld der Pflege: die IAV-Stellen in baden-Württemberg’, in G Igl, S Kühnert and G Naegle (eds) SGB XI als Herausforderung für die Kommunen, Hanover: Vincentz.