3.1 England: from dirigiste towards a regulatory state

The NHS system in the United Kingdom was, prior to 1991, an archetype of what Majone referred to as a ‘dirigiste’ state: public ownership (hospitals), planning (volume and capacity planning) and centralised administration (global budgets). The NHS continues to be free at the point of delivery funded by taxation; with salaried hospital specialists as employees of, and General Practitioners (GPs) as independent contractors with, the NHS; and the private sector's role for short stay care is marginal. For the first 30 years of the NHS, the emphasis by both Labour and Conservative administrations was tackling inequities in access to health care and seeking to improve services that had been neglected (for the elderly, mentally ill and ‘mentally handicapped’). Equity was sought from 1948, by making access to services free at the point of delivery and controlling the distribution of GPs and from 1977, by allocating resources for hospital and community health services to health authorities on an equitable basis based on their populations’ estimated relative need. Planning was the means of tackling neglected services (Bevan et al., Reference Bevan, Copeman, Perrin and Rosser1980; Webster, Reference Webster1988; Klein, Reference Klein2007).

Following the election of the Conservatives under Margaret Thatcher, the government sought to make the NHS more efficient and moved away from the ‘dirigiste’ state with the introduction of the ‘internal market’ in 1991 (Secretaries of State for Health, Wales, Northern Ireland and Scotland, 1989), which was intended to lead to competition between providers of care but not between District Health Authorities (DHAs), which may be seen as local insurers in which enrolment was determined by place of residence. This meant that DHAs no longer funded providers in a hierarchical arrangement but contracted with them. Providers became NHS Trusts independent of health authorities, but directly accountable to Ministers, and were funded by a system in which ‘money followed the patient’. GPs could opt to become GP fundholders, who contracted for diagnostic and elective services and managed budgets for these services and their own prescribing costs (Bevan and Robinson, Reference Bevan and Robinson2005; Klein, Reference Klein2007). As part of the policies of the internal market, new kinds of audit were introduced into the NHS: of value for money by the Audit Commission (until this passed to the Healthcare Commission in 2004); and medical audit by doctors (Klein, Reference Klein2007). Although this meant the demise of the ‘dirigiste’ state, there was little development of a regulatory state as the Department of Health in effect fulfilled most of the regulatory functions.

From 1997, the Labour government following its election abolished the idea of provider competition and GP-fundholding (on grounds that this had created a two-tier NHS) retained the purchaser/provider split and initially sought a ‘third way’ between ‘command and control’ and a market (Secretary of State for Health, 1997). The legacies of that failed search were the creation of two important regulators (Klein, Reference Klein2007; Bevan, Reference Bevan2011). First, the National Institute for Clinical Excellence, which recommends guidelines and whether (mainly new) treatments ought to be made available for the NHS based on assessments of their cost-effectiveness. This regulator aimed to end the perceived problem of ‘postcode rationing’. Second, the creation of an independent quality regulator in the NHS: the Commission for Health Improvement (CHI).

From 2000, the Labour government sought to improve NHS performance by a systems of ‘targets and terror’ and ‘naming and shaming’ (Secretary of State for Health, 2000). CHI took over from the Department of Health publication of an assessment of performance of NHS organisations in annual ‘star ratings’. Those NHS organisations that performed well were granted ‘earned autonomy’, whereas those that performed poorly were subject to heavy monitoring and their chief executives were at risk of being sacked (Bevan and Hood, Reference Bevan and Hood2006).

From 2002, the government developed the ‘new Labour market’ of provider competition with pluralism in delivery (Secretary of State for Health, 2002). The rationale was that regulation by targets could only improve NHS performance from appalling to mediocre: for a high-performing NHS markets driven by patient choice were required (Barber, Reference Barber2007; Le Grand Reference Le Grand2007). The Labour government created a new status for high-performing NHS Trusts, independent of the Department of Health, as NHS Foundation Trusts, which were subject to approval and regulation by a new regulator, Monitor. Additionally a small number of Independent Sector Treatment Centres were created to provide diagnostic and elective services and good value for money (Secretary of State for Health, 2002; Audit and Health Care Commission, 2008). CHI, which had been designed to regulate the NHS, was replaced by the Healthcare Commission, with the objective of creating a ‘level playing field’ in regulating quality of care in NHS and independent providers. From 2009, the Healthcare Commission was in turn replaced by the Care Quality Commission (CQC), which is responsible for regulating quality of public and private providers of health and social care (Bevan, Reference Bevan2011). The Department of Health also created the Co-operation and Competition Panel (CCP), in 2009, as an independent advisory body to the Department of Health, Monitor and NHS-funded stakeholders. CCP also makes recommendations on formal cases to the Department on mergers, acquisitions, joint ventures and other transactions between NHS organisations, complaints on the conduct of service providers and commissioners and disputes over procurement and advertising (Department of Health, 2010a). Two key functions remained with the Department of Health: the allocation of resources to purchasers and the setting of hospital prices at standard national tariffs. We consider below the extent to which the Coalition government's proposals for the NHS in England (Secretary of State for Health, 2010) move it further towards a ‘regulatory’ state.

3.2 The Netherlands: between corporatism and etatism

The Dutch health-care system is a classical example of a Bismarckian corporatist health insurance system with predominantly public financing and a private delivery of health care in which national associations of health-care providers, insurers, trade unions and employers play an important intermediary role (Okma, Reference Okma1997; Van der Grinten, Reference van der Grinten2001; Helderman et al., Reference Helderman, Schut, van der Grinten and van de Ven2005). Although a corporatist state may also rely on planning and centralised administration, it differs from the ‘dirigiste’ state on the aspect of public ownership. In the corporatist state, the state is dependent on private actors for the delivery of public goods and services. Care is provided by independent professionals (GPs, independent specialists and physiotherapists) and by private institutions (hospitals, home-care organisations and nursing homes). In the course of the development of the post-war welfare state, the Netherlands evolved from being a predominantly decentralised unitary state towards a highly centralised unitary state, combining both corporatist and etatist policy arrangements (Helderman et al., Reference Helderman, Schut, van der Grinten and van de Ven2005). Except for the supervision of quality, which was – and still is – the exclusive task of the state funded but independent Inspectorate for Health Care and its predecessors, corporatism was present nearly everywhere in Dutch health care.

Dutch health care has always been governed by several boards and councils and many of these used to include representatives from the health-care sector and independent experts. One of these is the Sickness Fund Council (Ziekenfondsraad), established in 1949 in order to administer the Sickness Fund Decree. In 1964, the Council was converted into a more formalised council composed of the representatives of employers and employees, representatives of the sickness funds, their contract partners (the health-care providers) and independent members appointed by the Crown (Okma, Reference Okma1997; Helderman, Reference Helderman2007).

Health-care prices used to be regulated by the Central Office on Hospital Prices (COZ), founded in 1965 with the enactment of the Hospital Prices Act. According to this Act, hospital price setting was to be determined by negotiations between the sickness funds and hospitals, and approved by the COZ. In 1982, the COZ was converted into the Central Negotiating Board Tariffs Health Care (COTG), which was charged with the task of implementing the Health Care Tariffs Act (Wet Tarieven Gezondheidszorg). The representatives of sickness funds and private health insurers, however, had little incentive to negotiate the lowest possible prices (Schut, Reference Schut1995). Sickness funds were fully and retrospectively reimbursed for their members’ medical expenses and the private health insurers had more of an interest in high rather than low health-care prices because rising health-care expenditure would push up their income from premiums (given the inelastic demand for health care).

The problems with the COTG can be taken as an example of more general problems of corporatist governance in the Netherlands in the 1970s and 1980s (Hemerijck, Reference Hemerijck1992). Although the Minister of Health was formally entitled to give binding instructions to the COTG, in practice, the legal basis for hierarchical instructions by the Minister turned out to be quite narrow (Schut, Reference Schut1995; Helderman, Reference Helderman2007). From the 1970s onwards, the need to rationalise volume and capacity planning and to contain expenditures on health care entailed a greater independence of these semi-public boards in order to avoid state-capture by the organised interests of health insurers and providers. It is mainly because of a tight system of supply-side regulation and budget caps that the government succeeded in gaining control over health-care expenditure, as a result of which the proportion of GDP spent on health services has remained stable at around 8.5% throughout the 1980s (Organisation for Economic Co-operation and Development (OECD, 2000). But these hierarchical interventions led on their turn to a growing dissatisfaction among physicians and hospitals. In the mid-1980s, the post-war corporatist relationships between the government, the health-care insurers and the health-care providers were under serious pressure.

From the late 1980s onwards, successive Dutch coalition governments have worked on the development and implementation of a national health insurance scheme, complemented with a system of regulated competition in the health insurance and health-care provision market. The reforms started with the advisory report of the Dekker committee in 1987. The Dekker Committee was an ad hoc committee, selected on the basis of individual merit rather than corporatist representation of health insurers, hospital, physicians and social partners. The Dekker Committee published its advice in March 1987 under the significant title ‘Willingness to Change’ (Commissie Dekker, 1987) in which it proposed to replace all separate health-care financing schemes by one mandatory national health insurance scheme, while a system of regulated competition should create the incentives for both health insurers and providers to improve the efficiency and quality of health-care delivery. The legal distinction between sickness funds and private health insurers would have to be abolished so that both would be allowed to offer ‘basic’ benefits as well as optional supplementary health insurance. This was finally implemented in the Health Insurance Act (ZVW) of 2006. Prior to 2006, more than 60% of the population fell under the Sickness Fund Act but the remainder had to take out private insurance coverage. After 2006, 100% of the population was required to have mandatory private health insurance.

3.3 Italy: regional sovereignty vs central responsibility

An important event in Italy's post-war history is the abolition of the Bismarckian health insurance system and the creation of the National Health Service (Servizio Sanitario Nazionale – SSN) in 1978. This latter was modelled in important ways on the British NHS with overarching principles of universalism, comprehensiveness of care, progressive tax funding and maximisation of resource efficiency. Italy had provided health protection through a system of sickness funds, not unlike the Netherlands. But in comparison with the Netherlands, the Italian funds were less stable financially, running chronic deficits and having to be periodically bailed out by the state. The majority of hospitals and ambulatory clinics were independent and themselves debt ridden mainly due to chronic delays in reimbursement by the sickness funds. In contrast with the Netherlands, moreover, centralised government and corporatism had a bad reputation in Italy, mainly because of their association with Fascism. The 1948 Italian Constitution had already provided for 21 regions and autonomous provinces to be created. In the case of health care, the combination of regionalism and the health-care model chosen in 1978 was to have important consequences for the regulatory arrangements eventually adopted.

The original blueprint of the SSN seemed to intend that the principal locus of power be at the centre: for example, levels of aggregate health spending were to be set by Parliament, rather than endogenously within the health sector, and to be funded by a central block grant. The regions would be responsible for setting broad policy in their territory and for allocating the state transfer among the new Local Health Authorities. To begin with, the new regions were probably incapable of doing much else since they lacked a professional civil service with a regional ethos and had to make do with demotivated staff from the defunct sickness funds and from the central ministries downsized after regionalisation. The real operational power was to lie with local government. Day-to-day administration of the SSN would be overseen by a local management board made up of nominees of local authorities reflecting the political party composition of their elected assemblies. The aim here was to promote ‘democratic participation’ and put an end to a tradition where the hospitals and medical barons had set spending patterns. But the management boards quickly degenerated into political machines, generating systematically waste and inefficiency, clientelism and even at times corruption (Ferrera, Reference Ferrera1995).

Most hospitals were nationalised with the creation of the SSN, but one of the many deals struck to get the health reform bill through Parliament allowed teaching, research and certain ecclesiastical hospitals to remain independent but with the right to contract with the SSN. In addition, the Constitution's protection of free enterprise meant that the SSN – unlike the British NHS in 1946 – could not unilaterally absorb private health facilities. This meant, for example with hospital care, that in many geographical areas and in many specialties, the SSN relied to a significant degree on private providers, between 15% and 20% on average and a third and over in a few regions (Neri, Reference Neri2009). Pluralism in provision implies pluralism in demand and, although many regions disputed the legality of this, in practice SSN patients enjoyed significant freedom of choice of provider. According to the SSN statute, private operators were not supposed to compete with public providers, but rather complement them. Conventions negotiated between the state and regions and the organisations representing the private providers set out the obligations and rights of the parties involved. These conventions were essentially rudimentary contracts, incomplete leaving ample room for opportunistic behaviour. De facto, private operators competed among themselves and with public providers for SSN patients. External transaction costs borne by the SSN were probably low given the scarce resort to contract, but the internal transaction costs associated with large highly vertically integrated organisations were likely quite high.

Attempts by the centre to solve these problems by direct intervention encountered legal obstacles. Numerous Constitutional Court sentences in the 1980s recognised the regions’ sovereignty over organisational and administrative matters, including matters concerning the local health authorities. The fact that competition was not officially envisaged meant that no need was felt for recognition to control it. The principle regulatory route taken to tackle local poor performance, provided of course that audit can be considered regulatory, was the promotion of the internal and external audit. Perhaps the most innovative initiative was the regional College of Auditors created by national legislation in 1982 to supervise the more than 650 Local Health Authorities. Regional laws set the investigative mandate of the College, the financial and human resources it enjoyed to carry out its activities, frequency of controls, powers of access to information and health authority facilities, and there were wide interregional differences in all these aspects. The thrust of its audit activity tended however to regard regularity and conformity. The College, however, was a hybrid in that it was responsible to both the Ministry of Treasury and to the region. (France and Prisco, Reference France and Prisco1986) and also, formally, was considered an internal organ of the Local Health Authority.

The Colleges of Auditors jostled for room in a crowded field. Other organs with audit powers included: the national Court of Accounts, the Inspectorate General of Finance of the Treasury, the Central Planning Directorate and the Anti-Fraud Unit, both of the Ministry of Health, and finally the inspection services of the single regions. All tended to emphasise regularity and conformity over performance. A form of ‘audit chaos’ was being created that is overlapping or duplication of auditing combined with serious audit gaps (Buglione and France, Reference Buglione and France1990). Toward the end of this period the central state began the slow process of building up its technical capacity to control sub-central government, which would become a decisive factor in the post-2000 period. Forces in and outside Parliament had began to lobby for reform as early as 1983. But the 1978 arrangements would essentially remain intact for over a decade (France and Taroni, Reference France and Taroni2005).

4.1 England: the regulatory impacts of provider markets

Market-oriented reforms in the English NHS created three interrelated regulatory challenges: the regulation of quality, prices and the functioning of the ‘market’. In the ‘internal market’, there was no external regulatory oversight of quality. Provider competition was intended to be on both price and quality, but there was little comparative information on either. The regulatory rule for prices was that these ought to equal costs. To remedy the lack of pressure on DHAs as purchasers, because of the absence of purchaser competition, the government regulated DHAs by the Purchaser Efficiency Index that required them to reduce costs per episode by 3% each year and also put pressure on them to reduce waiting times.

In the ‘new Labour’ market, the Healthcare Commission regulated and reported on quality of public and private providers. Tariffs for providers were set by the Department of Health in the system of ‘Payment by Results’. This was modelled on the US Prospective Payment System by Diagnosis-Related Groups (DRGs). The objective was that competition be on quality and not on price.

Studies of these different markets have generally worked within one of two paradigms (Kuhn, 1962). Analyses of the impacts of the markets on the various players have generally concluded that both markets were ineffective. Econometric analyses, in contrast, claim to find a causal relationship between greater intensity of competition and reduced hospital quality in the ‘internal market’, and improved hospital quality in the ‘New Labour market’. In these analyses, hospital quality is principally measured by hospital mortality rates, which Bevan and Skellern (Reference Bevan and Skellern2011) argue do not measure quality of services for which hospitals compete and different outcomes from these markets reflect their different regulatory designs. But sadly in each, scandals over poor quality of care emerged.

In the late 1990s, a series of notorious scandals, including that of excess deaths following paediatric cardiac surgery at the Bristol Royal Infirmary, highlighted the need for an external regulator of quality (Abbasi, Reference Abbasi1998; Bevan, Reference Bevan2008). The introduction of pluralism in, and competition between, providers, from 2002, meant a dramatic shift in the role of the regulator of quality. In 1997, as the government abandoned the policy of competition and pluralism of providers, CHI was seen as the regulator responsible for quality assurance in NHS providers based on inspections of each organisation organised around a visit. The objective of creating a ‘level playing field’ for the regulation of quality by public and private providers was seen to require moving to a system of ‘light touch’ regulation. The Healthcare Commission assessed quality based on routinely available data and self-declarations by trusts, which were supposed to be validated by a sample of inspections. There were weaknesses in the system of ‘light touch’ regulation by the Health Care Commission and it failed to work in harmony with monitor, the regulator of foundation trusts. Monitor approved the application for Mid-Staffordshire NHS Foundation Trust when it was being investigated by the Health Care Commission, which revealed a scandal in quality of care that has been the subject of three enquiries and is currently the subject of a public enquiry (Bevan, Reference Bevan2011). The Healthcare Commission was in an ambiguous position being both responsible for quality improvement and also, in effect, a regulator of public and private providers eligible to enter into the NHS market. The Coalition Government removed that ambiguity for CQC: its role is only the latter. Unlike its predecessors, CQC charges all providers a fee to be registered.

In 2010, the new Conservative-Liberal Coalition Government in its White Paper Liberating the NHS (Secretary of State for Health, 2010), proposed the abolition of the remaining bureaucratic hierarchical arrangements of the NHS in England, which suggests a transformation into a ‘regulatory state’. Purchasing would be done by GP consortia. The remaining NHS Trusts would become NHS Foundation Trusts. A new NHS Commissioning Board will regulate purchasing and allocate their resources. Monitor would become the economic regulator and set tariffs. CQC, the regulator of quality, would also provide a new independent consumer champion, HealthWatch England to inform patients of quality of care (Department of Health, 2010b). The House of Commons Public Accounts Committee (2011) raised questions over accountability in these proposed reforms. These proposals have been revised (Department of Health, 2011): including clinical commissioning groups replacing GP consortia and Monitor becoming responsible for competition and integration. It is, however, unclear as to how regulatory roles will develop between the Secretary of State for Health, the NHS Commissioning Board and the other regulators, and hence the extent to which the English NHS will move into a ‘regulatory state’.

4.2 The Netherlands: towards a model of ‘regulated’ competition

Different from in England, where the internal market was more or less introduced overnight, the Dutch reforms evolved in a much more gradual and incremental way (Schut, Reference Schut1995; Helderman et al., Reference Helderman, Schut, van der Grinten and van de Ven2005). Already in 1992, the regional monopolies of the sickness funds were abolished, and sickness funds were permitted to define their own geographical market. At the same time, sickness funds were required to have biennial open enrolment periods, during which enrollees were free to switch between sickness funds, irrespective of their health status. The main effect of these initial steps had been a large number of mergers between health insurers (sickness funds as well as private health insurers) and between hospitals, and a considerable reinforcement of regional co-operation among health-care professionals, such as GPs, pharmacists and physiotherapists. These market dynamics in turn required the development of specialised regulatory agencies aimed at safeguarding competition and consumer protection and equity.

The regulatory reforms that followed need to be understood within the context of two more general developments in the Dutch welfare state in the 1990s. Already in 1994, the Cabinet of the Social Democrat Prime Minister Wim Kok initiated a series of reforms aimed at reducing the number of corporatist advisory bodies in various segments of the welfare state by terminating the participation of sectoral interest groups. In the second half of the 1990s, many administrative boards and advisory councils were successfully reformed so that partisan interests were no longer formally represented. Second, as a product of European integration and the Cabinet's programme of introducing more competition into social services sectors, a new stringent Competition Act was adopted in 1998 under the responsibility of the Ministry of Economic Affairs and to be monitored by the Dutch Competition Authority (NMa). Given that competition – although still a marginal phenomenon – was now part of the health-care policy system as well, the NMa was assigned the role of ‘market-umpire’ in Dutch health care.

Most regulatory reforms that were undertaken, however, were not aimed at creating new agencies, as in England, but at converting the formerly corporatist bodies into semi-independent regulatory boards. The convergence between sickness funds and private health insurers, for example, had important consequences for the corporatist structure of the Sickness Fund Council. In January 2000, the Sickness Fund Council was converted into the Health Care Insurance Board (CVZ), governed by nine independent members appointed by the Minister of Health. It became responsible for the management of the Central Health Insurance Fund from which the risk-equalization subsidies are paid to the health insurers. The CVZ has the authority to instruct health insurers on administrative procedures, the registration of their enrollees, the collection of statistics, annual reports and the conditions of service staff. In addition, the Board has to inform the Minister of Health about all matters concerning the health insurance market and issues of equity related to the mandatory basic package.

In the 1990s, supervision of health-care insurers was placed in the hands of the Supervisory Board for Health Care Insurance (College van Toezicht op de Zorgverzekeringen, CTZ), whereas health-care prices used to be regulated by the COTG. Under the market-oriented reforms in the 1990s, the COTG has been converted into to the Board for Health Care Tariffs (CTG), now consisting of nine independent members appointed by the Minister of Health, similar to the composition of the CVZ mentioned above. In October 2006, both the CTG and the CTZ were integrated in the newly erected Dutch Health Care Authority (NZa: Nederlandse Zorg Autoriteit), which is now the most important market regulator in Dutch health care. Although there is some redundancy between the NZa and the NMa, the government felt that it needed a specialised agency to prepare and monitor the conditions for regulated competition in health care, including supervising health insurers and health-care providers; examining whether the information that purchasers provide to their potential enrollees is complete and truthful; whether the offered insurance packages are in accordance with the law; and to ensure transparency and consumer information in the market (Ministry of Health, Welfare and Sports (VWS, 2003).

With the enactment of the new ZVW on 1 January 2006, the formerly bifurcated health insurance system has finally been replaced by a single mandatory national health insurance scheme, which guarantees universal access to basic health-care services to be provided by both the former sickness funds and the private health insurers (Helderman, Reference Helderman2007). The reforms led to a full convergence of non-profit sickness funds and for-profit health insurers operating within a universal National Health Insurance regime on a national health insurance market. The new ZVW was complemented with the Health Care Market Organization Act, which regulates the newly created health insurance market and health-care provision market. Today, the Dutch health insurance market is dominated by four large insurance companies (NZa, 2010), monitored by three semi-independent regulatory agencies. Although the Minister of Health Care remains ultimately responsible for their functioning and for any political decision concerning health care, they have deliberately been located at arms length of the Ministry in order to free them from partisan politics and to enable them to develop specialised expertise. Once created, these agencies started to play an active role in the further development of regulated competition in Dutch health care. In a number of important decisions, for example, the NMa has forbidden horizontal price-fixing and market-sharing agreements, entry regulations and collective contracting practices by GPs, physiotherapists, pharmacists and other independent medical practitioners. Another example is the development of Diagnosis Related Treatment Combinations (DRGs) and the active role that the NZa played in this. Initially, the development of the DRGs was left to the professional associations of physicians. But when these came up with no less than 30,000 different DRGs, the government decided to bypass the professional associations and to ask the NZa to develop a more transparent system with a smaller number of DRGs.

To conclude, the conversion of formerly corporatist bodies into newly created independent regulatory agencies, assigned to different regulatory objectives in health care, can be interpreted as a transformation from a predominantly corporatist structured system towards a more regulatory managed health-care system. The creation of semi-independent regulatory agencies probably did ease the way for the government to pass through major policy change in the 1990s and 2000s. This does not imply the end of the deeply rooted practice of formal and informal consultations and negotiations between governmental actors and organised interest groups in Dutch health care. In a small country such as the Netherlands, where organised interests meet each other regularly, it is virtually impossible (and also unwise) to ‘exit’ the health policy arena for any involved stakeholder (Okma and de Roo, Reference Okma and de Roo2009). But it is also true that there is an inherent tension between the economic logic of the (quasi-) market and corporatism in the sense that the interdependencies and collaborative practices that fit in with the logic of corporatism come close to what an economist, following the logic of the market, would label as cartelisation (Streeck and Schmitter, Reference Streeck and Schmitter1985). The creation of semi-independent regulatory agencies may find its justification in the increasingly cumbersome balance between public and private interests in Dutch health care. But whether these newly created semi-independent agencies are more immune from capture by market forces, than the state was from organised interest forces in the former corporatist system, remains to be seen.

4.3 Italy's two speed devolution model

In the early 1990s, Italy was caught in a severe political and financial crisis, which opened a window of opportunity for drastic reforms (France and Taroni, Reference France and Taroni2005). Legislation in 1992 and 1993 confirmed the power that the regions had accumulated de facto over the 1980s. The legislation also transferred to the regions the responsibilities that local government had had in health care. The aim was to clarify the locus of responsibility for spending and, therefore, for funding health care. The new inter-governmental division of labour allocated the task of promoting the performance of the SSN to the regions. The health policy developments in Italy bore a (superficial) resemblance to policy innovations in England: internal markets in health care and the New Public Management. But a crucial difference between the two countries was that the English reforms were centrally designed and executed while in Italy, for constitutional reasons, central government plans had necessarily to be contained in framework legislation. The responsibility for detailed specification and implementation of the reforms was the competence of the region. The purchaser–provider split was far less marked in Italy and there was no GP-fundholding, but services were to be paid for by the Health Care Enterprises (which replaced the local health authorities) using DRG based tariffs for hospital care. SSN markets were further opened to private providers, and competition between public and private providers for SSN patients was not only formally recognised but also encouraged.

A distinction must be made here between these government functions that the regions took over with devolution and the regulatory function, which they began exercising after 1993. Much of the implementation of reform was concerned, on the one hand, with improving existing managerial capacity at the regional and local level and on the other with creating regulatory capacity ex novo, the need for which was probably perceived only after the event. The Health Care Enterprise was given strong managerial autonomy and headed by a general manager appointed by, and accountable to, the region. Regulation required now that there was competition between providers, creating a need for payment systems, accreditation, contracts and agreements and quality control. In the following years, an imposing body of regional legislation was approved regulating accreditation and the organisation and management of Health Care Enterprises (Mapelli, Reference Mapelli2007).

Regional regulatory models differed considerably one from another. Some regions stood out for the energetic use they made of their autonomy while others were distinguished by their inertness. The central authorities in Rome were quick in recognising the need for well designed tariff systems, but were dilatory in pressing the regions for accreditation arrangements and mechanisms for controlling private providers (Falcitelli and Langiano, Reference Falcitelli and Langiano2004; Arcangeli, Reference Arcangeli2010). Most regions welcomed technical guidance from the central level (Ministry of Health and through the National Agency for Regional Health Services), but overall they opposed state attempts to impose national tariff ceilings and to restrict region-specific tariff policies. Where the state did act with alacrity was in the enactment of the essential care levels (livelli essenziali di assistenza – LEAs), a national health entitlement, aimed at preventing the regions from cutting back services provided. In 2001, the LEAs were granted constitutional status and became operational after the state and regions agreed on how these levels were to be defined and financed (Torbica and Fattore, Reference Torbica and Fattore2005).

Regional regulatory activity further intensified in the new century. Some regions, like Emilia–Romagna, and Umbria were energetic in setting rules, standards and guidelines and in enforcing compliance (Camera dei Deputati, 2003–2010). Instead, Lombardia tended to adopt a hands off approach, limiting itself to defining the prerequisites for accreditation, leaving most other matters to its health-care enterprises (Arcangeli, Reference Arcangeli2010). The state promoted increased regulatory action by the regions in the hope that it would improve resource utilisation and help to contain expenditures. For the same reason, over the decade some major revenue sources were ceded to the regions and in 2000 a government plan was published for an ambitious form of fiscal federalism (Arachi and Zanardi, Reference Arachi and Zanardi2004). The aim seems to have been to shift the blame from central government to the regions for unpopular measures taken to eliminate deficits (e.g. increasing taxes or patient co-payments). This worked for a while as real public health spending fell, but the deficits reappeared and state bailing-out of the regions resumed (Bordignon and Turati, Reference Bordignon and Turati2009).

To conclude, the manner in which regulation of health care has evolved in Italy has depended crucially on its decentralised system of government and the state of inter-governmental relations. In the early years of the SSN, the state limited its intervention to the promotion of internal and a kind of external audit. The introduction of internal markets, although quite limited, encouraged a regulatory orientation although there was considerable interregional disparity. It was attempted to alleviate the effects of decentralised government by voluntary cooperation, but this produced only limited results. Once again a certain similarity exists between Italy and England with national government treatment of regions in Italy mirroring the English approach to high and poor performing providers. Regions with high state capacity that manage expenditures within their budgets are granted autonomy to innovate. The state uses its financial leverage instead to persuade/compel the chronic deficit spenders to encourage them to use well-stocked regulatory tool boxes. These asymmetrical arrangements permit ‘two speed devolution’, appropriate for a country such as Italy, characterised by a regionalist form of government and by significant divergence in levels of economic development and administrative capacity.